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

受賞区分：国際学会・会議・シンポジウム等の賞  受賞国：アメリカ合衆国

受賞区分：国際学会・会議・シンポジウム等の賞  受賞国：アメリカ合衆国

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：日本国

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

出版者・発行元：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

The intestine is an essential digestive organ that can cause serious health problems once diseased. This paper proposes a method for intestine segmentation to intestine obstruction diagnosis assistance called multi-dimensional U-Net (M U-Net). We employ two encoders to extract features from two-dimensional (2D) CT slices and three-dimensional (3D) CT patches. These two encoders collaborate to enhance the segmentation accuracy of the model. Additionally, we incorporate deep supervision with the M U-Net to reduce the limitation of training with sparse label data sets. The experimental results demonstrated that the Dice of the proposed method was 73.22%, the recall was 79.89%, and the precision was 70.61%.

DOI： 10.1007/978-3-031-47076-9_14

Scopus

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：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

researchmap

記述言語：英語   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

Web of Science

Scopus

PubMed

記述言語：英語   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

Web of Science

Scopus

PubMed

記述言語：英語   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

Web of Science

Scopus

PubMed

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

researchmap

記述言語：英語   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

Web of Science

Scopus

PubMed

記述言語：英語   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

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

Scopus

PubMed

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

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

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

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

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

Web of Science

Scopus

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

Web of Science

Scopus

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

researchmap

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

担当区分：筆頭著者,　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  

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

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

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

記述言語：日本語   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

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

出版者・発行元：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

Developing a generalized segmentation model capable of simultaneously delineating multiple organs and diseases is highly desirable. Federated learning (FL) is a key technology enabling the collaborative development of a model without exchanging training data. However, the limited access to fully annotated training data poses a major challenge to training generalizable models. We propose “ConDistFL”, a framework to solve this problem by combining FL with knowledge distillation. Local models can extract the knowledge of unlabeled organs and tumors from partially annotated data from the global model with an adequately designed conditional probability representation. We validate our framework on four distinct partially annotated abdominal CT datasets from the MSD and KiTS19 challenges. The experimental results show that the proposed framework significantly outperforms FedAvg and FedOpt baselines. Moreover, the performance on an external test dataset demonstrates superior generalizability compared to models trained on each dataset separately. Our ablation study suggests that ConDistFL can perform well without frequent aggregation, reducing the communication cost of FL. Our implementation will be available at https://github.com/NVIDIA/NVFlare/tree/main/research/condist-fl.

DOI： 10.1007/978-3-031-47401-9_30

Scopus

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

記述言語：日本語   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

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  

Polyp segmentation from colonoscopy videos is an essential task in medical image processing for detecting early cancer. However, segmenting a precise boundary is still challenging, even with powerful deep neural networks. We consider the difficulty can be caused by: (1) the ambiguity boundary and (2) some complicated shape makes polyps hard to segment. To address these problems, we propose the Boundary-aware Feature and Prediction Refinement framework (BaFPR) for polyp segmentation. Specifically, we design a segmentation decoder for representation learning with boundary prior and propose a novel consistency loss to learn clues from the polar coordinate. The decoder mainly consists of a boundary prior module (BPM) and a bi-directional fusion module (BiFM). BPM is designed to learn the boundary prior, while BiFM learns to fuse representations of BPM and multi-scale representations from an encoder. To handle these complicated shapes of polyps, we maintain an extra segmentation network that learns with polar transformations of data to provide extra clues for the main segmentation network by our proposed consistency loss. We evaluated BaFPR with five challenging datasets for polyp segmentation and the results showed that our proposal consistently improves the segmentation performance of polyps. Code available at: https://github.com/MoriLabNU/BaFPR.

DOI： 10.1080/21681163.2022.2155579

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：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

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

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.39.78

CiNii Research

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：株式会社医学書院  

DOI： 10.11477/mf.1408202455

CiNii Research

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s11548-022-02767-0

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s11548-022-02773-2

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：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

This work proposes a new feature extraction method to analyse patterns of the substantia nigra in Parkinson disease. Recent imaging techniques such that neuromelanin-sensitive MRI enable us to recognise the region of the substantia nigra and capture early Parkinson-disease-related changes. However, automated feature extraction of Parkinson-disease-related changes and their geometrical interpretation are still challenging. To tackle these challenges, we introduce a fifth-order tensor expression of multi-sequence MRI data such as T1-weighted, T2-weighted, and neuromelanin images and its tensor decomposition. Reconstruction from the selected components of the decomposition visualises the discriminative patterns of the substantia nigra between normal and Parkinson-disease patients. We collected multi-sequence MRI data from 155 patients for experiments. Using the collected data, we validate the proposed method and analyse discriminative patterns in the substantia nigra. Experimental results show that the geometrical interpretation of selected features coincides with neuropathological characteristics.

DOI： 10.1007/978-3-031-18814-5_7

Scopus

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：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

Scopus

記述言語：日本語   出版者・発行元：(一社)日本救急医学会  

researchmap

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：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

Depth values are essential information to automate surgical robots and achieve Augmented Reality technology for minimally invasive surgery. Although depth-pose self-supervised monocular depth estimation performs impressively for autonomous driving scenarios, it is more challenging to predict accurate depth values for laparoscopic images due to the following two aspects: (i) the laparoscope’s motions contain many rotations, leading to pose estimation difficulties for the depth-pose learning strategy; (ii) the smooth surface reduces photometric error even if the matching pixels are inaccurate between adjacent frames. This paper proposes a novel self-supervised monocular depth estimation for laparoscopic images with geometric constraints. We predict the scene coordinates as an auxiliary task and construct dual-task consistency between the predicted depth maps and scene coordinates under a unified camera coordinate system to achieve pixel-level geometric constraints. We extend the pose estimation into a Siamese process to provide stronger and more balanced geometric constraints in a depth-pose learning strategy by leveraging the order of the adjacent frames in a video sequence. We also design a weight mask for depth estimation based on our consistency to alleviate the interference from predictions with low confidence. The experimental results showed that the proposed method outperformed the baseline on depth and pose estimation. Our code is available at https://github.com/MoriLabNU/GCDepthL.

