受賞区分：国内学会・会議・シンポジウム等の賞  受賞国：日本国

受賞国：ドイツ連邦共和国

受賞国：アメリカ合衆国

受賞国：アメリカ合衆国

受賞国：アメリカ合衆国

受賞国：日本国

受賞国：アメリカ合衆国

受賞国：日本国

受賞国：日本国

受賞国：日本国

受賞国：日本国

受賞国：日本国

受賞国：日本国

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：British Journal of Ophthalmology  

Aim To develop an artificial intelligence (AI) algorithm that diagnoses cataracts/corneal diseases from multiple conditions using smartphone images. Methods This study included 6442 images that were captured using a slit-lamp microscope (6106 images) and smartphone (336 images). An AI algorithm was developed based on slit-lamp images to differentiate 36 major diseases (cataracts and corneal diseases) into 9 categories. To validate the AI model, smartphone images were used for the testing dataset. We evaluated AI performance that included sensitivity, specificity and receiver operating characteristic (ROC) curve for the diagnosis and triage of the diseases. Results The AI algorithm achieved an area under the ROC curve of 0.998 (95% CI, 0.992 to 0.999) for normal eyes, 0.986 (95% CI, 0.978 to 0.997) for infectious keratitis, 0.960 (95% CI, 0.925 to 0.994) for immunological keratitis, 0.987 (95% CI, 0.978 to 0.996) for cornea scars, 0.997 (95% CI, 0.992 to 1.000) for ocular surface tumours, 0.993 (95% CI, 0.984 to 1.000) for corneal deposits, 1.000 (95% CI, 1.000 to 1.000) for acute angle-closure glaucoma, 0.992 (95% CI, 0.985 to 0.999) for cataracts and 0.993 (95% CI, 0.985 to 1.000) for bullous keratopathy. The triage of referral suggestion using the smartphone images exhibited high performance, in which the sensitivity and specificity were 1.00 (95% CI, 0.478 to 1.00) and 1.00 (95% CI, 0.976 to 1.000) for’urgent’, 0.867 (95% CI, 0.683 to 0.962) and 1.00 (95% CI, 0.971 to 1.000) for’semi-urgent’, 0.853 (95% CI, 0.689 to 0.950) and 0.983 (95% CI, 0.942 to 0.998) for’routine’ and 1.00 (95% CI, 0.958 to 1.00) and 0.896 (95% CI, 0.797 to 0.957) for’observation’, respectively. Conclusions The AI system achieved promising performance in the diagnosis of cataracts and corneal diseases.

DOI： 10.1136/bjo-2023-324488

Scopus

PubMed

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Sensors  

In the context of Minimally Invasive Surgery, surgeons mainly rely on visual feedback during medical operations. In common procedures such as tissue resection, the automation of endoscopic control is crucial yet challenging, particularly due to the interactive dynamics of multi-agent operations and the necessity for real-time adaptation. This paper introduces a novel framework that unites a Hierarchical Quadratic Programming controller with an advanced interactive perception module. This integration addresses the need for adaptive visual field control and robust tool tracking in the operating scene, ensuring that surgeons and assistants have optimal viewpoint throughout the surgical task. The proposed framework handles multiple objectives within predefined thresholds, ensuring efficient tracking even amidst changes in operating backgrounds, varying lighting conditions, and partial occlusions. Empirical validations in scenarios involving single, double, and quadruple tool tracking during tissue resection tasks have underscored the system’s robustness and adaptability. The positive feedback from user studies, coupled with the low cognitive and physical strain reported by surgeons and assistants, highlight the system’s potential for real-world application.

DOI： 10.3390/s23249865

Scopus

PubMed

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Gastrointestinal Endoscopy  

Background and Aims: Gastric cancer (GC) is associated with chronic gastritis. To evaluate the risk, the Operative Link on Gastric Intestinal Metaplasia Assessment (OLGIM) system was constructed and showed a higher GC risk in stage III or IV patients, determined by the degree of intestinal metaplasia (IM). Although the OLGIM system is useful, evaluating the degree of IM requires substantial experience to produce precise scoring. Whole-slide imaging is becoming routine, but most artificial intelligence (AI) systems in pathology are focused on neoplastic lesions. Methods: Hematoxylin and eosin–stained slides were scanned. Images were divided into each gastric biopsy tissue sample and labeled with an IM score. IM was scored as follows: 0 (no IM), 1 (mild IM), 2 (moderate IM), and 3 (severe IM). Overall, 5753 images were prepared. A deep convolutional neural network (DCNN) model, ResNet50, was used for classification. Results: ResNet50 classified images with and without IM with a sensitivity of 97.7% and specificity of 94.6%. IM scores 2 and 3, involved as criteria of stage III or IV in the OLGIM system, were classified by ResNet50 in 18%. The respective sensitivity and specificity values of classifying IM between scores 0 and 1 and 2 and 3 were 98.5% and 94.9%, respectively. The IM scores classified by pathologists and the AI system were different in only 438 images (7.6%), and we found that ResNet50 tended to miss small foci of IM but successfully identified minimal IM areas that pathologists missed during the review. Conclusions: Our findings suggested that this AI system would contribute to evaluating the risk of GC accuracy, reliability, and repeatability with worldwide standardization.

DOI： 10.1016/j.gie.2023.06.056

Scopus

PubMed

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：European Spine Journal  

Purpose: Postoperative complication prediction helps surgeons to inform and manage patient expectations. Deep learning, a model that finds patterns in large samples of data, outperform traditional statistical methods in making predictions. This study aimed to create a deep learning-based model (DLM) to predict postoperative complications in patients with cervical ossification of the posterior longitudinal ligament (OPLL). Methods: This prospective multicenter study was conducted by the 28 institutions, and 478 patients were included in the analysis. Deep learning was used to create two predictive models of the overall postoperative complications and neurological complications, one of the major complications. These models were constructed by learning the patient's preoperative background, clinical symptoms, surgical procedures, and imaging findings. These logistic regression models were also created, and these accuracies were compared with those of the DLM. Results: Overall complications were observed in 127 cases (26.6%). The accuracy of the DLM was 74.6 ± 3.7% for predicting the overall occurrence of complications, which was comparable to that of the logistic regression (74.1%). Neurological complications were observed in 48 cases (10.0%), and the accuracy of the DLM was 91.7 ± 3.5%, which was higher than that of the logistic regression (90.1%). Conclusion: A new algorithm using deep learning was able to predict complications after cervical OPLL surgery. This model was well calibrated, with prediction accuracy comparable to that of regression models. The accuracy remained high even for predicting only neurological complications, for which the case number is limited compared to conventional statistical methods.

DOI： 10.1007/s00586-023-07562-2

DOI： 10.1007/s00586-023-07562-2

Scopus

PubMed

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：IEICE Transactions on Information and Systems  

Social relationships (e.g., couples, opponents) are the foundational part of society. Social relation atmosphere describes the overall interaction environment between social relationships. Discovering social relation atmosphere can help machines better comprehend human behaviors and improve the performance of social intelligent applications. Most existing research mainly focuses on investigating social relationships, while ignoring the social relation atmosphere. Due to the complexity of the expressions in video data and the uncertainty of the social relation atmosphere, it is even difficult to define and evaluate. In this paper, we innovatively analyze the social relation atmosphere in video data. We introduce a Relevant Visual Concept (RVC) from the social relationship recognition task to facilitate social relation atmosphere recognition, because social relationships contain useful information about human interactions and surrounding environments, which are crucial clues for social relation atmosphere recognition. Our approach consists of two main steps: (1) we first generate a group of visual concepts that preserve the inherent social relationship information by utilizing a 3D explanation module; (2) the extracted relevant visual concepts are used to supplement the social relation atmosphere recognition. In addition, we present a new dataset based on the existing Video Social Relation Dataset. Each video is annotated with four kinds of social relation atmosphere attributes and one social relationship. We evaluate the proposed method on our dataset. Experiments with various 3D ConvNets and fusion methods demonstrate that the proposed method can effectively improve recognition accuracy compared to end-to-end ConvNets. The visualization results also indicate that essential information in social relationships can be discovered and used to enhance social relation atmosphere recognition.

DOI： 10.1587/transinf.2023PCP0008

Scopus

CiNii Research

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Journal of Clinical Medicine  

Spinal cord tumors are infrequently identified spinal diseases that are often difficult to diagnose even with magnetic resonance imaging (MRI) findings. To minimize the probability of overlooking these tumors and improve diagnostic accuracy, an automatic diagnostic system is needed. We aimed to develop an automated system for detecting and diagnosing spinal schwannomas and meningiomas based on deep learning using You Only Look Once (YOLO) version 4 and MRI. In this retrospective diagnostic accuracy study, the data of 50 patients with spinal schwannomas, 45 patients with meningiomas, and 100 control cases were reviewed, respectively. Sagittal T1-weighted (T1W) and T2-weighted (T2W) images were used for object detection, classification, training, and validation. The object detection and diagnosis system was developed using YOLO version 4. The accuracies of the proposed object detections based on T1W, T2W, and T1W + T2W images were 84.8%, 90.3%, and 93.8%, respectively. The accuracies of the object detection for two spine surgeons were 88.9% and 90.1%, respectively. The accuracies of the proposed diagnoses based on T1W, T2W, and T1W + T2W images were 76.4%, 83.3%, and 84.1%, respectively. The accuracies of the diagnosis for two spine surgeons were 77.4% and 76.1%, respectively. We demonstrated an accurate, automated detection and diagnosis of spinal schwannomas and meningiomas using the developed deep learning-based method based on MRI. This system could be valuable in supporting radiological diagnosis of spinal schwannomas and meningioma, with a potential of reducing the radiologist’s overall workload.

DOI： 10.3390/jcm12155075

Scopus

PubMed

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Communications of the ACM  

DOI： 10.1145/3589736

Scopus

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Scandinavian Journal of Gastroenterology  

Objectives: Meticulous inspection of the mucosa during colonoscopy, represents a lengthier withdrawal time, but has been shown to increase adenoma detection rate (ADR). We investigated if artificial intelligence-aided speed monitoring can improve suboptimal withdrawal time. Methods: We evaluated the implementation of a computer-aided speed monitoring device during colonoscopy at a large academic endoscopy center. After informed consent, patients ≥18 years undergoing colonoscopy between 5 March and 29 April 2021 were examined without the use of the speedometer, and with the speedometer between 29 April and 30 June 2021. All colonoscopies were recorded, and withdrawal time was assessed based on the recordings in a blinded fashion. We compared mean withdrawal time, percentage of withdrawal time ≥6 min, and ADR with and without the speedometer. Results: One hundred sixty-six patients in each group were eligible for analyses. Mean withdrawal time was 9 min and 6.6 s (95% CI: 8 min and 34.8 s to 9 min and 39 s) without the use of the speedometer, and 9 min and 9 s (95% CI: 8 min and 45 s to 9 min and 33.6 s) with the speedometer; difference 2.3 s (95% CI: −42.3–37.7, p = 0.91). The ADRs were 45.2% (95% CI: 37.6–52.8) without the speedometer as compared to 45.8% (95% CI: 38.2–53.4) with the speedometer (p = 0.91). The proportion of colonoscopies with withdrawal time ≥6 min without the speedometer was 85.5% (95% CI: 80.2–90.9) versus 86.7% (95% CI: 81.6–91.9) with the speedometer (p = 0.75). Conclusions: Use of speed monitoring during withdrawal did not increase withdrawal time or ADR in colonoscopy. ClinicalTrials.gov Identifier: NCT04710251.

DOI： 10.1080/00365521.2022.2154616

Scopus

PubMed

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Computer Assisted Radiology and Surgery  

The original version of this article unfortunately contained a mistake. The incorrect notations were given in the author’s affiliations.

DOI： 10.1007/s11548-023-02899-x

Scopus

PubMed

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Computer Assisted Radiology and Surgery  

Purpose: Endoscopic submucosal dissection (ESD) is a minimally invasive treatment for early gastric cancer. However, perforations may happen and cause peritonitis during ESD. Thus, there is a potential demand for a computer-aided diagnosis system to support physicians in ESD. This paper presents a method to detect and localize perforations from colonoscopy videos to avoid perforation ignoring or enlarging by ESD physicians. Method: We proposed a training method for YOLOv3 by using GIoU and Gaussian affinity losses for perforation detection and localization in colonoscopic images. In this method, the object functional contains the generalized intersection over Union loss and Gaussian affinity loss. We propose a training method for the architecture of YOLOv3 with the presented loss functional to detect and localize perforations precisely. Results: To qualitatively and quantitatively evaluate the presented method, we created a dataset from 49 ESD videos. The results of the presented method on our dataset revealed a state-of-the-art performance of perforation detection and localization, which achieved 0.881 accuracy, 0.869 AUC, and 0.879 mean average precision. Furthermore, the presented method is able to detect a newly appeared perforation in 0.1 s. Conclusions: The experimental results demonstrated that YOLOv3 trained by the presented loss functional were very effective in perforation detection and localization. The presented method can quickly and precisely remind physicians of perforation happening in ESD. We believe a future CAD system can be constructed for clinical applications with the proposed method.

DOI： 10.1007/s11548-022-02821-x

Scopus

PubMed

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Computer Assisted Radiology and Surgery  

Purpose: Segmentation tasks are important for computer-assisted surgery systems as they provide the shapes of organs and the locations of instruments. What prevents the most powerful segmentation approaches from becoming practical applications is the requirement for annotated data. Active learning provides strategies to dynamically select the most informative samples to reduce the annotation workload. However, most previous active learning literature has failed to select the frames that containing low-appearing frequency classes, even though the existence of these classes is common in laparoscopic videos, resulting in poor performance in segmentation tasks. Furthermore, few previous works have explored the unselected data to improve active learning. Therefore, in this work, we focus on these classes to improve the segmentation performance. Methods: We propose a class-wise confidence bank that stores and updates the confidence scores for each class and a new acquisition function based on a confidence bank. We apply confidence scores to explore an unlabeled dataset by combining it with a class-wise data mixture method to exploit unlabeled datasets without any annotation. Results: We validated our proposal on two open-source datasets, CholecSeg8k and RobSeg2017, and observed that its performance surpassed previous active learning studies with about 10 % improvement on CholecSeg8k, especially for classes with a low-appearing frequency. For robSeg2017, we conducted experiments with a small and large annotation budgets to validate situation that shows the effectiveness of our proposal. Conclusions: We presented a class-wise confidence score to improve the acquisition function for active learning and explored unlabeled data with our proposed class-wise confidence score, which results in a large improvement over the compared methods. The experiments also showed that our proposal improved the segmentation performance for classes with a low-appearing frequency.

