Faculty Profiles - ODA, Masahiro

Papers - ODA, Masahiro

Division display 41 - 60 of about 372 ／ All the affair displays >>

Skeleton-guided 3D convolutional neural network for tubular structure segmentation

Zhu, RY; Oda, M; Hayashi, Y; Kitasaka, T; Misawa, K; Fujiwara, M; Mori, K

INTERNATIONAL JOURNAL OF COMPUTER ASSISTED RADIOLOGY AND SURGERY 2024.9

Language：English Publisher：International Journal of Computer Assisted Radiology and Surgery

Purpose: Accurate segmentation of tubular structures is crucial for clinical diagnosis and treatment but is challenging due to their complex branching structures and volume imbalance. The purpose of this study is to propose a 3D deep learning network that incorporates skeleton information to enhance segmentation accuracy in these tubular structures. Methods: Our approach employs a 3D convolutional network to extract 3D tubular structures from medical images such as CT volumetric images. We introduce a skeleton-guided module that operates on extracted features to capture and preserve the skeleton information in the segmentation results. Additionally, to effectively train our deep model in leveraging skeleton information, we propose a sigmoid-adaptive Tversky loss function which is specifically designed for skeleton segmentation. Results: We conducted experiments on two distinct 3D medical image datasets. The first dataset consisted of 90 cases of chest CT volumetric images, while the second dataset comprised 35 cases of abdominal CT volumetric images. Comparative analysis with previous segmentation approaches demonstrated the superior performance of our method. For the airway segmentation task, our method achieved an average tree length rate of 93.0%, a branch detection rate of 91.5%, and a precision rate of 90.0%. In the case of abdominal artery segmentation, our method attained an average precision rate of 97.7%, a recall rate of 91.7%, and an F-measure of 94.6%. Conclusion: We present a skeleton-guided 3D convolutional network to segment tubular structures from 3D medical images. Our skeleton-guided 3D convolutional network could effectively segment small tubular structures, outperforming previous methods.

DOI： 10.1007/s11548-024-03215-x

整形外科領域の画像診断技術の深化深層学習を用いたMRI運動器腫瘍診断

伊藤定之, 中島宏彰, 世木直喜, 大内田隼, 小田昌宏, 山内一平, 森田圭則, 大出幸史, 長谷康弘, 森健策, 今釜史郎

日本整形外科学会雑誌 Vol. 98 ( 8 ) page： S1732 - S1732 2024.9

Analysis of the performance of the CorneAI for iOS in the classification of corneal diseases and cataracts based on journal photographs. Open Access

Taki Y, Ueno Y, Oda M, Kitaguchi Y, Ibrahim OMA, Aketa N, Yamaguchi T

Scientific reports Vol. 14 ( 1 ) page： 15517 2024.7

Language：English Publisher：Scientific Reports

CorneAI for iOS is an artificial intelligence (AI) application to classify the condition of the cornea and cataract into nine categories: normal, infectious keratitis, non-infection keratitis, scar, tumor, deposit, acute primary angle closure, lens opacity, and bullous keratopathy. We evaluated its performance to classify multiple conditions of the cornea and cataract of various races in images published in the Cornea journal. The positive predictive value (PPV) of the top classification with the highest predictive score was 0.75, and the PPV for the top three classifications exceeded 0.80. For individual diseases, the highest PPVs were 0.91, 0.73, 0.42, 0.72, 0.77, and 0.55 for infectious keratitis, normal, non-infection keratitis, scar, tumor, and deposit, respectively. CorneAI for iOS achieved an area under the receiver operating characteristic curve of 0.78 (95% confidence interval [CI] 0.5–1.0) for normal, 0.76 (95% CI 0.67–0.85) for infectious keratitis, 0.81 (95% CI 0.64–0.97) for non-infection keratitis, 0.55 (95% CI 0.41–0.69) for scar, 0.62 (95% CI 0.27–0.97) for tumor, and 0.71 (95% CI 0.53–0.89) for deposit. CorneAI performed well in classifying various conditions of the cornea and cataract when used to diagnose journal images, including those with variable imaging conditions, ethnicities, and rare cases.