DOI： 10.1007/978-3-031-16440-8_45

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: 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

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）   出版者・発行元：(一社)日本医用画像工学会  

黒質緻密部の3次元パターン表現に関する初期的検討について報告する.黒質緻密部における神経細胞の減少はパーキンソン病疾患者において観察される特徴のひとつである.この黒質緻密部の変化はT2強調画像や神経メラニン強調画像を介して捉えることができ,黒質緻密部の左右差や体積の減少として観察できると報告されている.一方で,得られた画像データを用いて黒質緻密部の3次元パターンの変化を計算機上で解析するためには,何らかの特徴表現が必要となる.本稿では黒質緻密部の神経メラニンMRIの信号強度比をテンソルにて表し,これらテンソルの分解に基づいた特徴表現を提案する.提案した特徴表現にて健常者・パーキンソン病患者のパターン分布を調査した結果,2つのクラスで異なる分布を表現できていることを確認した.(著者抄録)

researchmap

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  

DOI： 10.1053/j.gastro.2022.03.053

Scopus

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：IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops  

The stereomicroscope, which is an optical microscope, is used to observe the organoids cultured in cranial windows. A cranial window is a light accessible observation window made on the brain of mice through craniotomy. Organoids research is often conducted on cranial windows. Hence, the observation of blood vessels in them is important for organoid research, like organoid vascularization. Therefore, achieving a simple, low-cost method that extracts blood vessel structures would significantly help researchers observe the blood vessels in cranial windows from microscopic images. However, wide-field optical microscopic images taken by stereomicroscope suffer from low contrast and dura mater occlusion, complicating the observation of the blood vessels in such images. To address such problems and assist researchers who are observing vascular structures, we propose a method that segments blood vessels in cranial windows from low-contrast and wide-field microscopic images. Our method is based on the Attention U-Net framework and clDice, which considers the connectivity of blood vessels. In addition, for low-contrast and partial occlusion problems, we used contrast enhancement and dehazing as preprocessing steps. Our method achieved a Dice score of 75.56%, a clDice score of 79.95%, and the Accuracy of 91.41% on our microscopic image dataset, suggesting that our method can extract blood vessels from low-contrast and wide-field microscopic images better than other methods.

DOI： 10.1109/CVPRW56347.2022.00203

Web of Science

Scopus

DOI： 10.1136/annrheumdis-2022-eular.933

Web of Science

出版者・発行元：International Journal of Hematology  

Acute lymphoblastic leukemia (ALL) is the most common childhood cancer. As overall cure rates of childhood ALL have improved, reduction of overall treatment intensity while still ensuring excellent outcomes is imperative for low-risk patients. We report the outcomes of patients treated following the standard-risk protocol from the prospective Japan Association of Childhood Leukemia Study (JACLS) ALL-02 study, which was conducted between 2002 and 2008 for patients with newly diagnosed ALL aged 1–18 years. Of 1138 patients with B-cell precursor ALL, 388 (34.1%) were allocated to this protocol. Excellent outcomes were achieved despite the overall treatment intensity being lower than that of most contemporary protocols: 4 years event-free survival (EFS) was 92.3% and 4 years overall survival 98.2%. Patients with high hyperdiploidy (HHD) involving triple trisomy (trisomy of chromosomes 4, 10, and 17) or ETV6-RUNX1 had even better outcomes (4 years EFS 97.6% and 100%, respectively). Unique characteristics of this protocol include a selection of low-risk patients with a low initial WBC count and good early treatment response and reduction of cumulative doses of chemotherapeutic agents while maintaining dose density. In Japan, we are currently investigating the feasibility of this protocol while incorporating minimal residual disease into the patient stratification strategy.

DOI： 10.1007/s12185-022-03315-x

Scopus

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1117/1.JMI.9.2.024003

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: 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

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  

This paper presents an automated real-time esophagus achalasia (achalasia) detection method for esophagoscopy assistance. Achalasia is a well-recognized primary esophageal motor disorder of unknown etiology. To diagnose the achalasia, endoscopic evaluation of the esophagus and stomach is recommended to ensure that there is not a malignancy causing the disease or esophageal squamous cell carcinoma complicating achalasia. However, esophagoscopy is low sensitive in the early-stage of achalasia, only about half of patients with early-stage achalasia can be identified. Thus, a quantitative detection system of real-time esophagoscopy video is required for diagnosis assistance of achalasia. This paper presents to use of a convolutional neural network (CNN) to detect all achalasia frames in esophagoscopy videos. The features of achalasia cannot be easily distinguished. To better extract features from esophagoscopy frames, we introduce dense pooling connections and dilated convolutions in the CNN. We trained and evaluated our network with an original dataset that is extracted from several esophagoscopy videos of achalasia patients. Furthermore, we develop a real-time achalasia detection ComputerAided Diagnosis (CAD) system with the trained network. The CAD system can detect each frame from the input esophagoscopy videos with only 0.1 milliseconds delay. The real-time achalasia detection system achieved 0.872 accuracy, and 0.943 AUC score on our dataset.