DOI： 10.1007/s11548-022-02773-2

Web of Science

Scopus

PubMed

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Computer Assisted Radiology and Surgery  

Purpose: This paper aims to propose a deep learning-based method for abdominal artery segmentation. Blood vessel structure information is essential to diagnosis and treatment. Accurate blood vessel segmentation is critical to preoperative planning. Although deep learning-based methods perform well on large organs, segmenting small organs such as blood vessels is challenging due to complicated branching structures and positions. We propose a 3D deep learning network from a skeleton context-aware perspective to improve segmentation accuracy. In addition, we propose a novel 3D patch generation method which could strengthen the structural diversity of a training data set. Method: The proposed method segments abdominal arteries from an abdominal computed tomography (CT) volume using a 3D fully convolutional network (FCN). We add two auxiliary tasks to the network to extract the skeleton context of abdominal arteries. In addition, our skeleton-based patch generation (SBPG) method further enables the FCN to segment small arteries. SBPG generates a 3D patch from a CT volume by leveraging artery skeleton information. These methods improve the segmentation accuracies of small arteries. Results: We used 20 cases of abdominal CT volumes to evaluate the proposed method. The experimental results showed that our method outperformed previous segmentation accuracies. The averaged precision rate, recall rate, and F-measure were 95.5%, 91.0%, and 93.2%, respectively. Compared to a baseline method, our method improved 1.5% the averaged recall rate and 0.7% the averaged F-measure. Conclusions: We present a skeleton context-aware 3D FCN to segment abdominal arteries from an abdominal CT volume. In addition, we propose a 3D patch generation method. Our fully automated method segmented most of the abdominal artery regions. The method produced competitive segmentation performance compared to previous methods.

DOI： 10.1007/s11548-022-02767-0

Web of Science

Scopus

PubMed

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Urology  

Objective: To develop a convolutional neural network to recognize the seminal vesicle and vas deferens (SV-VD) in the posterior approach of robot-assisted radical prostatectomy (RARP) and assess the performance of the convolutional neural network model under clinically relevant conditions. Methods: Intraoperative videos of robot-assisted radical prostatectomy performed by the posterior approach from 3 institutions were obtained between 2019 and 2020. Using SV-VD dissection videos, semantic segmentation of the seminal vesicle-vas deferens area was performed using a convolutional neural network-based approach. The dataset was split into training and test data in a 10:3 ratio. The average time required by 6 novice urologists to correctly recognize the SV-VD was compared using intraoperative videos with and without segmentation masks generated by the convolutional neural network model, which was evaluated with the test data using the Dice similarity coefficient. Training and test datasets were compared using the Mann–Whitney U-test and chi-square test. Time required to recognize the SV-VD was evaluated using the Mann–Whitney U-test. Results: From 26 patient videos, 1 040 images were created (520 SV-VD annotated images and 520 SV-VD non-displayed images). The convolutional neural network model had a Dice similarity coefficient value of 0.73 in the test data. Compared with original videos, videos with the generated segmentation mask promoted significantly faster seminal vesicle and vas deferens recognition (P < .001). Conclusion: The convolutional neural network model provides accurate recognition of the SV-VD in the posterior approach RARP, which may be helpful, especially for novice urologists.

DOI： 10.1016/j.urology.2022.12.006

Scopus

PubMed

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Computer Assisted Radiology and Surgery  

Purpose: A surgical navigation system helps surgeons understand anatomical structures in the operative field during surgery. Patient-to-image registration, which aligns coordinate systems between the CT volume and a positional tracker, is vital for accurate surgical navigation. Although a point-based rigid registration method using fiducials on the body surface is often utilized for laparoscopic surgery navigation, precise registration is difficult due to such factors as soft tissue deformation. We propose a method that compensates a transformation matrix computed using fiducials on the body surface based on the analysis of positional information in the database. Methods: We built our database by measuring the positional information of the fiducials and the guidance targets in both the CT volume and positional tracker coordinate systems through previous surgeries. We computed two transformation matrices: using only the fiducials and using only the guidance targets in all the data in the database. We calculated the differences between the two transformation matrices in each piece of data. The compensation transformation matrix was computed by averaging these difference matrices. In this step, we selected the data from the database based on the similarity of the fiducials and the configuration of the guidance targets. Results: We evaluated our proposed method using 20 pieces of data acquired during laparoscopic gastrectomy for gastric cancer. The locations of blood vessels were used as guidance targets for computing target registration error. The mean target registration errors significantly decreased from 33.0 to 17.1 mm before and after the compensation. Conclusion: This paper described a registration error compensation method using a database for image-guided laparoscopic surgery. Since our proposed method reduced registration error without additional intraoperative measurements during surgery, it increases the accuracy of surgical navigation for laparoscopic surgery.

DOI： 10.1007/s11548-022-02804-y

Scopus

PubMed

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Techniques and Innovations in Gastrointestinal Endoscopy  

DOI： 10.1016/j.tige.2022.11.004

Scopus

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Progress in Biomedical Optics and Imaging - Proceedings of SPIE  

This paper presents surgical area recognition from laparoscopic images in laparoscopic gastrectomy. Laparoscopic gastrectomy is performed as a minimally invasive procedure for removing gastric cancer. In this surgery, surgeons should cut the blood vessels around the stomach before resecting cancer. Since this type of surgery requires higher surgical skill, a surgical assistance system has been developed to enhance surgeons' abilities. Recognition of the surgical area related to the blood vessels from laparoscopic videos provides essential information in the operative field to the surgical assistance system. Therefore, we develop a method for recognizing laparoscopic images into the surgical area. The proposed method classifies the laparoscopic images into seven scenes using deep neural networks. We introduce the label smoothing in time direction to obtain a soft label. Bayesian neural networks are used to classify the laparoscopic images and estimate the uncertainty. After the classification, we modify the predictions on each laparoscopic image using the estimated uncertainty and temporal information. We evaluated the proposed method using 10,818 images from 10 videos recorded during laparoscopic gastrectomy for gastric cancer. Five-fold cross-validation was performed for the performance evaluation. The mean classification accuracy was 84.0%. The experimental results showed that the proposed method could recognize the surgical area from laparoscopic images.

DOI： 10.1117/12.2654775

Scopus

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Progress in Biomedical Optics and Imaging - Proceedings of SPIE  

In this paper, we present a segmentation method for laparoscopic images using semantically similar groups for multi-class semantic segmentation. Accurate semantic segmentation is a key problem for computer assisted surgeries. Common segmentation models do not explicitly learn similarities between classes. We propose a model that, in addition to learning to segment an image into classes, also learns to segment it into human-defined semantically similar groups. We modify the LinkNet34 architecture by adding a second decoder with an auxiliary task of segmenting the image into these groups. The feature maps of the second decoder are merged into the final decoder. We validate our method against our base model LinkNet34 and a larger LinkNet50. We find that our proposed modification increased the performance both with mean Dice (average +1.5%) and mean Intersection over Union metrics (average +2.8%) on two laparoscopic datasets.

DOI： 10.1117/12.2654636

Scopus

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：公益社団法人 日本生体医工学会  

<p>【背景】膀胱癌は経尿道手術後に再発が多い腫瘍であり、膀胱鏡での腫瘍の見落としが原因とされている。内視鏡での観察は、従来の白色光(WLI)の他に、NBIを使用するが、いずれの腫瘍検出精度は検者の技量・経験に依存するため、検査の再現性・客観性が少ないことが課題である。近年、人工知能(AI)が多くの医療分野で活用されており、AIによる検査は、客観性・再現性を持った上で、エキスパートレベルと同程度の診断能を持つ可能性があるとされている。今回、WLI/NBI膀胱鏡画像を用いて、AIによる腫瘍検出の精度を検証した。【方法】2019年から2021年まで、経尿道的膀胱腫瘍切除術(TURBT)の際に、WLI/NBIを用いて観察を行った症例の手術動画から膀胱鏡画像を作成し、腫瘍を含む画像を腫瘍画像、腫瘍を含まない画像を正常画像と定義した。腫瘍画像内の膀胱腫瘍を矩形でアノテーションを行い、テストデータ用の画像を用いてAIによる感度、特異度、陽性的中率を評価した。AIでの物体検出はtiny-YOLOを用い、腫瘍検出精度の検証を行った。【結果】WLIとNBIから、それぞれ腫瘍画像をそれぞれtiny-YOLOで学習を行い、腫瘍画像(WLI: 525枚、NBI:219枚)と正常画像(WLI:98枚、NBI:108枚)で精度検証を行った。AIによる物体検出の感度/特異度/陽性的中率は、WLIで87.8%/88.8%/97.7%、NBIで82.2%/81.4%/90.0%であった。【結論】膀胱鏡画像において、AIにより比較的良好に腫瘍検出が可能であった。更なる精度改善、リアルタイム検出への課題について、文献的考察を加え報告する。</p>

DOI： 10.11239/jsmbe.annual61.255_2

CiNii Research

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：公益社団法人 日本生体医工学会  

<p>看護職は心身不調に由来する顕在化していない労働生産性低下状態（プレゼンティズム）が他職種より多く、不規則な勤務形態や強い心身ストレスにより、離職に至る率も高いのが現状である。類似研究は腰痛に注目してウェアラブルデバイスで筋硬度測定によりプレゼンティズム発生を予測するが、性別や運動経験の有無など個人の特性を大きく反映するため予測が難しいと報告がある。そこで本研究では、プレゼンティズムの主な要因で筋骨格系疼痛の原因となる看護行動を、Internet of Things (IoT)センシングで抽出・判別することを目的とする。既設のBLE 屋内測位(AoA法）アンテナを利用した加速度センサデバイスによる看護行動データの収集と、観察者の直接観察による看護行動に対するラベルデータ（看護行動、実施場所、行動の開始・終了時刻を入力）作成を実施した。看護行動は身体プレゼンティズムリスク(腰痛・膝痛・頚肩腕痛)に分類し定義した。1分毎の行動データ（位置情報の平均値、加速度の最大・最小・平均値・標準偏差）、及び時刻を入力データとし、ラベルデータを教師データとして機械学習を行い、行動認識モデルを構築した。看護行動割合は、腰痛43.6%、膝痛3.3%、頚肩腕痛7.4%であった。IoTセンシングで個人ごとのプレゼンティズムに発展する行動をモニタリングすることにより、プレゼンティズムを予防することが可能と示唆された。</p>

DOI： 10.11239/jsmbe.annual61.225_1

CiNii Research

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

CiNii Research

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

We present a general framework for medical image segmentation from limited supervision, reducing the reliance on fully and densely labeled data. Our method is simple, jointly trains triple diverse models, and adopts a mix augmentation scheme, and thus is called TriMix. TriMix imposes consistency under a more challenging perturbation, i.e., combining data augmentation and model diversity on the tri-training framework. This straightforward strategy enables TriMix to serve as a strong and general learner learning from limited supervision using different kinds of imperfect labels. We conduct extensive experiments to show TriMix’s generic purpose for semi- and weakly-supervised segmentation tasks. Compared to task-specific state-of-the-arts, TriMix achieves competitive performance and sometimes surpasses them by a large margin. The code is available at https://github.com/MoriLabNU/TriMix.

DOI： 10.1007/978-3-031-26351-4_12

Scopus

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Progress in Biomedical Optics and Imaging - Proceedings of SPIE  

In this paper, we propose a scheme that includes automated extraction of thrombus regions and quantitative analysis of thrombosis in confocal laser scanning microscope (CLSM) blood flow image sequence. Making thrombosis model in animal models play an important role in the development of antithrombotic drugs and ascertaining thrombosis mechanisms. Making thrombosis model in cerebral cortex of mice is usually observed using a CLSM in the fluorescence mode. However, some small changes of thrombus regions are not easily observed in CLSM blood flow image sequences. In addition, it is not easy for researchers to quantitatively analyze the degree of thrombosis. Therefore, we propose a scheme to achieve automatic thrombosis region extraction and quantitative analysis. In which, our thrombosis region extraction method uses analysis of changing pattern of thrombosis regions in CLSM blood flow image sequence. Experimental results showed that our scheme can help biological researchers observe and analyze the changes of thrombosis in animal models and reduced the use of fluorescent thrombus markers.

DOI： 10.1117/12.2654632

Scopus

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Progress in Biomedical Optics and Imaging - Proceedings of SPIE  

This article describes a method for bronchial nomenclature using real bronchoscopic (RB) images and pre-built knowledge base of branches. The bronchus has a complex tree-like structure, which increases the difficulty of bronchoscopy. Therefore, a bronchoscopic navigation system is used to help physicians during examination. Conventional navigation system used preoperative CT images and real bronchoscopic images to obtain the camera pose for navigation, whose accuracy is influenced by organ deformation. We propose a bronchial nomenclature method to estimate branch names for bronchoscopic navigation. This method consists of a bronchus knowledge base construction model, a camera motion estimation module, an anatomical structure tracking module, and a branch name estimation module. The knowledge base construction module is used to find the relationship of each branch. The anatomical tracking module is used to track the bronchial orifice (BO) extracted in each RB frame. The camera motion estimation module is used to estimate the camera motion between two frames. The branch name estimation module uses the pre-built bronchus knowledge base and BO tracking results to find the name of each branch. Experimental results showed that it is possible to estimate branch names using only RB images and the pre-built knowledge base of branches.

DOI： 10.1117/12.2654508

Scopus

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Progress in Biomedical Optics and Imaging - Proceedings of SPIE  

Neuromelanin magnetic resonance imaging (NM-MRI) has been widely used in the diagnosis of Parkinson’s disease (PD) for its significantly enhanced contrast between the PD-related structure, the substantia nigra (SN) and surrounding tissues. To develop the computer-aided diagnosis (CAD) system of PD and reduce the labor burden of clinicians, precise and automatic segmentation of SN is becoming more and more desired. This paper proposes a novel network combining the priority gating attention and Bayesian learning for improving the accuracy of fully automatic SN segmentation from NM-MRI. Different from the conventional gated attention model, the proposed network uses the prior SN probability map for guiding the attention computation and reducing the potential disruptions introduced by the background. Additionally, to lower the risks of over-fitting and estimate the confidence scores for the segmentation results, Bayesian learning with Monte Carlo dropout is applied in the training and testing phases. The quantitative results showed that the proposed network acquired the averaged Dice score of 79.46% in comparison with the baseline model 77.93%.

DOI： 10.1117/12.2655112

Scopus

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Progress in Biomedical Optics and Imaging - Proceedings of SPIE  

This paper proposes a deformable tissue model that introduces octree lattice vertex layout and cubic constraints to the orthodox PBD (Position Based Dynamics) method. Surgical simulation is expected to provide a safe method for training in surgery, which is especially useful for preoperative education of inexperienced surgeons and/or for the case a prior attempt is required. To build a surgical simulator, it is necessary to develop organ models with deformations and interaction algorithms between surgical instruments and organ models, all of which must be performed in real time. Since existing surgical simulators focus on real-time performance, the resolution of organ models is limited. The proposed method restricts the vertex locations of the PBD method to the vertices of the octree lattice to save computation time while maintaining a high deformation resolution. To obtain appropriate results even for large deformations, three-dimensional constraints are applied to each octree cube as the constraints of the PBD method. In the simulations, we tested the overall deformation by dropping a liver model and the local deformation scene by laparoscopic clipping. As a result, we achieved deformation simulations at 26.5 fps for the model with approximately 2,672 cube elements and 20,659 vertices.

DOI： 10.1117/12.2654092

Scopus

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

This work proposes an innovative self-supervised approach to monocular depth estimation in laparoscopic scenarios. Previous methods independently predicted depth maps ignoring spatial coherence in local regions and temporal correlation between adjacent images. The proposed approach leverages spatio-temporal coherence to address the challenges of textureless areas and homogeneous colors in such scenes. This approach utilizes a multi-view depth estimation model to guide monocular depth estimation when predicting depth maps. Moreover, the minimum reprojection error is extended to construct a cost volume for the multi-view model using adjacent images. Additionally, a 3D consistency of the point cloud back-projected from predicted depth maps is optimized for the monocular depth estimation model. To benefit from spatial coherence, deformable patch-matching is introduced to the monocular and multi-view models to smooth depth maps in local regions. Finally, a cycled prediction learning for view synthesis and relative poses is designed to exploit the temporal correlation between adjacent images fully. Experimental results show that the proposed method outperforms existing methods in both qualitative and quantitative evaluations. Our code is available at https://github.com/MoriLabNU/MGMDepthL.