DOI： 10.1038/s41598-024-66296-3

Federated 3D multi-organ segmentation with partially labeled and unlabeled data

Zheng, Z; Hayashi, Y; Oda, M; Kitasaka, T; Misawa, K; Mori, K

INTERNATIONAL JOURNAL OF COMPUTER ASSISTED RADIOLOGY AND SURGERY 2024.5

Language：English Publisher：International Journal of Computer Assisted Radiology and Surgery

Purpose: This paper considers a new problem setting for multi-organ segmentation based on the following observations. In reality, (1) collecting a large-scale dataset from various institutes is usually impeded due to privacy issues; (2) many images are not labeled since the slice-by-slice annotation is costly; and (3) datasets may exhibit inconsistent, partial annotations across different institutes. Learning a federated model from these distributed, partially labeled, and unlabeled samples is an unexplored problem. Methods: To simulate this multi-organ segmentation problem, several distributed clients and a central server are maintained. The central server coordinates with clients to learn a global model using distributed private datasets, which comprise a small part of partially labeled images and a large part of unlabeled images. To address this problem, a practical framework that unifies partially supervised learning (PSL), semi-supervised learning (SSL), and federated learning (FL) paradigms with PSL, SSL, and FL modules is proposed. The PSL module manages to learn from partially labeled samples. The SSL module extracts valuable information from unlabeled data. Besides, the FL module aggregates local information from distributed clients to generate a global statistical model. With the collaboration of three modules, the presented scheme could take advantage of these distributed imperfect datasets to train a generalizable model. Results: The proposed method was extensively evaluated with multiple abdominal CT datasets, achieving an average result of 84.83% in Dice and 41.62 mm in 95HD for multi-organ (liver, spleen, and stomach) segmentation. Moreover, its efficacy in transfer learning further demonstrated its good generalization ability for downstream segmentation tasks. Conclusion: This study considers a novel problem of multi-organ segmentation, which aims to develop a generalizable model using distributed, partially labeled, and unlabeled CT images. A practical framework is presented, which, through extensive validation, has proved to be an effective solution, demonstrating strong potential in addressing this challenging problem.

DOI： 10.1007/s11548-024-03139-6

A clinical assessment of three-dimensional-printed liver model navigation for thrice or more repeated hepatectomy based on a conversation analysis

Igami, T; Maehigashi, A; Nakamura, Y; Hayashi, Y; Oda, M; Yokoyama, Y; Mizuno, T; Yamaguchi, J; Onoe, S; Sunagawa, M; Watanabe, N; Baba, T; Kawakatsu, S; Mori, K; Miwa, K; Ebata, T

SURGERY TODAY 2024.4

Language：English Publisher：Surgery Today

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

DOI： 10.1007/s00595-024-02835-9

Artificial intelligence-based diagnostic imaging system with virtual enteroscopy and virtual unfolded views to evaluate small bowel lesions in Crohn's disease.

Furukawa K, Oda M, Watanabe O, Nakamura M, Yamamura T, Maeda K, Mori K, Kawashima H

Revista espanola de enfermedades digestivas 2024.3

Anatomical attention can help to segment the dilated pancreatic duct in abdominal CT.

Shen C, Roth HR, Hayashi Y, Oda M, Sato G, Miyamoto T, Rueckert D, Mori K

International journal of computer assisted radiology and surgery 2024.3

Use of 3D-printed model of liver by experts and novices Reviewed

Akihiro Maehigashi, Kazuhisa Miwa, Masahiro Oda, Yoshihiko Nakamura, Kensaku Mori, Tsuyoshi Igami

Current Psychology 2024.2

Publishing type：Research paper (scientific journal)

This study investigated the influence of using three-dimensional (3D) computer and 3D-printed models on the spatial reasoning of experts and novices. The task of this study required general university students as novices in Experiment 1 and surgeons specializing in digestive surgery as experts in Experiment 2 to infer the cross sections of a liver, using a 3D-computer or 3D-printed model. The results of the experiments showed that the university students learned faster and inferred the liver structure more accurately with the 3D-printed model than with the 3D-computer model. Conversely, the surgeons showed the same task performance when using the 3D-computer and 3D-printed models; however, they performed the task with more confidence and less workload during the task with the 3D-printed model. Based on the results, the cognitive effects and advantages of using 3D-printed models for novices and experts have been discussed.

DOI： 10.1007/s12144-024-05676-4

YOLOv7-RepFPN: Improving real-time performance of laparoscopic tool detection on embedded systems Reviewed

Yuzhang Liu, Yuichiro Hayashi, Masahiro Oda, Takayuki Kitasaka, Kensaku Mori

Healthcare Technology Letters 2024.1

Publishing type：Research paper (scientific journal)

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

DOI： 10.1049/htl2.12072

Deep learning model for extensive smartphone-based diagnosis and triage of cataracts and multiple corneal diseases.