DOI： 10.1117/12.2613289

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：Computer Methods in Biomechanics and Biomedical Engineering: Imaging and Visualization  

This paper presents a new 3D-spatial feature extraction from a 2D colonoscopic image for polyp-size estimation. The polyp-size estimation poses potential demands on colonoscopy, since an endoscopist’s subjective estimation is apt to result in uncertain polyp-size determination. This uncertain determination is derived from the lack of 3D-spatial information. Even though a previous work clarified that depth estimation mitigates uncertainty in binary polyp-size classification, precise estimation of 3D structure from a colonoscopic image(s) remains an unsolved challenge in medical image analysis. This work proposes a 3D-spatial feature that expresses a polyp’s precise 3D shape to mitigate uncertainty in polyp-size determination. First, we introduce an accurate depth estimation method to capture the 3D structure of a colon. Next, we integrate depth estimation and polyp localisation to extract a 3D polyp shape as a feature. Finally, we achieve polyp-size estimation by statistical learning of extracted features. The experimental results demonstrated the validity both of our depth estimation and 3D-spatial feature. Compared with deep RGB and RGB-D convolutional neural networks (CNNs), a shallow CNN with the proposed 3D-spatial feature achieved a more accurate polyp-size estimation with a mean absolute error of 1.36 mm, whereas the one of the deep CNN is 3.11 mm.

DOI： 10.1080/21681163.2021.2004445

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：Gastrointestinal Endoscopy  

Background and Aims: Recently, the use of computer-aided detection (CADe) for colonoscopy has been investigated to improve the adenoma detection rate (ADR). We aimed to assess the efficacy of a regulatory-approved CADe in a large-scale study with high numbers of patients and endoscopists. Methods: This was a propensity score–matched prospective study that took place at a university hospital between July 2020 and December 2020. We recruited patients aged ≥20 years who were scheduled for colonoscopy. Patients with polyposis, inflammatory bowel disease, or incomplete colonoscopy were excluded. We used a regulatory-approved CADe system and conducted a propensity score matching–based comparison of the ADR between patients examined with and without CADe as the primary outcome. Results: During the study period, 2261 patients underwent colonoscopy with the CADe system or routine colonoscopy, and 172 patients were excluded in accordance with the exclusion criteria. Thirty endoscopists (9 nonexperts and 21 experts) were involved in this study. Propensity score matching was conducted using 5 factors, resulting in 1836 patients included in the analysis (918 patients in each group). The ADR was significantly higher in the CADe group than in the control group (26.4% vs 19.9%, respectively; relative risk, 1.32; 95% confidence interval, 1.12-1.57); however, there was no significant increase in the advanced neoplasia detection rate (3.7% vs 2.9%, respectively). Conclusions: The use of the CADe system for colonoscopy significantly increased the ADR in a large-scale prospective study including 30 endoscopists (Clinical trial registration number: UMIN000040677.)

DOI： 10.1016/j.gie.2021.07.022

Web of Science

Scopus

PubMed

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：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

Automatic segmentation of substantia nigra (SN), which is Parkinson’s disease-related tissue, is an important step toward accurate computer-aided diagnosis systems. Conventional methods for SN segmentation depend heavily on limited magnetic resonance imaging (MRI) modalities such as neuromelanin and quantitative susceptibility mapping, which require longer imaging times and are rare in public datasets. To enable a multi-modal investigation for SN anatomic alterations based on medical bigdata researches, the need for automated SN segmentation arises from commonly investigated T2-weighted MRIs. To improve the performance of the automated SN segmentation from a T2-weighted MRI and enhance the model generalization for cross-center researches, this paper proposes a novel test-time normalization (TTN) method to increase the geometric and intensity similarity between the query data and the model’s trained data. Our proposed method requires no additional training procedure or extra annotation for the unseen data. Our results showed that our proposed TTN achieved a mean Dice score of 71.08% in comparison with the baseline model’s 69.87% score with in-house dataset. Additionally, improved SN segmentation performance was observed from the unseen and unlabeled datasets.

DOI： 10.1007/978-3-031-16449-1_70

Scopus

researchmap

その他リンク： https://dblp.uni-trier.de/db/conf/miccai/miccai2022-7.html#HuIOHLSHKAKAM22

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  

Depth estimation is an essential tool in obtaining depth information for robotic surgery and augmented reality technology in the current laparoscopic surgery robot system. Since there is a lack of ground-truth for depth values and laparoscope motions during operation, depth estimation networks have difficulties in predicting depth maps from laparoscopic images under the supervised strategy. It is challenging to generate the correct depth maps for the different environments from abdominal images. To tackle these problems, we propose a novel monocular self-supervised depth estimation network with sparse nest architecture. We design a non-local block to capture broader and deeper context features that can further enhance the scene-variant generalisation capacity of the network for the differences in datasets. Moreover, we introduce an improved multi-mask feature in the loss function to tackle the classical occlusion problem based on the time-series information from stereo videos. We also use heteroscedastic aleatoric uncertainty to reduce the effect of noisy data for depth estimation. We compared our proposed method with other existing methods for different scenes in datasets. The experimental results show that the proposed model outperformed the state-of-the-art models qualitatively and quantitatively.

DOI： 10.1080/21681163.2021.2015723

Web of Science

Scopus

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：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

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：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

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)  

Federated learning (FL) for medical image segmentation becomes more challenging in multi-task settings where clients might have different categories of labels represented in their data. For example, one client might have patient data with “healthy” pancreases only while datasets from other clients may contain cases with pancreatic tumors. The vanilla federated averaging algorithm makes it possible to obtain more generalizable deep learning-based segmentation models representing the training data from multiple institutions without centralizing datasets. However, it might be sub-optimal for the aforementioned multi-task scenarios. In this paper, we investigate heterogeneous optimization methods that show improvements for the automated segmentation of pancreas and pancreatic tumors in abdominal CT images with FL settings.