DOI： 10.1007/978-3-031-43996-4_41

Scopus

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

Semantic segmentation of laparoscopic images is an important issue for intraoperative guidance in laparoscopic surgery. However, acquiring and annotating laparoscopic datasets is labor-intensive, which limits the research on this topic. In this paper, we tackle the Domain-Adaptive Semantic Segmentation (DASS) task, which aims to train a segmentation network using only computer-generated simulated images and unlabeled real images. To bridge the large domain gap between generated and real images, we propose a Masked Frequency Consistency (MFC) module that encourages the network to learn frequency-related information of the target domain as additional cues for robust recognition. Specifically, MFC randomly masks some high-frequency information of the image to improve the consistency of the network’s predictions for low-frequency images and real images. We conduct extensive experiments on existing DASS frameworks with our MFC module and show performance improvements. Our approach achieves comparable results to fully supervised learning method on the CholecSeg8K dataset without using any manual annotation. The code is available at github.com/MoriLabNU/MFC.

DOI： 10.1007/978-3-031-43907-0_63

Scopus

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Progress in Biomedical Optics and Imaging - Proceedings of SPIE  

This paper introduces the improved method for the COVID-19 classification based on computed tomography (CT) volumes using a combination of a complex-architecture convolutional neural network (CNN) and orthogonal ensemble networks (OEN). The novel coronavirus disease reported in 2019 (COVID-19) is still spreading worldwide. Early and accurate diagnosis of COVID-19 is required in such a situation, and the CT scan is an essential examination. Various computer-aided diagnosis (CAD) methods have been developed to assist and accelerate doctors' diagnoses. Although one of the effective methods is ensemble learning, existing methods combine some major models which do not specialize in COVID-19. In this study, we attempted to improve the performance of a CNN for the COVID-19 classification based on chest CT volumes. The CNN model specializes in feature extraction from anisotropic chest CT volumes. We adopt the OEN, an ensemble learning method considering inter-model diversity, to boost its feature extraction ability. For the experiment, We used chest CT volumes of 1283 cases acquired in multiple medical institutions in Japan. The classification result on 257 test cases indicated that the combination could improve the classification performance.

DOI： 10.1117/12.2653792

Scopus

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Progress in Biomedical Optics and Imaging - Proceedings of SPIE  

This paper proposes an automated classification method of COVID-19 chest CT volumes using improved 3D MLP-Mixer. Novel coronavirus disease 2019 (COVID-19) spreads over the world, causing a large number of infected patients and deaths. Sudden increase in the number of COVID-19 patients causes a manpower shortage in medical institutions. Computer-aided diagnosis (CAD) system provides quick and quantitative diagnosis results. CAD system for COVID-19 enables efficient diagnosis workflow and contributes to reduce such manpower shortage. In image-based diagnosis of viral pneumonia cases including COVID-19, both local and global image features are important because viral pneumonia cause many ground glass opacities and consolidations in large areas in the lung. This paper proposes an automated classification method of chest CT volumes for COVID-19 diagnosis assistance. MLP-Mixer is a recent method of image classification using Vision Transformer-like architecture. It performs classification using both local and global image features. To classify 3D CT volumes, we developed a hybrid classification model that consists of both a 3D convolutional neural network (CNN) and a 3D version of the MLP-Mixer. Classification accuracy of the proposed method was evaluated using a dataset that contains 1205 CT volumes and obtained 79.5% of classification accuracy. The accuracy was higher than that of conventional 3D CNN models consists of 3D CNN layers and simple MLP layers.

DOI： 10.1117/12.2654706

Scopus

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：BioMed Research International  

Ossification of the ligaments progresses slowly in the initial stages, and most patients are unaware of the disease until obvious myelopathy symptoms appear. Consequently, treatment and clinical outcomes are not satisfactory. This study is aimed at developing an automated system for the detection of the thoracic ossification of the posterior longitudinal ligament (OPLL) using deep learning and plain radiography. We retrospectively reviewed the data of 146 patients with thoracic OPLL and 150 control cases without thoracic OPLL. Plain lateral thoracic radiographs were used for object detection, training, and validation. Thereafter, an object detection system was developed, and its accuracy was calculated. The performance of the proposed system was compared with that of two spine surgeons. The accuracy of the proposed object detection model based on plain lateral thoracic radiographs was 83.4%, whereas the accuracies of spine surgeons 1 and 2 were 80.4% and 77.4%, respectively. Our findings indicate that our automated system, which uses a deep learning-based method based on plain radiographs, can accurately detect thoracic OPLL. This system has the potential to improve the diagnostic accuracy of thoracic OPLL.

DOI： 10.1155/2023/8495937

Scopus

PubMed

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：公益社団法人 日本生体医工学会  

<p>【はじめに】術後成長発達する小児患者にとって、低侵襲手術は非常に重要である。しかし患者数は限られているため、しっかりとした手術を行うためにoff the job-training(OJT)が重要である。さらにOJTでの効率的な手技獲得には、手技を客観的に評価しfeed backを行うシステムが必須である。また安全で効率的な内視鏡手術を行うためには、臓器の位置関係の把握が必要であるため、術中ナビゲーションは重要な要件となる。これらの課題に対して、AIを用いた内視鏡手技評価および手術支援システムの構築に着手しており現状の成果を報告する。【方法と結果】食道閉鎖症モデルを用いた吻合手技を被験者に課し、各被験者の手技を最初に人の目で「check 表」「エラー項目」「時間」を用いて評価した。次にビデオから検出した鉗子の動きと人が判定した手技優劣の関係性をAIで学習させ、上位88％・下位95％の精度で手技優劣が自動判定可能となった。この結果を解析することで今まで必要だった50項目以上の肉眼チェックが、わずか７項目チェックするだけで手技の優劣を判断できることが明らかになった。現在食道閉鎖症の手術画像を用いて、食道・迷走神経・気管を深層学習させ、各種構造物の自動認識を進めている。【まとめ】AI画像解析により内視鏡手技の優劣をビデオで判定可能となった。この結果から新たに効率的な手技判断基準を定めることができた。術中ナビゲーションは現在精度のさらなる向上を目指している。</p>

DOI： 10.11239/jsmbe.annual61.127_2

CiNii Research

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Computer Methods in Biomechanics and Biomedical Engineering: Imaging and Visualization  

Physicians use an endoscopic navigation system during bronchoscopy to decrease the risk of getting lost in complex tree-structure like bronchus. Most existing navigation systems based on the camera pose estimated from bronchoscope tracking and/or deep learning. However, bronchoscope tracking-based method exists tracking error, and the pre-training of the model needs massive data. This paper describes an improved bronchoscope tracking procedure by adopting image domain translation technique to improve tracking performance. Specifically, our scheme consists of three modules, an RGB-D image domain translation module, an anatomical structure classification module and a structure-aware bronchoscope tracking module. The RGB-D image domain translation module translates a real bronchoscope (RB) image to its corresponding virtual bronchoscope image and depth image. The anatomical dependency module classifies the current scene into two categories: structureless and rich structure. The bronchoscope tracking module uses a modified video-CT bronchoscope tracking approach to estimate camera pose. Experimental results showed that the proposed method achieved higher tracking accuracy than the current state-of-the-art bronchoscope tracking methods.

DOI： 10.1080/21681163.2022.2152728

Web of Science

Scopus

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：日本医用画像工学会  

<p>腹腔鏡下手術は，外科手術の一つの方法として広く行われるようになった．腹腔鏡下手術は高度な技術を要するため，コンピューターを用いて外科医を支援するコンピューター支援外科の研究が行われている．これらは，手術ナビゲーションとして一般的にはとらえられている．近年のAI技術の発展に伴い，腹腔鏡下手術支援のために深層学習を利用して腹腔鏡から得られる手術映像を解析する研究も行われるようになってきた．本稿では，筆者らの研究グループで取り組んでいるAIを用いた腹腔鏡映像の解析に関する研究を紹介する．</p>

DOI： 10.11409/mit.40.164

CiNii Research

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：日本VR医学会  

DOI： 10.24764/jsmvr.2022.0_9

CiNii Research

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1136/annrheumdis-2022-eular.933

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：一般社団法人電子情報通信学会  

DOI： 10.1587/ieiceissjournal.27.1_16

CiNii Research

DOI： 10.1080/21681163.2021.2012835

DOI： 10.1080/21681163.2021.2004445

DOI： 10.1080/21681163.2021.2015723

DOI： 10.1117/1.JMI.9.2.024003

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Annals of surgery  

DOI： 10.1097/SLA.0000000000005319

Scopus

PubMed

DOI： 10.1016/j.gie.2021.10.019

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1097/DCR.0000000000002393

PubMed

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Computer Assisted Radiology and Surgery  

Purpose: Pancreatic duct dilation can be considered an early sign of pancreatic ductal adenocarcinoma (PDAC). However, there is little existing research focused on dilated pancreatic duct segmentation as a potential screening tool for people without PDAC. Dilated pancreatic duct segmentation is difficult due to the lack of readily available labeled data and strong voxel imbalance between the pancreatic duct region and other regions. To overcome these challenges, we propose a two-step approach for dilated pancreatic duct segmentation from abdominal computed tomography (CT) volumes using fully convolutional networks (FCNs). Methods: Our framework segments the pancreatic duct in a cascaded manner. The pancreatic duct occupies a tiny portion of abdominal CT volumes. Therefore, to concentrate on the pancreas regions, we use a public pancreas dataset to train an FCN to generate an ROI covering the pancreas and use a 3D U-Net-like FCN for coarse pancreas segmentation. To further improve the dilated pancreatic duct segmentation, we deploy a skip connection on each corresponding resolution level and an attention mechanism in the bottleneck layer. Moreover, we introduce a combined loss function based on Dice loss and Focal loss. Random data augmentation is adopted throughout the experiments to improve the generalizability of the model. Results: We manually created a dilated pancreatic duct dataset with semi-automated annotation tools. Experimental results showed that our proposed framework is practical for dilated pancreatic duct segmentation. The average Dice score and sensitivity were 49.9% and 51.9%, respectively. These results show the potential of our approach as a clinical screening tool. Conclusions: We investigate an automated framework for dilated pancreatic duct segmentation. The cascade strategy effectively improved the segmentation performance of the pancreatic duct. Our modifications to the FCNs together with random data augmentation and the proposed combined loss function facilitate automated segmentation.

DOI： 10.1007/s11548-021-02530-x

Web of Science

Scopus

PubMed

担当区分：筆頭著者  

DOI： 10.1016/j.gie.2021.07.022

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

Segmentation studies in medical image analysis are always associated with a particular task scenario. However, building datasets to train models to segment multiple types of organs and pathologies is challenging. For example, a dataset annotated for the pancreas and pancreatic tumors will result in a model that cannot segment other organs, like the liver and spleen, visible in the same abdominal computed tomography image. The lack of a well-annotated dataset is one limitation resulting in a lack of universal segmentation models. Federated learning (FL) is ideally suited for addressing this issue in the real-world context. In this work, we show that each medical center can use training data for distinct tasks to collaboratively build more generalizable segmentation models for multiple segmentation tasks without the requirement to centralize datasets in one place. The main challenge of this research is the heterogeneity of training data from various institutions and segmentation tasks. In this paper, we propose a multi-task segmentation framework using FL to learn segmentation models using several independent datasets with different annotations of organs or tumors. We include experiments on four publicly available single-task datasets, including MSD liver (w/ tumor), MSD spleen, MSD pancreas (w/ tumor), and KITS19. Experimental results on an external validation set to highlight the advantages of employing FL in multi-task organ and tumor segmentation.

DOI： 10.1007/978-3-031-18523-6_6

Web of Science

Scopus

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Computer Assisted Radiology and Surgery  

Purpose: Bronchoscopists rely on navigation systems during bronchoscopy to reduce the risk of getting lost in the complex bronchial tree-like structure and the homogeneous bronchus lumens. We propose a patient-specific branching level estimation method for bronchoscopic navigation because it is vital to identify the branches being examined in the bronchus tree during examination. Methods: We estimate the branching level by integrating the changes in the number of bronchial orifices and the camera motions among the frames. We extract the bronchial orifice regions from a depth image, which is generated using a cycle generative adversarial network (CycleGAN) from real bronchoscopic images. We calculate the number of orifice regions using the vertical and horizontal projection profiles of the depth images and obtain the camera-moving direction using the feature point-based camera motion estimation. The changes in the number of bronchial orifices are combined with the camera-moving direction to estimate the branching level. Results: We used three in vivo and one phantom case to train the CycleGAN model and four in vivo cases to validate the proposed method. We manually created the ground truth of the branching level. The experimental results showed that the proposed method can estimate the branching level with an average accuracy of 87.6%. The processing time per frame was about 61 ms. Conclusion: Experimental results show that it is feasible to estimate the branching level using the number of bronchial orifices and camera-motion estimation from real bronchoscopic images.

DOI： 10.1007/s11548-021-02460-8

DOI： 10.1007/s11548-021-02460-8

Web of Science

Scopus

PubMed

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Translational Gastroenterology and Hepatology  

Computer-aided diagnosis (CAD) for colonoscopy with use of artificial intelligence (AI) is catching increased attention of endoscopists. CAD allows automated detection and pathological prediction, namely optical biopsy, of colorectal polyps during real-time endoscopy, which help endoscopists avoid missing and/or misdiagnosing colorectal lesions. With the increased number of publications in this field and emergence of the AI medical device that have already secured regulatory approval, CAD in colonoscopy is now being implemented into clinical practice. On the other side, drawbacks and weak points of CAD in colonoscopy have not been thoroughly discussed. In this review, we provide an overview of CAD for optical biopsy of colorectal lesions with a particular focus on its clinical applications and limitations.

DOI： 10.21037/tgh.2019.12.14

Web of Science

Scopus

PubMed

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：日本医用画像工学会  

<p>本稿では，名古屋大学におけるCT像からのCOVID-19診断支援AI開発に対する取り組みを紹介する．新型コロナウイルス感染症（COVID-19）が急速に世界に広まり，多くの感染者と死亡者を生じている．このように多数の患者を迅速に診断する必要がある状況では，AIによる診断支援が有効と考える．われわれは患者のCT像から画像所見に基づくCOVID-19典型度を自動判別するAI開発を行った．AIによる自動判別に必要となる3つの要素技術として肺野セグメンテーション，肺野領域クラスタリング，COVID-19典型度推定を開発し，CT像からの自動判別AIの処理フロー全体を構築した．本AIを開発する上で，多数の医療機関で撮影された膨大な画像を格納した医療画像ビッグデータクラウド基盤を活用し，実際のCOVID-19患者のCT像に対して高い精度での自動判別が可能であることを確認した．</p>

DOI： 10.11409/mit.39.13

CiNii Research

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Spine  

STUDY DESIGN: A retrospective analysis of magnetic resonance imaging (MRI) was conducted. OBJECTIVE: This study aims to develop an automated system for the detection of spinal schwannoma, by employing deep learning based on object detection from MRI. The performance of the proposed system was verified to compare the performances of spine surgeons. SUMMARY OF BACKGROUND DATA: Several MRI scans were conducted for the diagnoses of patients suspected to suffer from spinal diseases. Typically, spinal diseases do not involve tumors on the spinal cord, although a few tumors may exist at the unexpectable level or without symptom by chance. It is difficult to recognize these tumors; in some cases, these tumors may be overlooked. Hence, a deep learning approach based on object detection can minimize the probability of overlooking these tumors. METHODS: Data from 50 patients with spinal schwannoma who had undergone MRI were retrospectively reviewed. Sagittal T1- and T2-weighted magnetic resonance imaging (T1WI and T2WI) were used in the object detection training and for validation. You Only Look Once version3 was used to develop the object detection system, and its accuracy was calculated. The performance of the proposed system was compared to that of two doctors. RESULTS: The accuracies of the proposed object detection based on T1W1, T2W1, and both T1W1 and T2W1 were 80.3%, 91.0%, and 93.5%, respectively. The accuracies of the doctors were 90.2% and 89.3%. CONCLUSION: Automated object detection of spinal schwannoma was achieved. The proposed system yielded a high accuracy that was comparable to that of the doctors.Level of Evidence: 4.