Ueno Y, Oda M, Yamaguchi T, Fukuoka H, Nejima R, Kitaguchi Y, Miyake M, Akiyama M, Miyata K, Kashiwagi K, Maeda N, Shimazaki J, Noma H, Mori K, Oshika T

The British journal of ophthalmology 2024.1

EndoSelf: Self-supervised Monocular 3D Scene Reconstruction of Deformable Tissues with Neural Radiance Fields on Endoscopic Videos

Li, WD; Hayashi, Y; Oda, M; Kitasaka, T; Misawa, K; Mori, K

MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION - MICCAI 2024, PT VI Vol. 15006 page： 241 - 251 2024

Towards better laparoscopic video segmentation: A class-wise contrastive learning approach with multi-scale feature extraction Open Access

Zhang L., Hayashi Y., Oda M., Mori K.

Healthcare Technology Letters 2024

Publisher：Healthcare Technology Letters

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

DOI： 10.1049/htl2.12069

M U-Net: Intestine Segmentation Using Multi-dimensional Features for Ileus Diagnosis Assistance

An Q., Oda H., Hayashi Y., Kitasaka T., Hinoki A., Uchida H., Suzuki K., Takimoto A., Oda M., Mori K.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) Vol. 14313 LNCS page： 135 - 144 2024

Publisher：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

Revisiting instrument segmentation: Learning from decentralized surgical sequences with various imperfect annotations Open Access

Zheng Z., Hayashi Y., Oda M., Kitasaka T., Mori K.

Healthcare Technology Letters 2024

Publisher：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

Intestine Segmentation from CT Volume based on Bidirectional Teaching

An, Q; Oda, H; Hayashi, Y; Kitasaka, T; Hinoki, A; Uchida, H; Suzuki, K; Takimoto, A; Oda, M; Mori, K

MEDICAL IMAGING 2024: IMAGE PROCESSING Vol. 12926 2024

Publisher：Progress in Biomedical Optics and Imaging - Proceedings of SPIE

This paper proposes an intestine segmentation method to segment intestines from CT volumes for helping clinicians diagnose intestine obstruction. For large-scale labeled datasets, fully-supervised methods have shown superior results. However, medical image segmentation is usually difficult to achieve accurate prediction due to the limited number of labeled data available for training. To address this challenge, we introduce a novel multi-view symmetrical network (MVS-Net) for intestine segmentation and incorporate bidirectional teaching to utilize unlabeled datasets. Specifically, we design the MVS-Net, which can use different sizes of convolution kernels instead of a fixed kernel size, enabling the network to capture multi-scale features from images’ different perceptual fields and ensure segmentation accuracy. Additionally, the pseudo-labels are generated by bidirectional teaching, which can make the network captures semantic information from large-scale unlabeled data for increasing the training data. We repeated the experiment five times, and used the averaged result on the intestines dataset to represent the segmentation accuracy of the proposed method. The experimental results showed the average Dice was 78.86%, the average recall 84.50%, and the average precision 75.94%, respectively.

DOI： 10.1117/12.3006623

Continual pretraining for enhanced multi-organ segmentation from CT images

Yang, YQ; Shen, C; Tang, YC; Roth, HR; Oda, M; Hayashi, Y; Misawa, K; Mori, K

MEDICAL IMAGING 2024: IMAGE PROCESSING Vol. 12926 2024

Publisher：Progress in Biomedical Optics and Imaging - Proceedings of SPIE

Self-supervised pretraining has shown great performance in improving the accuracy of downstream tasks. Although pretraining on a large dataset improves performances, it becomes challenging to further optimize the model by solely enlarging the dataset. In contrast, additional adaptation of pretrained models to the target domain has shown promise in NLP. Inspired by the success of continual pretraining, we investigated the efficacy of adapting the target domain dataset to a pretrained model in medical imaging, particularly in the context of segmentation. We present a study based on a self-supervised pretraining framework using the SwinUNETR backbone. In this study, we improved the generalizability of the self-supervised pretraining by adapting a foundational model pretrained on 5k CT volumes to data of the downstream segmentation task. In detail, we employed 385 abdominal CT volumes for the continual task-adaptive pretraining and 24 abdominal CT volumes for the downstream segmentation task, all sourced from the same dataset. Additionally, we conducted comparative experiments to demonstrate the benefits of this task-adapting pretraining approach. Our method has shown that continual pretraining helps to improve the performances, achieving an average Dice score for 10-class organ segmentation of 87.8%.