DOI： 10.1007/978-3-030-90874-4_10

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：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

This paper proposes an intestine segmentation method from CT volumes for the intestinal obstruction and the ileus diagnosis assistance. The previous method was built based on the 3D U-Net, whose computational cost was high. Nevertheless, there was no confirmation that the 3-dimensional network contributed to the segmentation performance. In this paper, we propose a method utilizing the 2D U-Net on behalf of the previous methods’ 3D U-Net. Experimental results using 110 CT volumes showed that both the proposed (2D U-Net) and previous (3D U-Net) methods achieved similar scores for the segmentation accuracy and system usefulness. In addition, the proposed method’s inference was 0.5 min on average, around 8-times faster than the 3D U-Net’s. Although subsequent processes still require more than 20 min for each case, utilizing lightweight networks is essential for practical use in the future, especially for emergency diagnosis.

DOI： 10.1007/978-3-030-90874-4_1

Scopus

担当区分：筆頭著者,　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 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

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s11548-021-02492-0

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：Proceedings of the IEEE International Conference on Computer Vision  

Segmentation accuracy and generalization ability are essential for deep learning models, especially in medical image segmentation. We present a novel, robust yet straightforward loss function to boost model accuracy and generalizability for medical image segmentation. We reformulate the graph cuts cost function to a loss function for supervised learning. The graph cuts loss innately focuses on a dual penalty to optimize the regional properties and boundary regularization. We benchmark the proposed loss on six public retinal vessel segmentation datasets with a comprehensive intra-dataset and cross-dataset evaluation. Results reveal that the proposed loss is more generalizable, narrowing the performance gap between different architectures. Besides, models trained with our loss show higher segmentation accuracy and better generalization ability than those trained with other mainstream losses. Moreover, we extend our loss to other segmentation tasks, e.g., left atrium and liver tumor segmentation. The proposed loss still achieves comparable performance to the state-of-the-art, demonstrating its potential for any N-D segmentation problem. The code is available at https://github.com/zzh_enggit/graphcutsloss.

DOI： 10.1109/ICCVW54120.2021.00369

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：Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)  

This paper proposes a super-resolution (SR) method, for performing SR on a poorly-aligned dataset. Super-resolution methods commonly needs aligned low-resolution (LR) and high-resolution (HR) images for training. For obtaining paired LR and HR images in medical imaging, we need to align low and high-resolution data using image registration technology. However, since the hardness of aligning LR and HR images, the aligned LR-HR dataset is always low quality. Conventional SR methods always fail to train using poorly-aligned datasets since these methods need high-quality LR-HR datasets. To tackle this problem, we propose a two-step framework for SR using poorly-aligned datasets. In the first step, we decompose image representation into two parts: one is a content code that captures the image content; the other is a style code that captures the image style and anatomy difference between LR / HR images. To perform SR of a given LR image, we input the content code and a latent variable simultaneously into the SR network to obtain an SR result. In the second step, using the trained SR network and an LR image, we search for a content code, and a style code for generating the most proper SR image. This is conducted by searching for the best content code and the best style code by latent space exploration. We conducted experiments using a poorly-aligned clinical-micro CT lung specimen dataset. Experimental results illustrated the proposed method outperformed conventional SR methods by increasing SSIM from 0.309 to 0.312, and have much more convincing perceptual quality than conventional SR methods.

DOI： 10.1007/978-3-030-87592-3_3

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：Medical Physics  

Purpose: For the planning and navigation of neurosurgery, we have developed a fully convolutional network (FCN)-based method for brain structure segmentation on magnetic resonance (MR) images. The capability of an FCN depends on the quality of the training data (i.e., raw data and annotation data) and network architectures. The improvement of annotation quality is a significant concern because it requires much labor for labeling organ regions. To address this problem, we focus on skip connection architectures and reveal which skip connections are effective for training FCNs using sparsely annotated brain images. Methods: We tested 2D FCN architectures with four different types of skip connections. The first was a U-Net architecture with horizontal skip connections that transfer feature maps at the same scale from the encoder to the decoder. The second was a U-Net++ architecture with dense convolution layers and dense horizontal skip connections. The third was a full-resolution residual network (FRRN) architecture with vertical skip connections that pass feature maps between each downsampled scale path and the full-resolution scale path. The last one was a hybrid architecture with a combination of horizontal and vertical skip connections. We validated the effect of skip connections on medical image segmentation from sparse annotation based on these four FCN architectures, which were trained under the same conditions. Results: For multiclass segmentation of the cerebrum, cerebellum, brainstem, and blood vessels from sparsely annotated MR images, we performed a comparative evaluation of segmentation performance among the above four FCN approaches: U-Net, U-Net++, FRRN, and hybrid architectures. The experimental results show that the horizontal skip connections in the U-Net architectures were effective for the segmentation of larger sized objects, whereas the vertical skip connections in the FRRN architecture improved the segmentation of smaller sized objects. The hybrid architecture with both horizontal and vertical skip connections achieved the best results of the four FCN architectures. We then performed an ablation study to explore which skip connections in the FRRN architecture contributed to the improved segmentation of blood vessels. In the ablation study, we compared the segmentation performance between architectures with a horizontal path (HP), an HP and vertical up paths (HP+VUPs), an HP and vertical down paths (HP+VDPs), and an HP and vertical up and down paths (FRRN). We found that the vertical up paths were effective in improving the segmentation of smaller sized objects. Conclusions: This paper investigated which skip connection architectures were effective for multiclass brain segmentation from sparse annotation. Consequently, using vertical skip connections with horizontal skip connections allowed FCNs to improve segmentation performance.