DOI： 10.1097/BRS.0000000000003749

Web of Science

Scopus

PubMed

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Computer Methods in Biomechanics and Biomedical Engineering: Imaging and Visualization  

We propose a depth estimation method from a single-shot monocular endoscopic image using Lambertian surface translation by domain adaptation and depth estimation using multi-scale edge loss. We employ a two-step estimation process including Lambertian surface translation from unpaired data and depth estimation. The texture and specular reflection on the surface of an organ reduce the accuracy of depth estimations. We apply Lambertian surface translation to an endoscopic image to remove these texture and reflections. Then, we estimate the depth by using a fully convolutional network (FCN). During the training of the FCN, improvement of the object edge similarity between an estimated image and a ground truth depth image is important for getting better results. We introduced a muti-scale edge loss function to improve the accuracy of depth estimation. We quantitatively evaluated the proposed method using real colonoscopic images. The estimated depth values were proportional to the real depth values. Furthermore, we applied the estimated depth images to automated anatomical location identification of colonoscopic images using a convolutional neural network. The identification accuracy of the network improved from 69.2% to 74.1% by using the estimated depth images.

DOI： 10.1080/21681163.2021.2012835

Scopus

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Progress in Biomedical Optics and Imaging - Proceedings of SPIE  

This paper proposes a novel self-supervised depth estimation method guided by a context encoder. Depth estimation from stereo laparoscopic images is essential to robotic surgical navigation systems and robotic surgical platform. Recent work has shown that depth estimation of stereo image pairs can be formulated as a self- supervised learning task without ground-truth. However, most architectures based on convolutional neural lead to lose some spatial information because of the consecutive pooling and convolution operations. In order to tackle this problem, we add a contextual encoding module to the previous method. The context encoder module is formed by dense atrous convolution block and spatial pyramid pooling block that are used to extract and merge features on different scales. Also, we add the edge-awared smoothness for predicted disparity maps. In addition, we output multi-scale disparity predictions and corresponding image reconstruction for loss calculating. In the experiments, we showed that the proposed method has about 7.79% improvement in SSIM and about 17.76% improvement in PSNR for stereo image pairs compared with previous method. Also, the disparity maps and reconstructed images given by the proposed method have significant enhancements compared with the previous method.

DOI： 10.1117/12.2582348

Scopus

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Progress in Biomedical Optics and Imaging - Proceedings of SPIE  

This paper describes a bronchial orifice (BO) segmentation method on real bronchoscopic video frames by using depth images. The BO is one of the anatomical characteristics in the bronchus, which is critical in clinical applications such as bronchus scene description and navigation path generation. Previous work used image appearance and the gradation of the real bronchoscopic image to segment orifice region, which behaved poorly in complex scenes including bubble or changes in illumination. To obtain a better segmentation result of BO even in the complex scenes, we propose a BO segmentation method using the distance between the bronchoscope camera and the bronchus lumen, which is represented by a depth image. Since the depth image is unavailable due to devices limitation, we use an image-to-image domain translation network named cycle generative adversarial network (CycleGAN) to estimate depth images from real bronchoscopic images. The BO regions are considered as the regions whose distances are larger than a distance threshold. We decide the distance threshold according to the depth images' projection profiles. Experimental results showed that the proposed method can find BO regions in the real bronchoscopic videos in real-time. We manually labeled BO regions as ground truth to evaluate the proposed method. The average Dice score of the proposed method was 77.0 %.

DOI： 10.1117/12.2582341

Scopus

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

This paper proposes a segmentation method of infection regions in the lung from CT volumes of COVID-19 patients. COVID-19 spread worldwide, causing many infected patients and deaths. CT image-based diagnosis of COVID-19 can provide quick and accurate diagnosis results. An automated segmentation method of infection regions in the lung provides a quantitative criterion for diagnosis. Previous methods employ whole 2D image or 3D volume-based processes. Infection regions have a considerable variation in their sizes. Such processes easily miss small infection regions. Patch-based process is effective for segmenting small targets. However, selecting the appropriate patch size is difficult in infection region segmentation. We utilize the scale uncertainty among various receptive field sizes of a segmentation FCN to obtain infection regions. The receptive field sizes can be defined as the patch size and the resolution of volumes where patches are clipped from. This paper proposes an infection segmentation network (ISNet) that performs patch-based segmentation and a scale uncertainty-aware prediction aggregation method that refines the segmentation result. We design ISNet to segment infection regions that have various intensity values. ISNet has multiple encoding paths to process patch volumes normalized by multiple intensity ranges. We collect prediction results generated by ISNets having various receptive field sizes. Scale uncertainty among the prediction results is extracted by the prediction aggregation method. We use an aggregation FCN to generate a refined segmentation result considering scale uncertainty among the predictions. In our experiments using 199 chest CT volumes of COVID-19 cases, the prediction aggregation method improved the dice similarity score from 47.6% to 62.1%.

DOI： 10.1007/978-3-030-90874-4_9

Scopus

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

Pancreatic duct dilation indicates a high risk of pancreatic ductal adenocarcinoma (PDAC), the deadliest cancer with a poor prognosis. Segmentation of dilated pancreatic duct from CT taken from patients without PDAC shows the potential to assist the early detection of PDAC. Most current researches include pancreatic duct segmentation as one additional class for patients who have already detected PDAC. However, the dilated pancreatic duct for people who have not yet developed PDAC is typically much smaller, making the segmentation difficult. Deep learning-based segmentation on tiny components is challenging because of the large imbalance between the target object and irrelevant regions. In this work, we explore an attention-guided approach for dilated pancreatic duct segmentation as a screening tool for pre-PDAC patients, enhancing the pancreas regions’ concentration and ignoring the unnecessary features. We employ a multi-scale aggregation to combine the information at different scales to improve the segmentation performance further. Our proposed multi-scale pancreatic attention-guided approach achieved a Dice score of 54.16% on dilated pancreatic duct dataset, which shows a significant improvement over prior techniques.

DOI： 10.1007/978-3-030-90874-4_5

Scopus

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Gastroenterological Endoscopy  

The global incidence and mortality rate of colorectal cancer remains high. Colonoscopy is regarded as the gold standard examination for detecting and eradicating neoplastic lesions. However, there are some uncertainties in colonoscopy practice that are related to limitations in human performance. First, approximately one-fourth of colorectal neoplasms are missed on a single colonoscopy. Second, it is still difficult for nonexperts to perform adequately regarding optical biopsy. Third, recording of some quality indicators (e.g. cecal intubation, bowel preparation, and withdrawal speed）which are related to adenoma detection rate, is sometimes incomplete. With recent improvements in machine learning techniques and advances in computer performance, artificial intelligence-assisted computer-aided diagnosis is being increasingly utilized by endoscopists. In particular, the emergence of deep-learning, data-driven machine learning techniques have made the development of computer-aided systems easier than that of conventional machine learning techniques, the former currently being considered the standard artificial intelligence engine of computer-aided diagnosis by colonoscopy. To date, computer-aided detection systems seem to have improved the rate of detection of neoplasms. Additionally, computer-aided characterization systems may have the potential to improve diagnostic accuracy in real-time clinical practice. Furthermore, some artificial intelligence-assisted systems that aim to improve the quality of colonoscopy have been reported. The implementation of computer-aided system clinical practice may provide additional benefits such as helping in educational poorly performing endoscopists and supporting real-time clinical decision-making. In this review, we have focused on computer-aided diagnosis during colonoscopy reported by gastroenterologists and discussed its status, limitations, and future prospects.

DOI： 10.11280/gee.63.1402

Scopus

CiNii Research

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：一般社団法人 映像情報メディア学会  

DOI： 10.3169/itej.75.325

CiNii Research

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Scientific Reports  

Inherited retinal disorder (IRD) is a leading cause of blindness, and CRX is one of a number of genes reported to harbour autosomal dominant (AD) and recessive (AR) causative variants. Eighteen patients from 13 families with CRX-associated retinal disorder (CRX-RD) were identified from 730 Japanese families with IRD. Ophthalmological examinations and phenotype subgroup classification were performed. The median age of onset/latest examination was 45.0/62.5 years (range, 15–77/25–94). The median visual acuity in the right/left eye was 0.52/0.40 (range, −0.08–2.00/−0.18–1.70) logarithm of the minimum angle of resolution (LogMAR) units. There was one family with macular dystrophy, nine with cone-rod dystrophy (CORD), and three with retinitis pigmentosa. In silico analysis of CRX variants was conducted for genotype subgroup classification based on inheritance and the presence of truncating variants. Eight pathogenic CRX variants were identified, including three novel heterozygous variants (p.R43H, p.P145Lfs*42, and p.P197Afs*22). A trend of a genotype-phenotype association was revealed between the phenotype and genotype subgroups. A considerably high proportion of CRX-RD in ADCORD was determined in the Japanese cohort (39.1%), often showing the mild phenotype (CORD) with late-onset disease (sixth decade). Frequently found heterozygous missense variants located within the homeodomain underlie this mild phenotype. This large cohort study delineates the disease spectrum of CRX-RD in the Japanese population.

DOI： 10.1038/s41598-020-65737-z

Scopus

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Scientific Reports  

Ossification of the posterior longitudinal ligament (OPLL) can occur throughout the entire spine and can sometimes lead to spinal disorder. Although patients with OPLL sometimes develop physical limitations because of pain, the characteristics of pain and effects on activities of daily living (ADL) have not been precisely evaluated in OPLL patients. Therefore, we conducted a multi-center prospective study to assess whether the symptoms of cervical OPLL are different from those of cervical spondylosis (CS). A total of 263 patients with a diagnosis of cervical OPLL and 50 patients with a diagnosis of CS were enrolled and provided self-reported outcomes, including responses to the Japanese Orthopaedic Association (JOA) Cervical Myelopathy Evaluation Questionnaire (JOACMEQ), JOA Back Pain Evaluation Questionnaire (JOABPEQ), visual analog scale (VAS), and SF-36 scores. The severity of myelopathy was significantly correlated with each domain of the JOACMEQ and JOABPEQ. There was a negative correlation between the VAS score for each domain and the JOA score. There were significantly positive correlations between the JOA score and the Mental Health, Bodily Pain, Physical Functioning, Role Emotional, and Role Physical domains of the SF-36. One-to-one matching resulted in 50 pairs of patients with OPLL and CS. Although there was no significant between-group difference in scores in any of the domains of the JOACMEQ or JOABPEQ, the VAS scores for pain or numbness in the buttocks or limbs were significantly higher in the CS group; however, there was no marked difference in low back pain, chest tightness, or numbness below the chest between the two study groups. The scores for the Role Physical and Body Pain domains of the SF-36 were significantly higher in the OPLL group than in the CS group, and the mean scores for the other domains was similar between the two groups. The results of this study revealed that patients with OPLL were likely to have neck and low back pain and restriction in ADL. No specific type of pain was found in patients with OPLL when compared with those who had CS.

DOI： 10.1038/s41598-020-62278-3

Scopus

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： https://doi.org/10.1080/21681163.2020.1835560

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1002/ags3.12384

PubMed

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Computer Assisted Radiology and Surgery  

Purpose: Due to the complex anatomical structure of bronchi and the resembling inner surfaces of airway lumina, bronchoscopic examinations require additional 3D navigational information to assist the physicians. A bronchoscopic navigation system provides the position of the endoscope in CT images with augmented anatomical information. To overcome the shortcomings of previous navigation systems, we propose using a technique known as visual simultaneous localization and mapping (SLAM) to improve bronchoscope tracking in navigation systems. Methods: We propose an improved version of the visual SLAM algorithm and use it to estimate nt-specific bronchoscopic video as input. We improve the tracking procedure by adding more narrow criteria in feature matching to avoid mismatches. For validation, we collected several trials of bronchoscopic videos with a bronchoscope camera by exploring synthetic rubber bronchus phantoms. We simulated breath by adding periodic force to deform the phantom. We compared the camera positions from visual SLAM with the manually created ground truth of the camera pose. The number of successfully tracked frames was also compared between the original SLAM and the proposed method. Results: We successfully tracked 29,559 frames at a speed of 80 ms per frame. This corresponds to 78.1% of all acquired frames. The average root mean square error for our technique was 3.02 mm, while that for the original was 3.61 mm. Conclusion: We present a novel methodology using visual SLAM for bronchoscope tracking. Our experimental results showed that it is feasible to use visual SLAM for the estimation of the bronchoscope camera pose during bronchoscopic navigation. Our proposed method tracked more frames and showed higher accuracy than the original technique did. Future work will include combining the tracking results with virtual bronchoscopy and validation with in vivo cases.

DOI： 10.1007/s11548-020-02241-9

Web of Science

Scopus

PubMed

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： https://doi.org/10.1080/21681163.2020.1835543

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Minimally Invasive Therapy and Allied Technologies  

Background: Accurate registration for surgical navigation of laparoscopic surgery is highly challenging due to vessel deformation. Here, we describe the design of a deformable model with improved matching accuracy by applying the finite element method (FEM). Material and methods: ANSYS software was used to simulate an FEM model of the vessel after pull-up based on laparoscopic gastrectomy requirements. The central line of the FEM model and the central line of the ground truth were drawn and compared. Based on the material and parameters determined from the animal experiment, a perigastric vessel FEM model of a gastric cancer patient was created, and its accuracy in a laparoscopic gastrectomy surgical scene was evaluated. Results: In the animal experiment, the FEM model created with Ogden foam material exhibited better results. The average distance between the two central lines was 6.5mm, and the average distance between their closest points was 3.8 mm. In the laparoscopic gastrectomy surgical scene, the FEM model and the true artery deformation demonstrated good coincidence. Conclusion: In this study, a deformable vessel model based on FEM was constructed using preoperative CT images to improve matching accuracy and to supply a reference for further research on deformation matching to facilitate laparoscopic gastrectomy navigation.