DOI： 10.1002/mp.15192

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: 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

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: 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

Web of Science

Scopus

PubMed

担当区分：筆頭著者   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）   出版者・発行元：公益社団法人 日本生体医工学会  

<p>深層学習は医用画像処理において重要な手法の一つとなった．医用画像処理に含まれる画像分類，検出，セグメンテーションにおいては深層学習が多くの問題を解決し，高い成果を挙げている．画像モダリティに着目すると，眼底画像や放射線画像を対象とした深層学習の適用例が多く，実用化レベルに達したものもある．本発表では，手法や画像モダリティの面から深層学習を用いた医用画像処理を概観し，応用例を紹介する．また，深層学習の新たな応用についても紹介する．深層学習は画像に限らず様々なデータを説明変数と目的変数とすることが可能である．この性質を利用し，画像，テキスト，センサ計測データ等を用いる新たな手法を構築できる．応用例と共に深層学習の新たな可能性について考える．</p>

DOI： 10.11239/jsmbe.Annual59.256

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1055/a-1475-3624

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

記述言語：英語   掲載種別：研究論文（学術雑誌）  

Web of Science

担当区分：筆頭著者,　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（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  

Subarachnoid Hemorrhage (SAH) detection is a critical, severe problem that confused clinical residents for a long time. With the rise of deep learning technologies, SAH detection made a significant breakthrough in recent ten years. Whereas, the performances are significantly degraded on imbalanced data, makes deep learning models have always suffered criticism. In this study, we present a DenseNet-LSTM network with Class-Balanced Loss and the transfer learning strategy to solve the SAH detection problem on an extremely imbalanced dataset. Compared to the previous works, the proposed framework not merely effectively integrate greyscale features the and spatial information from the consecutive CT scans, but also employ Class-Balanced loss and transfer learning to alleviate the adverse effects and broaden feature diversity respectively on an extreme SAH cases scarcity dataset, mimicking the actual situation of emergency departments. Comprehensive experiments are conducted on a dataset, consisted of 2,519 cases without hemorrhage cases and only 33 cases with SAH. Experimental results demonstrate the F-measure score of SAH detection achieved a remarkable improvement, the backbone DenseNet121 gained around 33% promotion after transfer learning, and on this basis, importing the Class-Balanced Loss and the LSTM structure, the F-measure score further increased 6.1% and 2.7% sequentially.

DOI： 10.1117/12.2582088

Web of Science

Scopus

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1097/BRS.0000000000003749

Scopus

PubMed

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1111/den.13847

Web of Science

Scopus

PubMed

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

記述言語：英語   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

記述言語：日本語   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。従来の白色光(WLI)に加え、NBIは腫瘍検出の精度が改善する報告があるが、検者の経験に依存するため、再現性・客観性が乏しいことが課題である。近年、客観的な診断ツールとして人工知能(AI)が活用されているが、膀胱鏡においてAIによる腫瘍検出の研究はあまり行われていない。今回、我々は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12月から2020年11月まで、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。腫瘍画像を学習データ、テストデータに分類し、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、医師個人のコンピューター上で環境構築を行った。【結果】腫瘍画像(WLI:809枚、NBI:684枚)をそれぞれtiny-YOLOで学習を行い、腫瘍画像(WLI:280枚、NBI:87枚)と正常画像(WLI:104枚、NBI:104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84.6%/76.9%/90.8%、NBIで78.1%/97.1%/95.8%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.Annual59.299

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>深層学習において本格的な実験や研究を行うためには，GPUを使った環境構築が必要となる．これは，CPUのみで計算した場合と比較して数十倍の計算速度が得られるためである．今回の講座では，GPU環境の構築方法を説明し，画像の領域分割を例にして，計算時間がどの程度変化するかを明らかにする．同時に，U-Netを用いた領域分割の方法，その正解データの構築についても言及する．</p>

DOI： 10.11409/mit.39.124

担当区分：最終著者   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  

DOI： 10.11409/mit.39.97

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）   出版者・発行元：一般社団法人 日本コンピュータ外科学会  

DOI： 10.5759/jscas.23.74

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s11548-020-02255-3

Web of Science

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1080/21681163.2020.1835560

Web of Science

Scopus

記述言語：英語   出版者・発行元：(一社)日本コンピュータ外科学会  

researchmap

記述言語：日本語   出版者・発行元：(一社)日本コンピュータ外科学会  

researchmap

記述言語：日本語   出版者・発行元：日本医用画像工学会  

researchmap

記述言語：日本語   出版者・発行元：(一社)日本コンピュータ外科学会  

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1053/j.gastro.2020.09.027

Scopus

PubMed

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1002/mp.14372

Web of Science

Scopus

PubMed

記述言語：英語   出版者・発行元：日本医用画像工学会  

researchmap

記述言語：日本語   出版者・発行元：日本医用画像工学会  

本稿では,Spherical K-means(SpK)により抽出された複数スケールの特徴量と位置特徴量を用いた胃壁μCT画像からの粘膜層,粘膜下層,筋層及び腫瘍の抽出手法について報告する.胃壁のμCT画像から解剖学的構造を抽出することで腫瘍の立体構造把握が可能である.従来手法では複数スケールの特徴抽出フィルタをSpKによって学習し,パッチから特徴抽出を行った.その後得られた特徴量に基づいてSVMによって分類を行っていた.しかし,SpKにより生成された特徴抽出フィルタは濃度値に基づく特徴量を抽出するため,粘膜層と筋層のように濃度値が類似した領域同士の特徴量が近しい値になる.本手法ではSpKにより得られる複数スケールの特徴量にパッチの位置特徴量を追加して分類する事によって,濃度値に基づく特徴量では分類できない領域の分類を可能にする.パッチの位置特徴量としてパッチの中心座標を用いる.本手法を胃壁μCT像に適用した結果,粘膜層,粘膜下層,筋層及び腫瘍の抽出に対するDICE係数は68.3%,63.2%,82.2%,69.1%であった.特に従来手法と比べ,腫瘍の抽出に対するDICE係数は26.1%向上した.(著者抄録)

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s11548-020-02241-9

Web of Science

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/j.gie.2020.07.060

Scopus

PubMed

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1080/13645706.2019.1625926

Web of Science

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/j.cgh.2019.09.009

Web of Science

Scopus

PubMed

researchmap

掲載種別：研究論文（学術雑誌）  

Web of Science

記述言語：日本語   出版者・発行元：(社)日本大腸検査学会  

EndoBRAIN(製造:サイバーネットシステム株式会社・販売:オリンパス株式会社)は、人工知能による診断支援システムとして、本邦初の薬機法承認を得た医療機器である。EndoBRAINは超拡大内視鏡(CF-H290ECI、オリンパス株式会社)によって取得される520倍の拡大画像をAIが解析することで、大腸病変が腫瘍なのかどうかを瞬時に予測し、医師に情報提供を行う。2012年から開始したEndoBRAINの研究開発は、昭和大学-名古屋大学-サイバーネットシステム株式会社での医工産連携のもと、日本医療研究開発機構(AMED)からのサポート下に継続的に実施され、2018年12月に薬機法承認を取得した。EndoBRAINの臨床的有用性については、2018年に791症例を対象とした大規模前向き試験の結果が公表されており、感度92.7%で腫瘍の鑑別が可能であると報告されている。また、最近公表された前向き試験の副次解析結果によると、EndoBRAINを使用することで、本来切除すべきでない非腫瘍性ポリープの治療件数を抑制することができるため、これにより最大164億円/年の医療費が削減しうることが分かってきた。本稿では、EndoBRAINの特徴・使用法、および薬機法承認取得までの経緯を紹介するとともに、その臨床的有用性および期待される医療費抑制効果について、最新の研究成果を総括する。(著者抄録)

researchmap

記述言語：日本語   出版者・発行元：(社)日本大腸検査学会  

researchmap

記述言語：英語   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  

This paper presents a super-resolution (SR) method with unpaired training
dataset of clinical CT and micro CT volumes. For obtaining very detailed
information such as cancer invasion from pre-operative clinical CT volumes of
lung cancer patients, SR of clinical CT volumes to $\m$}CT level is desired.
While most SR methods require paired low- and high- resolution images for
training, it is infeasible to obtain paired clinical CT and {\mu}CT volumes. We
propose a SR approach based on CycleGAN, which could perform SR on clinical CT
into $\mu$CT level. We proposed new loss functions to keep cycle consistency,
while training without paired volumes. Experimental results demonstrated that
our proposed method successfully performed SR of clinical CT volume of lung
cancer patients into $\mu$CT level.

arXiv

researchmap

その他リンク： http://arxiv.org/pdf/2004.03272v1

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.23922/jarc.2019-045

Web of Science

PubMed

記述言語：日本語   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

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1117/1.JMI.7.2.026001

Web of Science

Scopus

PubMed

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

Web of Science

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.18999/nagjms.82.1.25

Web of Science

Scopus

PubMed

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： 10.1117/12.2551022

Web of Science

Scopus

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： 10.1117/12.2549532

Web of Science

Scopus

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1117/12.2548910

Web of Science

Scopus

担当区分：筆頭著者   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）  

DOI： 10.5759/jscas.22.54

記述言語：英語   掲載種別：研究論文（学術雑誌）  

<p>EndoBRAIN（製造：サイバネットシステム社，販売：オリンパス社）は，内視鏡検査中に発見された大腸ポリープが腫瘍なのかどうかを瞬時に解析予測し，医師の診断支援を行うシステムである．EndoBRAINは人工知能に基づく診断支援システムとして，本邦初の薬事承認を2018年12月に得た医療機器であるが，承認達成は緊密な医工産官連携研究の賜物であった．昭和大学（医）-名古屋大学（工）-サイバネットシステム株式会社（産）での連携体制を構築し，複数の公的研究費（文部科学省科学研究費（新学術領域研究・多元計算解剖学を含む）・日本医療研究開発機構（AMED）研究費）のサポートを得ることで，着実に研究成果を蓄積し，薬事承認に至った．本稿では，EndoBRAINの研究開発概要を示すとともに，医療機器開発における鬼門である薬事承認と，その後の普及において不可欠とされる（しかしながらいまだAI医療機器については実現していない）保険承認に対するアプローチについて紹介したい．</p>

DOI： 10.11409/mit.38.213

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： 10.1117/12.2551024

Web of Science

Scopus

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： 10.1117/12.2550496

Scopus

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： 10.1117/12.2548929

Scopus

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： 10.1117/12.2549949

Scopus

担当区分：筆頭著者   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： 10.1117/12.2549951