DOI： 10.1080/13645706.2019.1625926

Web of Science

Scopus

PubMed

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Surgery  

Background: Identifying laparoscopic surgical videos using artificial intelligence (AI) facilitates the automation of several currently time-consuming manual processes, including video analysis, indexing, and video-based skill assessment. This study aimed to construct a large annotated dataset comprising laparoscopic colorectal surgery (LCRS) videos from multiple institutions and evaluate the accuracy of automatic recognition for surgical phase, action, and tool by combining this dataset with AI. Materials and methods: A total of 300 intraoperative videos were collected from 19 high-volume centers. A series of surgical workflows were classified into 9 phases and 3 actions, and the area of 5 tools were assigned by painting. More than 82 million frames were annotated for a phase and action classification task, and 4000 frames were annotated for a tool segmentation task. Of these frames, 80% were used for the training dataset and 20% for the test dataset. A convolutional neural network (CNN) was used to analyze the videos. Intersection over union (IoU) was used as the evaluation metric for tool recognition. Results: The overall accuracies for the automatic surgical phase and action classification task were 81.0% and 83.2%, respectively. The mean IoU for the automatic tool segmentation task for 5 tools was 51.2%. Conclusions: A large annotated dataset of LCRS videos was constructed, and the phase, action, and tool were recognized with high accuracy using AI. Our dataset has potential uses in medical applications such as automatic video indexing and surgical skill assessments. Open research will assist in improving CNN models by making our dataset available in the field of computer vision.

DOI： 10.1016/j.ijsu.2020.05.015

Web of Science

Scopus

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Techniques and Innovations in Gastrointestinal Endoscopy  

Because magnifying endoscopy is considered to be more accurate at predicting the histology of colorectal polyps than nonmagnifying endoscopy, it has been attracting a lot of attention, especially in Japan. However, use of magnifying endoscopy is not yet widespread because of its limited availability and the difficulty in interpreting the acquired images. Application of artificial intelligence (AI) is now changing this situation because it helps less-skilled endoscopists to accurately interpret magnified images. Research in this field initially focused on magnifying endoscopy with narrow-band imaging as the target of AI. Most previously published retrospective studies have reported over 90% sensitivity in differentiation of neoplastic lesions; however, automatically indicating the region of interest (ROI) of the polyps that AI should analyze has been found to be challenging. To address this practical problem, some researchers have started to adopt contact endomicroscopy as a target for AI. Contact endomicroscopy includes endocytoscopy (520-fold magnification, Olympus, Tokyo, Japan) and confocal laser endomicroscopy (1000-fold magnification, Mauna Kea, Paris, France). These forms of contact endomicroscopy provide ultramagnified images that make it unnecessary to manually select the ROI because the entire image acquired by contact endomicroscopy is the ROI of the targeted polyps. This strength of contact endomicroscopy has contributed to early implementation of this technology into clinical practice, which may change the utility of magnifying endoscopy in clinical settings and help increase its use globally in the near future.

DOI： 10.1016/j.tgie.2019.150632

Web of Science

Scopus

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Journal of Thoracic Disease  

Background: We have developed a surgical navigation system that presents virtual thoracoscopic images using computed tomography (CT) image data, as if you are observing intra-thoracic cavity in synchronization with the real thoracoscopic view. Using this system, we made it possible to simultaneously visualize the ‘area of lung cancer before induction therapy’ and the ‘optimal resection line for obtaining a safe surgical margin’ as a virtual thoracoscopic view. We applied this navigation system in the clinical setting in operations for lung cancer patients with chest wall invasion after induction chemoradiotherapy. Methods: The proposed surgical navigation system consisted of a three-dimensional (3D) positional tracker and a virtual thoracoscopy system. The 3D positional tracker was used to recognize the positional information of the real thoracoscope. The virtual thoracoscopy system generated virtual thoracoscopic views based on CT image data. Combined with these two technologies, patient-to-image registration was performed in two patients, and the results generated a virtual thoracoscopic view that was synchronized with the real thoracoscopic view. Results: The operations were started with video-assisted thoracic surgery (VATS), and the navigation system was activated at the same time. The virtual thoracoscopic view was synchronized with the real thoracoscopic view, which also simultaneously indicated the ‘area of lung cancer before induction therapy’ and the ‘optimal resection lines for obtaining a safe surgical margin’. We marked the optimal lines using an electric scalpel, and then performed lobectomy and chest wall resection with a sufficient surgical margin using these landmarks. Pathological examinations confirmed that the surgical margin was negative. No complications related to the navigation system were encountered during or after the procedures. Conclusions: Using this proposed navigation system, we could obtain a ‘CT-derived virtual intrathoracic 3D view of the patient’ that was aligned with the thoracoscopic view during surgery. The accurate identification of areas of cancer invasion before induction therapy using this system might be a useful for determining optimal surgical resection lines.

DOI： 10.21037/jtd.2019.12.108

Web of Science

Scopus

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/j.media.2019.101623

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1254/fpj.20013

PubMed

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

CiNii Research

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Progress in Biomedical Optics and Imaging - Proceedings of SPIE  

This paper presents an automated classification method of infective and non-infective diseases from anterior eye images. Treatments for cases of infective and non-infective diseases are different. Distinguishing them from anterior eye images is important to decide a treatment plan. Ophthalmologists distinguish them empirically. Quantitative classification of them based on computer assistance is necessary. We propose an automated classification method of anterior eye images into cases of infective or non-infective disease. Anterior eye images have large variations of the eye position and brightness of illumination. This makes the classification difficult. If we focus on the cornea, positions of opacified areas in the corneas are different between cases of the infective and non-infective diseases. Therefore, we solve the anterior eye image classification task by using an object detection approach targeting the cornea. This approach can be said as "anatomical structure focused image classification". We use the YOLOv3 object detection method to detect corneas of infective disease and corneas of non-infective disease. The detection result is used to define a classification result of an image. In our experiments using anterior eye images, 88.3% of images were correctly classified by the proposed method.

DOI： 10.1117/12.2549951

Web of Science

Scopus

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

The performance of deep learning based methods strongly relies on the number of datasets used for training. Many efforts have been made to increase the data in the medical image analysis field. However, unlike photography images, it is hard to generate centralized databases to collect medical images because of numerous technical, legal, and privacy issues. In this work, we study the use of federated learning between two institutions in a real-world setting to collaboratively train a model without sharing the raw data across national boundaries. We quantitatively compare the segmentation models obtained with federated learning and local training alone. Our experimental results show that federated learning models have higher generalizability than standalone training.

DOI： 10.1007/978-3-030-60548-3_19

Scopus

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：一般社団法人 映像情報メディア学会  

DOI： 10.3169/itej.74.909

CiNii Research

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Computer Assisted Radiology and Surgery  

Purpose : The purpose of this paper is to present a fully automated abdominal artery segmentation method from a CT volume. Three-dimensional (3D) blood vessel structure information is important for diagnosis and treatment. Information about blood vessels (including arteries) can be used in patient-specific surgical planning and intra-operative navigation. Since blood vessels have large inter-patient variations in branching patterns and positions, a patient-specific blood vessel segmentation method is necessary. Even though deep learning-based segmentation methods provide good segmentation accuracy among large organs, small organs such as blood vessels are not well segmented. We propose a deep learning-based abdominal artery segmentation method from a CT volume. Because the artery is one of small organs that is difficult to segment, we introduced an original training sample generation method and a three-plane segmentation approach to improve segmentation accuracy. Method : Our proposed method segments abdominal arteries from an abdominal CT volume with a fully convolutional network (FCN). To segment small arteries, we employ a 2D patch-based segmentation method and an area imbalance reduced training patch generation (AIRTPG) method. AIRTPG adjusts patch number imbalances between patches with artery regions and patches without them. These methods improved the segmentation accuracies of small artery regions. Furthermore, we introduced a three-plane segmentation approach to obtain clear 3D segmentation results from 2D patch-based processes. In the three-plane approach, we performed three segmentation processes using patches generated on axial, coronal, and sagittal planes and combined the results to generate a 3D segmentation result. Results : The evaluation results of the proposed method using 20 cases of abdominal CT volumes show that the averaged F-measure, precision, and recall rates were 87.1%, 85.8%, and 88.4%, respectively. This result outperformed our previous automated FCN-based segmentation method. Our method offers competitive performance compared to the previous blood vessel segmentation methods from 3D volumes. Conclusions : We developed an abdominal artery segmentation method using FCN. The 2D patch-based and AIRTPG methods effectively segmented the artery regions. In addition, the three-plane approach generated good 3D segmentation results.

DOI： 10.1007/s11548-019-02062-5

Web of Science

Scopus

PubMed

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1183/13993003.congress-2019.PA3168

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Micromachines  

Laparoscopic surgery is now a standard treatment for gastric cancer. Currently, the location of the gastric cancer is identified during laparoscopic surgery via the preoperative endoscopic injection of charcoal ink around the primary tumor; however, the wide spread of injected charcoal ink can make it difficult to accurately visualize the specific site of the tumor. To precisely identify the locations of gastric tumors, we developed a fluorescent detection system comprising clips with glass phosphor (Yb 3+ , Nd 3+ doped to Bi 2 O 3 -B 2 O 3 -based glasses, size: 2 mm × 1 mm × 3 mm) fixed in the stomach and a laparoscopic fluorescent detection system for clip-derived near-infrared (NIR) light (976 nm). We conducted two ex vivo experiments to evaluate the performance of this fluorescent detection system in an extirpated pig stomach and a freshly resected human stomach and were able to successfully detect NIR fluorescence emitted from the clip in the stomach through the stomach wall by the irradiation of excitation light (λ: 808 nm). These results suggest that the proposed combined NIR light-emitting clip and laparoscopic fluorescent detection system could be very useful in clinical practice for accurately identifying the location of a primary gastric tumor during laparoscopic surgery.

DOI： 10.3390/mi10020081

Scopus

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1117/12.2521618

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Progress in Biomedical Optics and Imaging - Proceedings of SPIE  

This paper presents a colonoscope tracking method utilizing a colon shape estimation method. CT colonography is used as a less-invasive colon diagnosis method. If colonic polyps or early-stage cancers are found, they are removed in a colonoscopic examination. In the colonoscopic examination, understanding where the colonoscope running in the colon is difficult. A colonoscope navigation system is necessary to reduce overlooking of polyps. We propose a colonoscope tracking method for navigation systems. Previous colonoscope tracking methods caused large tracking errors because they do not consider deformations of the colon during colonoscope insertions. We utilize the shape estimation network (SEN), which estimates deformed colon shape during colonoscope insertions. The SEN is a neural network containing long short-term memory (LSTM) layer. To perform colon shape estimation suitable to the real clinical situation, we trained the SEN using data obtained during colonoscope operations of physicians. The proposed tracking method performs mapping of the colonoscope tip position to a position in the colon using estimation results of the SEN. We evaluated the proposed method in a phantom study. We confirmed that tracking errors of the proposed method was enough small to perform navigation in the ascending, transverse, and descending colons.

DOI： 10.1117/12.2512729

Web of Science

Scopus

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1117/12.2293431

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1117/12.2293495

Web of Science

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

CiNii Research

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/978-3-030-00934-2_26

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1117/12.2293288

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：The Japanese Society of Medical Imaging Technology  

One of the most common tasks in medical imaging is semantic segmentation. Achieving this segmentation automatically has been an active area of research, but the task has been proven very challenging due to the large variation of anatomy across different patients. However, recent advances in deep learning have made it possible to significantly improve the performance of image recognition and semantic segmentation methods in the field of computer vision. Due to the data driven approaches of hierarchical feature learning in deep learning frameworks, these advances can be translated to medical images without much difficulty. Several variations of deep convolutional neural networks have been successfully applied to medical images. Especially fully convolutional architectures have been proven efficient for segmentation of 3D medical images. In this article, we describe how to build a 3D fully convolutional network (FCN) that can process 3D images in order to produce automatic semantic segmentations. The model is trained and evaluated on a clinical computed tomography (CT) dataset and shows stateof-the-art performance in multi-organ segmentation.

DOI： 10.11409/mit.36.63

CiNii Research

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/978-3-030-00937-3_21

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：The Japanese Society of Medical Imaging Technology  

In this paper, we propose a new airway segmentation algorithm from 3D chest CT volumes based on the volume of interest (VOI). The algorithm segments each bronchial branch by recognizing the airway regions from the trachea using the VOIs to segment each branch. A VOI is placed to envelop the branch currently being processed. Then a cavity enhancement filter is performed only inside the current VOI so that each branch is extracted. At the same time, we perform a leakage detection scheme to avoid any leakage regions inside the VOI. Next the gradient vector flow magnitude map and a tubular-likeness function are computed in each VOI. This assists the predictions of both the position and direction of the next child VOIs to detect the next child branches to continue the tracking algorithm. Finally, we unify all of the extracted airway regions to form a complete airway tree. We used a dataset that includes 50 standard-dose human chest CT volumes to evaluate our proposed algorithm. The average extraction rate was approximately 78.1% with a significantly decreased false positive rate compared to the previous method.

DOI： 10.11409/mit.36.133

CiNii Research

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：公益財団法人 日本心臓財団  

DOI： 10.11281/shinzo.49.1104

CiNii Research

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：SPIE-Intl Soc Optical Eng  

DOI： 10.1117/1.jmi.4.4.044502

CiNii Research

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：2017 IEEE International Conference on Cyborg and Bionic Systems, CBS 2017  

Bionic humanoid systems, which are elaborate human models with sensors, have been developed as a tool for quantitative evaluation of doctors' psychomotor skills and medical device performances. For creation of the elaborate human models, this study presents automated segmentation of head sectioned images using sparsely-Annotated data based on deep convolutional neural network. We applied the following fully convolutional networks (FCNs) to the sparse-Annotation-based segmentation: a standard FCN and a dilated convolution based FCN. To validate the availability of FCNs for segmentation of head structures from sparse annotation, we performed 8- and 243-label segmentation experiments using different two sets of head sectioned images in the Visible Korean Human project. In the segmentation experiments, only 10% of all images in each data set were used for training data. Both of the FCNs could achieve the mean segmentation accuracy of more than 85% in the 8-label segmentation. In the 243-label segmentation, though the mean segmentation accuracy was about 50%, the results suggested that the FCNs, especially the dilated convolution based FCNs, had potential to achieve accurate segmentation of anatomical structures, except for small-sized and complex-shaped tissues, even from sparse annotation.

DOI： 10.1109/CBS.2017.8266085

Scopus

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：一般社団法人 人工知能学会  

DOI： 10.11517/jsaialst.79.0_08

CiNii Research

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

This paper presents a fully automated atlas-based pancreas segmentation method from CT volumes utilizing 3D fully convolutional network (FCN) feature-based pancreas localization. Segmentation of the pancreas is difficult because it has larger inter-patient spatial variations than other organs. Previous pancreas segmentation methods failed to deal with such variations. We propose a fully automated pancreas segmentation method that contains novel localization and segmentation. Since the pancreas neighbors many other organs, its position and size are strongly related to the positions of the surrounding organs. We estimate the position and the size of the pancreas (localization) from global features by regression forests. As global features, we use intensity differences and 3D FCN deep learned features, which include automatically extracted essential features for segmentation. We chose 3D FCN features from a trained 3D U-Net, which is trained to perform multi-organ segmentation. The global features include both the pancreas and surrounding organ information. After localization, a patient-specific probabilistic atlas-based pancreas segmentation is performed. In evaluation results with 146 CT volumes, we achieved 60.6% of the Jaccard index and 73.9% of the Dice overlap.