Web of Science

Scopus

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）  

DOI： 10.1007/s11548-019-02062-5

Web of Science

Scopus

researchmap

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）  

DOI： 10.1049/htl.2019.0071

Web of Science

Scopus

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1049/htl.2019.0079

Web of Science

Scopus

PubMed

記述言語：日本語  

DOI： 10.21037/tgh.2019.12.14

Web of Science

記述言語：日本語   出版者・発行元：(一社)日本コンピュータ外科学会  

researchmap

記述言語：日本語   出版者・発行元：(一社)日本コンピュータ外科学会  

researchmap

記述言語：日本語   出版者・発行元：(一社)日本コンピュータ外科学会  

researchmap

記述言語：英語   出版者・発行元：(一社)日本コンピュータ外科学会  

researchmap

記述言語：日本語   出版者・発行元：(一社)日本コンピュータ外科学会  

researchmap

記述言語：日本語   出版者・発行元：(一社)日本コンピュータ外科学会  

researchmap

出版者・発行元：(株)メジカルビュー社  

DOI： 10.18885/j01843.2020039296

CiNii Research

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

Web of Science

Scopus

PubMed

記述言語：日本語   出版者・発行元：(一社)日本消化器内視鏡学会  

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1183/13993003.congress-2019.PA3168

Web of Science

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1111/den.13340

Web of Science

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1111/den.13317

Web of Science

Scopus

PubMed

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）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

Web of Science

記述言語：英語   掲載種別：研究論文（学術雑誌）  

Web of Science

記述言語：英語   掲載種別：研究論文（学術雑誌）  

Web of Science

記述言語：英語   掲載種別：研究論文（学術雑誌）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

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）  

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.3390/mi10020081

Web of Science

Scopus

PubMed

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/j.gie.2018.09.024

Web of Science

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

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）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

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）  

researchmap

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）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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： 10.1117/12.2521618

Web of Science

担当区分：筆頭著者   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）  

DOI： 10.11409/mit.37.123

researchmap

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： 10.1007/978-3-030-32226-7_39

Web of Science

Scopus

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： 10.1117/12.2513094

Web of Science

Scopus

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： 10.1007/978-3-030-32226-7_27

Web of Science

Scopus

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： 10.1117/12.2512766

Web of Science

Scopus

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： 10.1117/12.2512746

Web of Science

Scopus

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： 10.1007/978-3-030-32689-0_11

Web of Science

Scopus

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： 10.1117/12.2521646

Web of Science

Scopus

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： 10.1117/12.2512790

Web of Science

Scopus

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： 10.1007/978-3-030-11018-5_43

Web of Science

Scopus

記述言語：英語   掲載種別：研究論文（研究会，シンポジウム資料等）  

DOI： 10.5759/jscas.21.143

担当区分：筆頭著者   記述言語：英語   掲載種別：研究論文（研究会，シンポジウム資料等）  

本稿では，われわれがAMED プロジェクトにおいて行ってきた医用画像解析研究について紹介する．具体的には，非造影腹部CT 像からの血管診断支援およびCT 像からの腹部複数臓器自動セグメンテーションを含む放射線画像診断支援研究，胃および大腸の内視鏡画像の観察部位分類を含む内視鏡画像解析研究を行った．これらの研究を概説し，得られた成果等について述べる．

DOI： 10.11409/mit.37.84

記述言語：英語   掲載種別：研究論文（学術雑誌）  

Web of Science

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： 10.1117/12.2513093

Web of Science

Scopus

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： 10.1117/12.2512145

Web of Science

Scopus

CiNii Research

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

DOI： 10.18974/tvrsj.23.4_249

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）  

researchmap

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）  

researchmap

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）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.7326/M18-0249

Web of Science

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1183/13993003.congress-2018.PA860

Web of Science

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1111/ans.14331

Web of Science

Scopus

PubMed

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1053/j.gastro.2018.04.003

Web of Science

Scopus

PubMed

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

Web of Science

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/j.compmedimag.2018.03.001

Web of Science

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）  

researchmap

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）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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： 10.1117/12.2293288

Web of Science

Scopus

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）  

本稿では，デスクトップ型マイクロCTを用いた微細構造イメージングについて述べる．臨床の場で利用されるX線CT装置は，その解像度がおおよそ1ボクセルあたり0.5mmから1mm程度である．このようなイメージング装置を用いて得ることができる画像は，このボクセル解像度に準じた解剖構造を得ることができる．一方，マイクロCT装置を用いれば，1&mu;mから50&mu;m程度の解像度で撮影できる．本稿では，マイクロCTによって撮影された肺標本ならびに心臓標本を示し，その可能性について述べる．

DOI： 10.11409/mit.36.127

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： 10.1007/978-3-030-00934-2_68

Web of Science

Scopus

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1117/12.2293499

Web of Science

Scopus

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： 10.1117/12.2292172

Web of Science

Scopus

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： 10.1117/12.2293414

Web of Science

Scopus

担当区分：筆頭著者   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）  

ディープラーニングは医用画像処理のさまざまなタスク解決に使用することができる便利な手法である．ディープラーニングを研究で活用し，その性能を引き出すために注意を払うべき点がいくつかある．その中で，ディープラーニングにおける多数の手動指定パラメーターの選択，データオーギュメンテーションを使用する上で注意する点，ディープラーニングのランダム要素が与える影響について解説する．

DOI： 10.11409/mit.36.72

担当区分：筆頭著者   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）  

DOI： 10.11409/mit.36.45

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： 10.1007/978-3-030-01045-4_14

Web of Science

Scopus

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： 10.1117/12.2293486

Web of Science

Scopus

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： 10.1117/12.2287066

Web of Science

Scopus

記述言語：英語   掲載種別：研究論文（研究会，シンポジウム資料等）  

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

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： 10.1117/12.2293495

Web of Science

Scopus

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： 10.1007/978-3-030-00934-2_26

Web of Science

Scopus

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： 10.1117/12.2293431

Web of Science

Scopus

記述言語：英語   掲載種別：研究論文（学術雑誌）  

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

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： 10.1007/978-3-030-00937-3_48