DOI： 10.1007/978-3-319-67558-9_26

Web of Science

Scopus

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

CiNii Research

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：医用画像情報学会  

DOI： 10.11318/mii.34.1

CiNii Research

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

CiNii Research

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

CiNii Research

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1117/12.2255799

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

Web of Science

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Proceedings - 3rd IAPR Asian Conference on Pattern Recognition, ACPR 2015  

This paper presents a method for extracting an airway region from 3D chest CT volumes that uses a combination of tube enhancement filters, voxel classification based on machine learning methods and graph-cut algorithm. Lots of previous methods utilize region growing or level set algorithms without any prior knowledge of bronchi, which always fail when they reach to the peripheral bronchi. In this paper, a method of extraction based on airway shape and machine learning is proposed. The proposed method detects candidate voxels of bronchial regions by using two types of enhancement filters, and a classifier model is built for selecting the proper candidates regions based on intensity and shape features and finally the selected candidate voxels are connected by graph-cut algorithm. We applied this method on six cases of 3D chest CT volumes. The results show that this method can extract the smaller airway branches without leaking into the lung parenchyma areas.

DOI： 10.1109/ACPR.2015.7486563

Scopus

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1053/j.gastro.2016.04.004

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：一般社団法人 人工知能学会  

DOI： 10.11517/jsaialst.76.0_14

CiNii Research

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

In this paper, we present a preliminary report of a multiscale registration method between micro-focus X-ray CT (micro CT) volumes taken in different scales. 3D fine structures of target objects can be observed on micro CT volumes, which are difficult to observe on clinical CT volumes. Micro CT scanners can scan specimens in various resolutions. In their high resolution volumes, ultra fine structures of specimens can be observed, while scanned areas are limited to very small. On the other hand, in low resolution volumes, large areas can be captured, while fine structures of specimens are difficult to observe. The fusion volume of the high and low resolution volumes will have benefits of both. Because the difference of resolutions between the high and low resolution volumes may vary greatly, an intermediate resolution volume is required for successful fusion of volumes. To perform such volume fusion, a cascade multi-resolution registration technique is required. To register micro CT volumes that have quite different resolutions, we employ a cascade co-registration technique. In the cascade co-registration process, intermediate resolution volumes are used in a registration process of the high and low resolution volumes. In the registration between two volumes, we apply two steps registration techniques. In the first step, a block division is used to register two resolution volumes. Afterward, we estimate the fine spatial positions relating the registered two volumes using the Powell method. The registration result can be used to generate a fusion volume of the high and low resolution volumes.

DOI： 10.1007/978-3-319-43775-0_24

Scopus

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

CiNii Research

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

CiNii Research

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/j.media.2014.11.002

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1117/12.2214845

Web of Science

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

CiNii Research

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1117/12.2082488

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1088/1742-6596/619/1/012033

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1117/12.2082690

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1117/12.2081977

Web of Science

担当区分：責任著者   記述言語：日本語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1117/12.2081774

Web of Science

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Surgery Today  

Purposes: We performed a conversation analysis of the speech conducted among the surgical team during three-dimensional (3D)-printed liver model navigation for thrice or more repeated hepatectomy (TMRH). Methods: Seventeen patients underwent 3D-printed liver navigation surgery for TMRH. After transcription of the utterances recorded during surgery, the transcribed utterances were coded by the utterer, utterance object, utterance content, sensor, and surgical process during conversation. We then analyzed the utterances and clarified the association between the surgical process and conversation through the intraoperative reference of the 3D-printed liver. Results: In total, 130 conversations including 1648 segments were recorded. Utterance coding showed that the operator/assistant, 3D-printed liver/real liver, fact check (F)/plan check (Pc), visual check/tactile check, and confirmation of planned resection or preservation target (T)/confirmation of planned or ongoing resection line (L) accounted for 791/857, 885/763, 1148/500, 1208/440, and 1304/344 segments, respectively. The utterance’s proportions of assistants, F, F of T on 3D-printed liver, F of T on real liver, and Pc of L on 3D-printed liver were significantly higher during non-expert surgeries than during expert surgeries. Confirming the surgical process with both 3D-printed liver and real liver and performing planning using a 3D-printed liver facilitates the safe implementation of TMRH, regardless of the surgeon’s experience. Conclusions: The present study, using a unique conversation analysis, provided the first evidence for the clinical value of 3D-printed liver for TMRH for anatomical guidance of non-expert surgeons.

DOI： 10.1007/s00595-024-02835-9

Open Access

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Computer Assisted Radiology and Surgery  

Purpose: We propose a large-factor super-resolution (SR) method for performing SR on registered medical image datasets. Conventional SR approaches use low-resolution (LR) and high-resolution (HR) image pairs to train a deep convolutional neural network (DCN). However, LR–HR images in medical imaging are commonly acquired from different imaging devices, and acquiring LR–HR image pairs needs registration. Registered LR–HR images have registration errors inevitably. Using LR–HR images with registration error for training an SR DCN causes collapsed SR results. To address these challenges, we introduce a novel SR approach designed specifically for registered LR–HR medical images. Methods: We propose style-subnets-assisted generative latent bank for large-factor super-resolution (SGSR) trained with registered medical image datasets. Pre-trained generative models named generative latent bank (GLB), which stores rich image priors, can be applied in SR to generate realistic and faithful images. We improve GLB by newly introducing style-subnets-assisted GLB (S-GLB). We also propose a novel inter-uncertainty loss to boost our method’s performance. Introducing more spatial information by inputting adjacent slices further improved the results. Results: SGSR outperforms state-of-the-art (SOTA) supervised SR methods qualitatively and quantitatively on multiple datasets. SGSR achieved higher reconstruction accuracy than recently supervised baselines by increasing peak signal-to-noise ratio from 32.628 to 34.206 dB. Conclusion: SGSR performs large-factor SR while given a registered LR–HR medical image dataset with registration error for training. SGSR’s results have both realistic textures and accurate anatomical structures due to favorable quantitative and qualitative results. Experiments on multiple datasets demonstrated SGSR’s superiority over other SOTA methods. SR medical images generated by SGSR are expected to improve the accuracy of pre-surgery diagnosis and reduce patient burden.

DOI： 10.1007/s11548-023-03037-3

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Healthcare Technology Letters  

This study focuses on enhancing the inference speed of laparoscopic tool detection on embedded devices. Laparoscopy, a minimally invasive surgery technique, markedly reduces patient recovery times and postoperative complications. Real-time laparoscopic tool detection helps assisting laparoscopy by providing information for surgical navigation, and its implementation on embedded devices is gaining interest due to the portability, network independence and scalability of the devices. However, embedded devices often face computation resource limitations, potentially hindering inference speed. To mitigate this concern, the work introduces a two-fold modification to the YOLOv7 model: the feature channels and integrate RepBlock is halved, yielding the YOLOv7-RepFPN model. This configuration leads to a significant reduction in computational complexity. Additionally, the focal EIoU (efficient intersection of union) loss function is employed for bounding box regression. Experimental results on an embedded device demonstrate that for frame-by-frame laparoscopic tool detection, the proposed YOLOv7-RepFPN achieved an mAP of 88.2% (with IoU set to 0.5) on a custom dataset based on EndoVis17, and an inference speed of 62.9 FPS. Contrasting with the original YOLOv7, which garnered an 89.3% mAP and 41.8 FPS under identical conditions, the methodology enhances the speed by 21.1 FPS while maintaining detection accuracy. This emphasizes the effectiveness of the work.

DOI： 10.1049/htl2.12072

Scopus

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Healthcare Technology Letters  

The task of segmentation is integral to computer-aided surgery systems. Given the privacy concerns associated with medical data, collecting a large amount of annotated data for training is challenging. Unsupervised learning techniques, such as contrastive learning, have shown powerful capabilities in learning image-level representations from unlabelled data. This study leverages classification labels to enhance the accuracy of the segmentation model trained on limited annotated data. The method uses a multi-scale projection head to extract image features at various scales. The partitioning method for positive sample pairs is then improved to perform contrastive learning on the extracted features at each scale to effectively represent the differences between positive and negative samples in contrastive learning. Furthermore, the model is trained simultaneously with both segmentation labels and classification labels. This enables the model to extract features more effectively from each segmentation target class and further accelerates the convergence speed. The method was validated using the publicly available CholecSeg8k dataset for comprehensive abdominal cavity surgical segmentation. Compared to select existing methods, the proposed approach significantly enhances segmentation performance, even with a small labelled subset (1–10%) of the dataset, showcasing a superior intersection over union (IoU) score.

DOI： 10.1049/htl2.12069

Scopus

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Healthcare Technology Letters  

This paper focuses on a new and challenging problem related to instrument segmentation. This paper aims to learn a generalizable model from distributed datasets with various imperfect annotations. Collecting a large-scale dataset for centralized learning is usually impeded due to data silos and privacy issues. Besides, local clients, such as hospitals or medical institutes, may hold datasets with diverse and imperfect annotations. These datasets can include scarce annotations (many samples are unlabelled), noisy labels prone to errors, and scribble annotations with less precision. Federated learning (FL) has emerged as an attractive paradigm for developing global models with these locally distributed datasets. However, its potential in instrument segmentation has yet to be fully investigated. Moreover, the problem of learning from various imperfect annotations in an FL setup is rarely studied, even though it presents a more practical and beneficial scenario. This work rethinks instrument segmentation in such a setting and propose a practical FL framework for this issue. Notably, this approach surpassed centralized learning under various imperfect annotation settings. This method established a foundational benchmark, and future work can build upon it by considering each client owning various annotations and aligning closer with real-world complexities.

DOI： 10.1049/htl2.12068

Scopus

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Digestive Endoscopy  

Objectives: In this study we aimed to develop an artificial intelligence-based model for predicting postendoscopic retrograde cholangiopancreatography (ERCP) pancreatitis (PEP). Methods: We retrospectively reviewed ERCP patients at Nagoya University Hospital (NUH) and Toyota Memorial Hospital (TMH). We constructed two prediction models, a random forest (RF), one of the machine-learning algorithms, and a logistic regression (LR) model. First, we selected features of each model from 40 possible features. Then the models were trained and validated using three fold cross-validation in the NUH cohort and tested in the TMH cohort. The area under the receiver operating characteristic curve (AUROC) was used to assess model performance. Finally, using the output parameters of the RF model, we classified the patients into low-, medium-, and high-risk groups. Results: A total of 615 patients at NUH and 544 patients at TMH were enrolled. Ten features were selected for the RF model, including albumin, creatinine, biliary tract cancer, pancreatic cancer, bile duct stone, total procedure time, pancreatic duct injection, pancreatic guidewire-assisted technique without a pancreatic stent, intraductal ultrasonography, and bile duct biopsy. In the three fold cross-validation, the RF model showed better predictive ability than the LR model (AUROC 0.821 vs. 0.660). In the test, the RF model also showed better performance (AUROC 0.770 vs. 0.663, P = 0.002). Based on the RF model, we classified the patients according to the incidence of PEP (2.9%, 10.0%, and 23.9%). Conclusion: We developed an RF model. Machine-learning algorithms could be powerful tools to develop accurate prediction models.

DOI： 10.1111/den.14622

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Elsevier BV  

DOI： 10.1016/j.gie.2023.04.1087

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Computer Assisted Radiology and Surgery  

Purpose: Minimally invasive surgery (MIS) using a thoraco- or laparoscope is becoming a more common surgical technique. In MIS, a magnified view from a thoracoscope helps surgeons conduct precise operations. However, there is a risk of the visible area becoming narrow. To confirm that the operation field is safe, the surgeon will draw the thoracoscope back to check the marginal area of the target and insert it again many times during MIS. To reduce the surgeon’s load, we aim to visualize the entire thoracic cavity using a newly developed device called “panorama vision ring” (PVR). Method: The PVR is used instead of a wound retractor or a trocar. It is a ring-type socket with one big hole for the thoracoscope and four small holes for tiny cameras placed around the big hole. The views from the tiny cameras are fused into one wider view that visualizes the entire thoracic cavity. A surgeon can proceed with an operation by checking what exists outside of the thoracoscopic view. Also, she/he can check whether or not bleeding has occurred from the image of the entire cavity. Results: We evaluated the view-expansion ability of the PVR by using a three-dimensional full-scale thoracic model. The experimental results showed that the entire thoracic cavity could be visible in a panoramic view generated by the PVR. We also demonstrated pulmonary lobectomy in virtual MIS using the PVR. Surgeons could perform a pulmonary lobectomy while checking the entire cavity. Conclusion: We developed the PVR, which uses tiny auxiliary cameras to create a panoramic view of the entire thoracic cavity during MIS. We aim to make MIS safer for patients and more comfortable for surgeons through the development of the PVR.

DOI： 10.1007/s11548-023-02859-5

Scopus

PubMed

記述言語：日本語  

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Computer Methods in Biomechanics and Biomedical Engineering: Imaging and Visualization  

Oesophageal achalasia is a primary oesophageal motility disorder disease. To diagnose oesophagus achalasia, physicians recommend endoscopic evaluation of the oesophagus. However, a low sensitivity still accompanies esophagoscopy on oesophagus achalasia diagnosis. Thus, a quantitative diagnosis system is needed to support physicians diagnose achalasia from the esophagoscopy video. This paper proposes a Serial Multi-scale Network for classifying achalasia images from the esophagoscopy video. The proposed method contains two main components, a Dense-pooling Net, and a Serial Multi-scale Dilated encoder. We construct the Dense-pooling Net using a convolution neural network with dense mixed-pooling connections to extract features. We design the Serial Multi-scale Dilated encoder based on a residual-style dilated encoder. We combine the dilated encoder and spatial attention modules to focus on features we need. We trained and evaluated our method with a dataset that was extracted from several esophagoscopy videos of achalasia patients. The evaluation results reveal a state-of-the-art accuracy of achalasia diagnosis. Furthermore, we developed a real-time computer-aided achalasia diagnosis system with the trained network. In the real-time test, the achalasia diagnosis system can stably output the diagnosis results in only (Formula presented.) seconds. The extended experiments demonstrate that the constructed diagnosis system can diagnose achalasia from esophagoscopy videos.

DOI： 10.1080/21681163.2022.2159534

Scopus

CiNii Research

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Computer Methods in Biomechanics and Biomedical Engineering: Imaging and Visualization  

We propose a spatio-temporal mixing kinematic data estimation method to estimate the shape of the colon with deformations caused by colonoscope insertion. Endoscope tracking or a navigation system that navigates physicians to target positions is needed to reduce such complications as organ perforations. Although many previous methods focused to track bronchoscopes and surgical endoscopes, few number of colonoscope tracking methods were proposed because the colon largely deforms during colonoscope insertion. The deformation causes significant tracking errors. Colon deformation should be considered in the tracking process. We propose a colon shape estimation method using a Kinematic Spatio-Temporal data Mixer (KST-Mixer) that can be used during colonoscope insertions to the colon. Kinematic data of a colonoscope and the colon, including positions and directions of their centerlines, are obtained using electromagnetic and depth sensors. The proposed method separates the data into sub-groups along the spatial and temporal axes. The KST-Mixer extracts kinematic features and mix them along the axes multiple times. We evaluated colon shape estimation accuracies in phantom studies. The proposed method achieved 11.92 mm mean Euclidean distance error, the smallest of the previous methods. Statistical analysis indicated that the proposed method significantly reduced the error compared to the previous methods.

DOI： 10.1080/21681163.2022.2151938

Scopus

CiNii Research

記述言語：日本語  

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Progress in Biomedical Optics and Imaging - Proceedings of SPIE  

Medical image analysis approaches such as data augmentation and domain adaption need huge amounts of realistic medical images. Generating realistic medical images by machine learning is a feasible approach. We propose L-former, a lightweight Transformer for realistic medical image generation. L-former can generate more reliable and realistic medical images than recent generative adversarial networks (GANs). Meanwhile, L-former does not consume as high computational cost as conventional Transformer-based generative models. L-former uses Transformers to generate low-resolution feature vectors at shallow layers, and uses convolutional neural networks to generate high-resolution realistic medical images at deep layers. Experimental results showed that L-former outperformed conventional GANs by FID scores 33.79 and 76.85 on two datasets, respectively. We further conducted a downstream study by using the images generated by L-former to perform a super-resolution task. A high PSNR score of 27.87 proved L-former’s ability to generate reliable images for super-resolution and showed its potential for applications in medical diagnosis.

DOI： 10.1117/12.2653776

Web of Science

Scopus

DOI： 10.1080/21681163.2022.2155579

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Scientific Reports  

This study presents a novel framework for classifying and visualizing pneumonia induced by COVID-19 from CT images. Although many image classification methods using deep learning have been proposed, in the case of medical image fields, standard classification methods are unable to be used in some cases because the medical images that belong to the same category vary depending on the progression of the symptoms and the size of the inflamed area. In addition, it is essential that the models used be transparent and explainable, allowing health care providers to trust the models and avoid mistakes. In this study, we propose a classification method using contrastive learning and an attention mechanism. Contrastive learning is able to close the distance for images of the same category and generate a better feature space for classification. An attention mechanism is able to emphasize an important area in the image and visualize the location related to classification. Through experiments conducted on two-types of classification using a three-fold cross validation, we confirmed that the classification accuracy was significantly improved; in addition, a detailed visual explanation was achieved comparison with conventional methods.

DOI： 10.1038/s41598-022-24936-6

Scopus

PubMed

その他リンク： https://www.nature.com/articles/s41598-022-24936-6

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Computer Assisted Radiology and Surgery  

Purpose: Precise polyp detection and localisation are essential for colonoscopy diagnosis. Statistical machine learning with a large-scale data set can contribute to the construction of a computer-aided diagnosis system for the prevention of overlooking and miss-localisation of a polyp in colonoscopy. We propose new visual explaining methods for a well-trained object detector, which achieves fast and accurate polyp detection with a bounding box towards a precise automated polyp localisation. Method: We refine gradient-weighted class activation mapping for more accurate highlighting of important patterns in processing a convolutional neural network. Extending the refined mapping into multiscaled processing, we define object activation mapping that highlights important object patterns in an image for a detection task. Finally, we define polyp activation mapping to achieve precise polyp localisation by integrating adaptive local thresholding into object activation mapping. We experimentally evaluate the proposed visual explaining methods with four publicly available databases. Results: The refined mapping visualises important patterns in each convolutional layer more accurately than the original gradient-weighted class activation mapping. The object activation mapping clearly visualises important patterns in colonoscopic images for polyp detection. The polyp activation mapping localises the detected polyps in ETIS-Larib, CVC-Clinic and Kvasir-SEG database with mean Dice scores of 0.76, 0.72 and 0.72, respectively. Conclusions: We developed new visual explaining methods for a convolutional neural network by refining and extending gradient-weighted class activation mapping. Experimental results demonstrated the validity of the proposed methods by showing that accurate visualisation of important patterns and localisation of polyps in a colonoscopic image. The proposed visual explaining methods are useful for the interpreting and applying a trained polyp detector.

DOI： 10.1007/s11548-022-02696-y

Web of Science

Scopus

PubMed

DOI： 10.1007/978-3-031-18814-5_7

DOI： 10.1007/978-3-031-16449-1_70

記述言語：日本語  

DOI： 10.1117/12.2613317

DOI： 10.1117/12.2613298

DOI： 10.1117/12.2612830

DOI： 10.1117/12.2612829

DOI： 10.1117/12.2611672

DOI： 10.1117/12.2611422

DOI： 10.1117/12.2611311

DOI： 10.1117/12.2612827

DOI： 10.1117/12.2612535

DOI： 10.1117/12.2611441

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Computer Assisted Radiology and Surgery  

Purpose: Artery contrasted computed tomography (CT) enables accurate observations of the arteries and surrounding structures, thus being widely used for the diagnosis of diseases such as aneurysm. To avoid the complications caused by contrast agent, this paper proposes an aorta-aware deep learning method to synthesize artery contrasted CT volume form non-contrast CT volume. Methods: By introducing auxiliary multi-resolution segmentation tasks in the generator, we force the proposed network to focus on the regions of aorta and the other vascular structures. Then, the segmentation results produced by the auxiliary tasks were used to extract aorta. The detection of abnormal CT images containing aneurysm was implemented by estimating the maximum axial radius of aorta. Results: In comparison with the baseline models, the proposed network with auxiliary tasks achieved better performances with higher peak signal-noise ratio value. In aorta regions which are supposed to be the main region of interest in many clinic scenarios, the average improvement can be up to 0.33dB. Using the synthesized artery contrasted CT, the F score of aneurysm detection achieved 0.58 at slice level and 0.85 at case level. Conclusion: This study tries to address the problem of non-contrast to artery contrasted CT modality translation by employing a deep learning model with aorta awareness. The auxiliary tasks help the proposed model focus on aorta regions and synthesize results with clearer boundaries. Additionally, the synthesized artery contrasted CT shows potential in identifying slices with abdominal aortic aneurysm, and may provide an option for patients with contrast agent allergy.

DOI： 10.1007/s11548-021-02492-0

Web of Science

Scopus

PubMed

担当区分：最終著者  

DOI： 10.1002/mp.15192

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Computer Assisted Radiology and Surgery  

Purpose: The size information of detected polyps is an essential factor for diagnosis in colon cancer screening. For example, adenomas and sessile serrated polyps that are ≥ 10 mm are considered advanced, and shorter surveillance intervals are recommended for smaller polyps. However, sometimes the subjective estimations of endoscopists are incorrect and overestimate the sizes. To circumvent these difficulties, we developed a method for automatic binary polyp-size classification between two polyp sizes: from 1 to 9 mm and ≥ 10 mm. Method: We introduce a binary polyp-size classification method that estimates a polyp’s three-dimensional spatial information. This estimation is comprised of polyp localisation and depth estimation. The combination of location and depth information expresses a polyp’s three-dimensional shape. In experiments, we quantitatively and qualitatively evaluate the proposed method using 787 polyps of both protruded and flat types. Results: The proposed method’s best classification accuracy outperformed the fine-tuned state-of-the-art image classification methods. Post-processing of sequential voting increased the classification accuracy and achieved classification accuracy of 0.81 and 0.88 for polyps ranging from 1 to 9 mm and others that are ≥ 10 mm. Qualitative analysis revealed the importance of polyp localisation even in polyp-size classification. Conclusions: We developed a binary polyp-size classification method by utilising the estimated three-dimensional shape of a polyp. Experiments demonstrated accurate classification for both protruded- and flat-type polyps, even though the flat type have ambiguous boundary between a polyp and colon wall.

DOI： 10.1007/s11548-021-02477-z

DOI： 10.1007/s11548-021-02477-z

Web of Science

Scopus

PubMed

その他リンク： https://link.springer.com/article/10.1007/s11548-021-02477-z/fulltext.html

担当区分：最終著者  

DOI： 10.1007/978-3-030-90874-4_10

DOI： 10.1007/978-3-030-90874-4_1

記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Endoscopy  

DOI： 10.1055/a-1261-2944

Web of Science

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Computer Assisted Radiology and Surgery  

Purpose: A three-dimensional (3D) structure extraction technique viewed from a two-dimensional image is essential for the development of a computer-aided diagnosis (CAD) system for colonoscopy. However, a straightforward application of existing depth-estimation methods to colonoscopic images is impossible or inappropriate due to several limitations of colonoscopes. In particular, the absence of ground-truth depth for colonoscopic images hinders the application of supervised machine learning methods. To circumvent these difficulties, we developed an unsupervised and accurate depth-estimation method. Method: We propose a novel unsupervised depth-estimation method by introducing a Lambertian-reflection model as an auxiliary task to domain translation between real and virtual colonoscopic images. This auxiliary task contributes to accurate depth estimation by maintaining the Lambertian-reflection assumption. In our experiments, we qualitatively evaluate the proposed method by comparing it with state-of-the-art unsupervised methods. Furthermore, we present two quantitative evaluations of the proposed method using a measuring device, as well as a new 3D reconstruction technique and measured polyp sizes. Results: Our proposed method achieved accurate depth estimation with an average estimation error of less than 1 mm for regions close to the colonoscope in both of two types of quantitative evaluations. Qualitative evaluation showed that the introduced auxiliary task reduces the effects of specular reflections and colon wall textures on depth estimation and our proposed method achieved smooth depth estimation without noise, thus validating the proposed method. Conclusions: We developed an accurate depth-estimation method with a new type of unsupervised domain translation with the auxiliary task. This method is useful for analysis of colonoscopic images and for the development of a CAD system since it can extract accurate 3D information.

DOI： 10.1007/s11548-021-02398-x

Web of Science

Scopus

PubMed

担当区分：筆頭著者  

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Gastrointestinal Endoscopy  

Background and Aims: Artificial intelligence (AI)–assisted polyp detection systems for colonoscopic use are currently attracting attention because they may reduce the possibility of missed adenomas. However, few systems have the necessary regulatory approval for use in clinical practice. We aimed to develop an AI-assisted polyp detection system and to validate its performance using a large colonoscopy video database designed to be publicly accessible. Methods: To develop the deep learning–based AI system, 56,668 independent colonoscopy images were obtained from 5 centers for use as training images. To validate the trained AI system, consecutive colonoscopy videos taken at a university hospital between October 2018 and January 2019 were searched to construct a database containing polyps with unbiased variance. All images were annotated by endoscopists according to the presence or absence of polyps and the polyps’ locations with bounding boxes. Results: A total of 1405 videos acquired during the study period were identified for the validation database, 797 of which contained at least 1 polyp. Of these, 100 videos containing 100 independent polyps and 13 videos negative for polyps were randomly extracted, resulting in 152,560 frames (49,799 positive frames and 102,761 negative frames) for the database. The AI showed 90.5% sensitivity and 93.7% specificity for frame-based analysis. The per-polyp sensitivities for all, diminutive, protruded, and flat polyps were 98.0%, 98.3%, 98.5%, and 97.0%, respectively. Conclusions: Our trained AI system was validated with a new large publicly accessible colonoscopy database and could identify colorectal lesions with high sensitivity and specificity. (Clinical trial registration number: UMIN 000037064.)

DOI： 10.1016/j.gie.2020.07.060

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Gastroenterology  

Background & Aims: In accordance with guidelines, most patients with T1 colorectal cancers (CRC) undergo surgical resection with lymph node dissection, despite the low incidence (∼10%) of metastasis to lymph nodes. To reduce unnecessary surgical resections, we used artificial intelligence to build a model to identify T1 colorectal tumors at risk for metastasis to lymph node and validated the model in a separate set of patients. Methods: We collected data from 3134 patients with T1 CRC treated at 6 hospitals in Japan from April 1997 through September 2017 (training cohort). We developed a machine-learning artificial neural network (ANN) using data on patients’ age and sex, as well as tumor size, location, morphology, lymphatic and vascular invasion, and histologic grade. We then conducted the external validation on the ANN model using independent 939 patients at another hospital during the same period (validation cohort). We calculated areas under the receiver operator characteristics curves (AUCs) for the ability of the model and US guidelines to identify patients with lymph node metastases. Results: Lymph node metastases were found in 319 (10.2%) of 3134 patients in the training cohort and 79 (8.4%) of /939 patients in the validation cohort. In the validation cohort, the ANN model identified patients with lymph node metastases with an AUC of 0.83, whereas the guidelines identified patients with lymph node metastases with an AUC of 0.73 (P < .001). When the analysis was limited to patients with initial endoscopic resection (n = 517), the ANN model identified patients with lymph node metastases with an AUC of 0.84 and the guidelines identified these patients with an AUC of 0.77 (P = .005). Conclusions: The ANN model outperformed guidelines in identifying patients with T1 CRCs who had lymph node metastases. This model might be used to determine which patients require additional surgery after endoscopic resection of T1 CRCs. UMIN Clinical Trials Registry no: UMIN000038609

DOI： 10.1053/j.gastro.2020.09.027

DOI： 10.1053/j.gastro.2020.09.027

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Computer Assisted Radiology and Surgery  

Purpose: In recent years, fully convolutional networks (FCNs) have been applied to various medical image segmentation tasks. However, it is difficult to generate a large amount of high-quality annotation data to train FCNs for medical image segmentation. Thus, it is desired to achieve high segmentation performances even from incomplete training data. We aim to evaluate performance of FCNs to clean noises and interpolate labels from noisy and sparsely given label images. Methods: To evaluate the label cleaning and propagation performance of FCNs, we used 2D and 3D FCNs to perform volumetric brain segmentation from magnetic resonance image volumes, based on network training on incomplete training datasets from noisy and sparse annotation. Results: The experimental results using pseudo-incomplete training data showed that both 2D and 3D FCNs could provide improved segmentation results from the incomplete training data, especially by using three orthogonal annotation images for network training. Conclusion: This paper presented a validation for label cleaning and propagation based on FCNs. FCNs might have the potential to achieve improved segmentation performances even from sparse annotation data including possible noises by manual annotation, which can be an important clue to more efficient annotation.

DOI： 10.1007/s11548-021-02312-5

Web of Science

Scopus

PubMed

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1117/12.2580641

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1117/12.2581818

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1117/12.2582066

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1117/12.2582088

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1117/12.2582660

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1117/12.2581261

担当区分：筆頭著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s00595-020-02071-x

記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Digestive Endoscopy  

The global incidence and mortality rate of colorectal cancer remains high. Colonoscopy is regarded as the gold standard examination for detecting and eradicating neoplastic lesions. However, there are some uncertainties in colonoscopy practice that are related to limitations in human performance. First, approximately one-fourth of colorectal neoplasms are missed on a single colonoscopy. Second, it is still difficult for non-experts to perform adequately regarding optical biopsy. Third, recording of some quality indicators (e.g. cecal intubation, bowel preparation, and withdrawal speed) which are related to adenoma detection rate, is sometimes incomplete. With recent improvements in machine learning techniques and advances in computer performance, artificial intelligence-assisted computer-aided diagnosis is being increasingly utilized by endoscopists. In particular, the emergence of deep-learning, data-driven machine learning techniques have made the development of computer-aided systems easier than that of conventional machine learning techniques, the former currently being considered the standard artificial intelligence engine of computer-aided diagnosis by colonoscopy. To date, computer-aided detection systems seem to have improved the rate of detection of neoplasms. Additionally, computer-aided characterization systems may have the potential to improve diagnostic accuracy in real-time clinical practice. Furthermore, some artificial intelligence-assisted systems that aim to improve the quality of colonoscopy have been reported. The implementation of computer-aided system clinical practice may provide additional benefits such as helping in educational poorly performing endoscopists and supporting real-time clinical decision-making. In this review, we have focused on computer-aided diagnosis during colonoscopy reported by gastroenterologists and discussed its status, limitations, and future prospects.