Web of Science

Scopus

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  

DOI： 10.1117/12.2293936

Web of Science

Scopus

記述言語：日本語   出版者・発行元：一般社団法人 日本機械学会  

DOI： 10.1299/jsmemmt.2018.12.c20

CiNii Research

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1117/1.JMI.4.4.044502

Web of Science

Scopus

PubMed

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1055/s-0043-105486

Web of Science

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1016/j.media.2017.03.006

Web of Science

Scopus

PubMed

記述言語：英語   出版者・発行元：日本医用画像工学会  

本論文では,3次元医用画像向けの新しい教師なしセグメンテーションの手法を提案する.提案手法は2つの段階に分けられる.1つ目の段階では,JULEを用いた深層表現学習をおこなう.JULEは,CNNから出力される表現のクラスタリングと,クラスタラベルを教師信号としたCNNの更新を繰り返す手法である.2つ目の段階では,訓練済みのCNNから生成された深層表現に対し,K-means法を適用してセグメンテーションをおこなう.評価には肺がん標本のマイクロCT画像を用い,浸潤領域,非浸潤領域,正常領域という3つの領域に分けることを試みた.マイクロCT画像上で病理組織学的特徴に基づいたセグメンテーションをおこなうことは,将来の病理診断精度の向上につながる.セグメンテーション結果の定性評価から,深層表現が3次元医用画像のセグメンテーションに有用であることが示された.(著者抄録)

researchmap

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）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

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）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

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）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s11548-017-1549-x

Web of Science

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s11548-017-1542-4

Web of Science

Scopus

記述言語：英語   掲載種別：研究論文（学術雑誌）  

Web of Science

記述言語：英語   掲載種別：研究論文（学術雑誌）  

Web of Science

記述言語：英語   掲載種別：研究論文（学術雑誌）  

Web of Science

記述言語：英語   掲載種別：研究論文（学術雑誌）  

Web of Science

記述言語：英語   掲載種別：研究論文（学術雑誌）  

researchmap

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/s11548-016-1492-2

Web of Science

Scopus

PubMed

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

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）  

researchmap

記述言語：日本語   掲載種別：研究論文（学術雑誌）  

researchmap

記述言語：英語   掲載種別：研究論文（その他学術会議資料等）  

researchmap

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）  

DOI： 10.1007/s11548-016-1456-6

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： 10.1117/12.2254949

Web of Science

Scopus

記述言語：日本語   掲載種別：研究論文（学術雑誌）  

<p>[目的] 3Dプリンタは手術支援の一つとしての期待が高まっているが、現在の汎用機で、どの位正確に病変を表すことが可能であるかに関して不明である。そこで本研究では、異なった3Dプリンタ造形法と素材を用いて心臓モデルを作成し有用性を検討した。［方法］CT画像から抽出した心臓データをstlファイルに変換することにより、３Ｄプリンタによる心臓模型をつくった。その際、二種類の3Dプリンタにおける造形法を試した。一つ目は材料にPLAを使用した熱溶解積層法、二つ目は材料にAR-G1Lを使用したインクジェット法である。[結果] 熱溶解積層法を用いて作成したものは、原寸大では印刷できず75％に縮小し印刷した。その結果、血管部分の樹脂が崩れてしまった、また、サイズが小さくなってしまったことに加え、モデル全体が単色であることから、血管がどこにあるかということが一目で分かりにくく、さらに材質が固く力を加えた際に壊れやすいという問題があった。インクジェット方式を用いて作成したものは熱溶解積層法とは異なり原寸大での作成が可能であった。病変部位を正確に表現できた。さらにゴム素材を使用可能でき塑性に富んでいた。[結論]インクジェット方式を採用し作成した心臓模型は塑性に優れ有用である。</p>

DOI： 10.11239/jsmbe.55Annual.273

researchmap

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）  

<p>本講演では、機械学習を利用した内視鏡画像の自動診断手法について紹介する。特に、超拡大内視鏡画像を用いた大腸ポリープの類型判別の自動診断手法を例に挙げながら、内視鏡画像診断における機械学習の有用性について議論したい。パーセプトロン、統計的パターン認識に始まる機械学習は長年研究が行われてきたが、GPGPUなどに代表される手軽なハイパフォーマンスコンピューティング技術の発展により、非常に複雑なアーキテクチャを持つニューラルネットワークを用いたパターン認識などが可能となった。一方、内視鏡画像の診断には高度な技術が必要とされ、医師間の差による診断のブレも発生する。そこで、機械学習を用いた内視鏡画像の自動診断方法について２，３紹介する。一つ目の方法は、Hand-crafted特徴量（微分特徴量）などを求め、その特徴量により大腸ポリープの組織型をSVM (Support Vector Machine)による自動分類するものである。二つ目の方法は、CNN (Convolutional Neural Network)を用いて内視鏡画像の自動分類を行うものである。これらの手法について、技術的な側面に焦点をあて解説を行う。機械学習において重要な学習データ生成法についても合わせて議論したい。</p>

DOI： 10.11239/jsmbe.55Annual.344

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： 10.1007/978-3-319-67543-5_6

Web of Science

Scopus

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： 10.1117/12.2254782

Web of Science

Scopus

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： 10.1117/12.2254233

Web of Science

Scopus

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： 10.1007/978-3-319-67434-6_11

Web of Science

Scopus

担当区分：筆頭著者,　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： 10.1007/978-3-319-67558-9_26

Web of Science

Scopus

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： 10.1117/12.2254095

Web of Science

Scopus

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： 10.1007/978-3-319-66185-8_23

Scopus

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： 10.1007/978-3-319-66182-7_78

Scopus

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