DOI： 10.1111/den.13847

Web of Science

Scopus

PubMed

記述言語：日本語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Progress in Biomedical Optics and Imaging - Proceedings of SPIE  

Endoscopic submucosal dissection is a minimally invasive treatment for early gastric cancer. In endoscopic submucosal dissection, a physician directly removes the mucosa around the lesion under internal endoscopy by using the flush knife. However, the flush knife may accidentally pierce the colonic wall and generate a perforation on it. If physicians overlooking a small perforation, a patient may need emergency open surgery, since a perforation can easily cause peritonitis. For the prevention of overlooking of perforations, a computer-aided diagnosis system has a potential demand. We believe automatic perforation detection and localization function is very useful for the analysis of endoscopic submucosal dissection videos for the development of a computeraided diagnosis system. At current stage, the research of perforation detection and localization progress slowly, automatic image-based perforation detection is very challenge. Thus, we devote to the development of detection and localization of perforations in colonoscopic videos. In this paper, we proposed a supervised-learning method for perforations detection and localization in colonoscopic videos. This method uses dense layers in YOLO-v3 instead of residual units, and a combination of binary cross entropy and generalized intersection over union loss as the loss function in the training process. This method achieved 0.854 accuracy, 0.850 AUC score and 0.884 mean average precision for perforation detection and localization, respectively, as an initial study

DOI： 10.1117/12.2582300

Web of Science

Scopus

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：International Journal of Computer Assisted Radiology and Surgery  

Purpose: An endocytoscope is a new type of endoscope that enables users to perform conventional endoscopic observation and ultramagnified observation at the cell level. Although endocytoscopy is expected to improve the cost-effectiveness of colonoscopy, endocytoscopic image diagnosis requires much knowledge and high-level experience for physicians. To circumvent this difficulty, we developed a robust endocytoscopic (EC) image classification method for the construction of a computer-aided diagnosis (CAD) system, since real-time CAD can resolve accuracy issues and reduce interobserver variability. Method: We propose a novel feature extraction method by introducing higher-order symmetric tensor analysis to the computation of multi-scale topological statistics on an image, and we integrate this feature extraction with EC image classification. We experimentally evaluate the classification accuracy of our proposed method by comparing it with three deep learning methods. We conducted this comparison by using our large-scale multi-hospital dataset of about 55,000 images of over 3800 patients. Results: Our proposed method achieved an average 90% classification accuracy for all the images in four hospitals even though the best deep learning method achieved 95% classification accuracy for images in only one hospital. In the case with a rejection option, the proposed method achieved expert-level accurate classification. These results demonstrate the robustness of our proposed method against pit pattern variations, including differences of colours, contrasts, shapes, and hospitals. Conclusions: We developed a robust EC image classification method with novel feature extraction. This method is useful for the construction of a practical CAD system, since it has sufficient generalisation ability.

DOI： 10.1007/s11548-020-02255-3

Web of Science

Scopus

PubMed

記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：(一社)日本消化器外科学会  

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1002/mp.14442

担当区分：筆頭著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者   記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Gastrointestinal Endoscopy  

Background and Aims: Artificial intelligence (AI) is being implemented in colonoscopy practice, but no study has investigated whether AI is cost saving. We aimed to quantify the cost reduction using AI as an aid in the optical diagnosis of colorectal polyps. Methods: This study is an add-on analysis of a clinical trial that investigated the performance of AI for differentiating colorectal polyps (ie, neoplastic versus non-neoplastic). We included all patients with diminutive (≤5 mm) rectosigmoid polyps in the analyses. The average colonoscopy cost was compared for 2 scenarios: (1) a diagnose-and-leave strategy supported by the AI prediction (ie, diminutive rectosigmoid polyps were not removed when predicted as non-neoplastic), and (2) a resect-all-polyps strategy. Gross annual costs for colonoscopies were also calculated based on the number and reimbursement of colonoscopies conducted under public health insurances in 4 countries. Results: Overall, 207 patients with 250 diminutive rectosigmoid polyps (104 neoplastic, 144 non-neoplastic, and 2 indeterminate) were included. AI correctly differentiated neoplastic polyps with 93.3% sensitivity, 95.2% specificity, and 95.2% negative predictive value. Thus, 105 polyps were removed and 145 were left under the diagnose-and-leave strategy, which was estimated to reduce the average colonoscopy cost and the gross annual reimbursement for colonoscopies by 18.9% and US$149.2 million in Japan, 6.9% and US$12.3 million in England, 7.6% and US$1.1 million in Norway, and 10.9% and US$85.2 million in the United States, respectively, compared with the resect-all-polyps strategy. Conclusions: The use of AI to enable the diagnose-and-leave strategy results in substantial cost reductions for colonoscopy.

DOI： 10.1016/j.gie.2020.03.3759

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1002/mp.14372

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（研究会，シンポジウム資料等）  

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（研究会，シンポジウム資料等）  

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

記述言語：日本語  

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

担当区分：最終著者   記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

担当区分：最終著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

担当区分：責任著者   記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：(社)日本大腸検査学会  

記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Journal of Medical Imaging  

Purpose: High-resolution cardiac imaging and fiber analysis methods are required to understand cardiac anatomy. Although refraction-contrast x-ray CT (RCT) has high soft tissue contrast, it cannot be commonly used because it requires a synchrotron system. Microfocus x-ray CT (μCT) is another commercially available imaging modality. Approach: We evaluate the usefulness of μCT for analyzing fibers by quantitatively and objectively comparing the results with RCT. To do so, we scanned a rabbit heart by both modalities with our original protocol of prepared materials and compared their image-based analysis results, including fiber orientation estimation and fiber tracking. Results: Fiber orientations estimated by two modalities were closely resembled under the correlation coefficient of 0.63. Tracked fibers from both modalities matched well the anatomical knowledge that fiber orientations are different inside and outside of the left ventricle. However, the μCT volume caused incorrect tracking around the boundaries caused by stitching scanning. Conclusions: Our experimental results demonstrated that μCT scanning can be used for cardiac fiber analysis, although further investigation is required in the differences of fiber analysis results on RCT and μCT.

DOI： 10.1117/1.JMI.7.2.026001

Web of Science

Scopus

PubMed

CiNii Research

担当区分：筆頭著者,　最終著者,　責任著者   記述言語：日本語  

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

担当区分：最終著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

担当区分：最終著者,　責任著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

担当区分：最終著者   記述言語：英語   掲載種別：研究論文（学術雑誌）  

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（研究会，シンポジウム資料等）  

担当区分：最終著者   記述言語：日本語   掲載種別：研究論文（研究会，シンポジウム資料等）  

担当区分：最終著者   記述言語：日本語   掲載種別：研究論文（研究会，シンポジウム資料等）  

担当区分：最終著者   記述言語：日本語   掲載種別：研究論文（研究会，シンポジウム資料等）  

担当区分：最終著者,　責任著者   記述言語：日本語   掲載種別：研究論文（研究会，シンポジウム資料等）  

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Healthcare Technology Letters  

This Letter presents a stable polyp-scene classification method with low false positive (FP) detection. Precise automated polyp detection during colonoscopies is essential for preventing colon-cancer deaths. There is, therefore, a demand for a computer-assisted diagnosis (CAD) system for colonoscopies to assist colonoscopists. A high-performance CAD system with spatiotemporal feature extraction via a three-dimensional convolutional neural network (3D CNN) with a limited dataset achieved about 80% detection accuracy in actual colonoscopic videos. Consequently, further improvement of a 3D CNN with larger training data is feasible. However, the ratio between polyp and non-polyp scenes is quite imbalanced in a large colonoscopic video dataset. This imbalance leads to unstable polyp detection. To circumvent this, the authors propose an efficient and balanced learning technique for deep residual learning. The authors’ method randomly selects a subset of non-polyp scenes whose number is the same number of still images of polyp scenes at the beginning of each epoch of learning. Furthermore, they introduce post-processing for stable polyp-scene classification. This post-processing reduces the FPs that occur in the practical application of polyp-scene classification. They evaluate several residual networks with a large polyp-detection dataset consisting of 1027 colonoscopic videos. In the scene-level evaluation, their proposed method achieves stable polyp-scene classification with 0.86 sensitivity and 0.97 specificity.

DOI： 10.1049/htl.2019.0079

Web of Science

Scopus

PubMed

CiNii Research

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Healthcare Technology Letters  

A realistic image generation method for visualisation in endoscopic simulation systems is proposed in this study. Endoscopic diagnosis and treatment are performed in many hospitals. To reduce complications related to endoscope insertions, endoscopic simulation systems are used for training or rehearsal of endoscope insertions. However, current simulation systems generate non-realistic virtual endoscopic images. To improve the value of the simulation systems, improvement of the reality of their generated images is necessary. The authors propose a realistic image generation method for endoscopic simulation systems. Virtual endoscopic images are generated by using a volume rendering method from a CT volume of a patient. They improve the reality of the virtual endoscopic images using a virtual-to-real image-domain translation technique. The image-domain translator is implemented as a fully convolutional network (FCN). They train the FCN by minimising a cycle consistency loss function. The FCN is trained using unpaired virtual and real endoscopic images. To obtain high-quality image-domain translation results, they perform an image cleansing to the real endoscopic image set. They tested to use the shallow U-Net, U-Net, deep U-Net, and U-Net having residual units as the image-domain translator. The deep U-Net and U-Net having residual units generated quite realistic images.

DOI： 10.1049/htl.2019.0071

Web of Science

Scopus

PubMed

CiNii Research

記述言語：英語   掲載種別：研究論文（学術雑誌）  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：日本語  

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/j.compmedimag.2019.101642

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Clinical Gastroenterology and Hepatology  

Background & Aims: Precise optical diagnosis of colorectal polyps could improve the cost-effectiveness of colonoscopy and reduce polypectomy-related complications. However, it is difficult for community-based non-experts to obtain sufficient diagnostic performance. Artificial intelligence-based systems have been developed to analyze endoscopic images; they identify neoplasms with high accuracy and low interobserver variation. We performed a multi-center study to determine the diagnostic accuracy of EndoBRAIN, an artificial intelligence-based system that analyzes cell nuclei, crypt structure, and microvessels in endoscopic images, in identification of colon neoplasms. Methods: The EndoBRAIN system was initially trained using 69,142 endocytoscopic images, taken at 520-fold magnification, from patients with colorectal polyps who underwent endoscopy at 5 academic centers in Japan from October 2017 through March 2018. We performed a retrospective comparative analysis of the diagnostic performance of EndoBRAIN vs that of 30 endoscopists (20 trainees and 10 experts); the endoscopists assessed images from 100 cases produced via white-light microscopy, endocytoscopy with methylene blue staining, and endocytoscopy with narrow-band imaging. EndoBRAIN was used to assess endocytoscopic, but not white-light, images. The primary outcome was the accuracy of EndoBrain in distinguishing neoplasms from non-neoplasms, compared with that of endoscopists, using findings from pathology analysis as the reference standard. Results: In analysis of stained endocytoscopic images, EndoBRAIN identified colon lesions with 96.9% sensitivity (95% CI, 95.8%–97.8%), 100% specificity (95% CI, 99.6%–100%), 98% accuracy (95% CI, 97.3%–98.6%), a 100% positive-predictive value (95% CI, 99.8%–100%), and a 94.6% negative-predictive (95% CI, 92.7%–96.1%); these values were all significantly greater than those of the endoscopy trainees and experts. In analysis of narrow-band images, EndoBRAIN distinguished neoplastic from non-neoplastic lesions with 96.9% sensitivity (95% CI, 95.8–97.8), 94.3% specificity (95% CI, 92.3–95.9), 96.0% accuracy (95% CI, 95.1–96.8), a 96.9% positive-predictive value, (95% CI, 95.8–97.8), and a 94.3% negative-predictive value (95% CI, 92.3–95.9); these values were all significantly higher than those of the endoscopy trainees, sensitivity and negative-predictive value were significantly higher but the other values are comparable to those of the experts. Conclusions: EndoBRAIN accurately differentiated neoplastic from non-neoplastic lesions in stained endocytoscopic images and endocytoscopic narrow-band images, when pathology findings were used as the standard. This technology has been authorized for clinical use by the Japanese regulatory agency and should be used in endoscopic evaluation of small polyps more widespread clinical settings. UMIN clinical trial no: UMIN000028843.

DOI： 10.1016/j.cgh.2019.09.009

Scopus

PubMed

記述言語：日本語  

記述言語：日本語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Digestive Endoscopy  

With recent breakthroughs in artificial intelligence, computer-aided diagnosis (CAD) for upper gastrointestinal endoscopy is gaining increasing attention. Main research focuses in this field include automated identification of dysplasia in Barrett's esophagus and detection of early gastric cancers. By helping endoscopists avoid missing and mischaracterizing neoplastic change in both the esophagus and the stomach, these technologies potentially contribute to solving current limitations of gastroscopy. Currently, optical diagnosis of early-stage dysplasia related to Barrett's esophagus can be precisely achieved only by endoscopists proficient in advanced endoscopic imaging, and the false-negative rate for detecting gastric cancer is approximately 10%. Ideally, these novel technologies should work during real-time gastroscopy to provide on-site decision support for endoscopists regardless of their skill; however, previous studies of these topics remain ex vivo and experimental in design. Therefore, the feasibility, effectiveness, and safety of CAD for upper gastrointestinal endoscopy in clinical practice remain unknown, although a considerable number of pilot studies have been conducted by both engineers and medical doctors with excellent results. This review summarizes current publications relating to CAD for upper gastrointestinal endoscopy from the perspective of endoscopists and aims to indicate what is required for future research and implementation in clinical practice.

DOI： 10.1111/den.13317

Web of Science

Scopus

PubMed

CiNii Research

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1111 den.13317

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：英語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語  

記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

記述言語：英語   掲載種別：研究論文（学術雑誌）  

記述言語：日本語   掲載種別：研究論文（その他学術会議資料等）  

記述言語：英語   掲載種別：研究論文（学術雑誌）   出版者・発行元：Digestive Endoscopy  

Background and Aim: Application of artificial intelligence in medicine is now attracting substantial attention. In the field of gastrointestinal endoscopy, computer-aided diagnosis (CAD) for colonoscopy is the most investigated area, although it is still in the preclinical phase. Because colonoscopy is carried out by humans, it is inherently an imperfect procedure. CAD assistance is expected to improve its quality regarding automated polyp detection and characterization (i.e. predicting the polyp's pathology). It could help prevent endoscopists from missing polyps as well as provide a precise optical diagnosis for those detected. Ultimately, these functions that CAD provides could produce a higher adenoma detection rate and reduce the cost of polypectomy for hyperplastic polyps. Methods and Results: Currently, research on automated polyp detection has been limited to experimental assessments using an algorithm based on ex vivo videos or static images. Performance for clinical use was reported to have >90% sensitivity with acceptable specificity. In contrast, research on automated polyp characterization seems to surpass that for polyp detection. Prospective studies of in vivo use of artificial intelligence technologies have been reported by several groups, some of which showed a >90% negative predictive value for differentiating diminutive (≤5 mm) rectosigmoid adenomas, which exceeded the threshold for optical biopsy. Conclusion: We introduce the potential of using CAD for colonoscopy and describe the most recent conditions for regulatory approval for artificial intelligence-assisted medical devices.

DOI： 10.1111/den.13340

Web of Science

Scopus

PubMed

CiNii Research

記述言語：日本語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語  

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）