DOI： 10.1117/12.3006630

Boosting substantia nigra segmentation from T2 weighted MRI via test-time normalization and distance-reweighted loss

Hu, T; Itoh, H; Oda, M; Saiki, S; Hattori, N; Kamagata, K; Sako, W; Ishikawa, K; Aoki, S; Mori, K

COMPUTER-AIDED DIAGNOSIS, MEDICAL IMAGING 2024 Vol. 12927 2024

Publisher：Progress in Biomedical Optics and Imaging - Proceedings of SPIE

Substantia nigra (SN) has been reported as significantly related to the progression of Parkinson’s Disease (PD). Fully automated segmentation of SN is an important step for developing an interpretable computer-aided diagnosis system for PD. Based on the deep learning techniques, this paper proposes a novel distance-reweighted loss function and combines it with the test-time normalization (TTN) to boost the fully automated SN segmentation accuracy from low contrast T2 weighted MRI. The proposed loss encourages the model to focus on the suspicious regions with vague boundaries, and the involved TTN narrows the gap between an input MRI volume and the reference MRI volumes in test-time. The results showed that both the proposed loss and TTN could help improve the segmentation accuracy. By combining the proposed loss and TTN, the averaged Dice coefficient achieved 70.90% from T2 weighted MRI, compared to 68.17% by the baseline method.

DOI： 10.1117/12.3008521

Adaptive Octree Cube Refinement Depending on Grasping Position for Deformable Organ Models

Miyazaki, R; Hayashi, Y; Oda, M; Mori, K

IMAGE-GUIDED PROCEDURES, ROBOTIC INTERVENTIONS, AND MODELING, MEDICAL IMAGING 2024 Vol. 12928 2024

Publisher：Progress in Biomedical Optics and Imaging - Proceedings of SPIE

This paper describes an adaptive octree cube refinement method for deformable organ models. Surgical simulation is one of the most promising ways for surgical training. Various types of surgical simulators have been researched and developed. Laparoscopic surgery simulators are already in practical use. They have been evaluated for their effectiveness in learning surgical techniques. To realize a high-quality simulator, it is important to efficiently process organ deformation models according to the content of the surgical simulation so that both high-resolution and real-Time processing. In this study, we extend adaptive mesh refinement, which increases mesh resolution in the manipulation region, and apply it to an octree cube structure. The refinement process of the octree cube structure is performed based on the distance from the grasping position of the gallbladder model. This approach improves the resolution of the octree in the area near the grasping position where relatively large deformations occur. In addition, it makes it easier to detect interference between the grasp model and the high-resolution grid of the octree. Simulation results showed that there were 199 cubes before and 339 cubes after refinement, and the FPS decreased from 44.1 FPS to 32.4 FPS on a standard CPU and GPU PC, which is still within real-Time processing.

DOI： 10.1117/12.3006816

HTSeg: Hybrid Two-Stage Segmentation Framework for Intestine Segmentation from CT Volumes

An Q., Oda H., Hayashi Y., Kitasaka T., Takimoto A., Hinoki A., Uchida H., Suzuki K., Oda M., Mori K.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) Vol. 15196 LNCS page： 32 - 41 2024

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

This paper proposes a semi-supervised intestine segmentation method from CT volumes. Our method can use densely and sparsely annotated CT volumes for training to reduce the labor of manually annotating intestines. The proposed Hybrid Two-stage Segmentation (HTSeg) framework consists of two networks, a 2D swin-transformer-based network as the first stage and a 3D network as the second stage. In the first stage, we use 6964 labeled CT slices to train the 2D Swin U-Net. The trained 2D Swin U-Net is used to generate pseudo-labels for sparse annotation data. In the second stage, we use sparsely annotated datasets with pseudo-labels and densely annotated datasets to train a 3D multi-view symmetrical network (MVSNet). Experimental results showed that the Dice score of the proposed method was 74.70%, which was 1.03% higher than just using MVSNet. Compared with the other four previous methods (3D U-Net, CPS, EM, MT), the proposed method produced competitive segmentation performance. The code can be found at: https://github.com/MoriLabNU/semi-pseudo-labels.

DOI： 10.1007/978-3-031-73083-2_4

Performance Improvement for Medical Image Classification Model by using Gradient-based Analytical Feature Selection

Toda, R; Itoh, H; Oda, M; Hayashi, Y; Otake, Y; Hashimoto, M; Akashi, T; Aoki, S; Mori, K

COMPUTER-AIDED DIAGNOSIS, MEDICAL IMAGING 2024 Vol. 12927 2024

Publisher：Progress in Biomedical Optics and Imaging - Proceedings of SPIE

This paper presents a gradient-based analytical method for improving medical image classification. The automated classification of diseases is important in computer-aided diagnosis. In addition to accurate classification, its explainability is an essential factor toward its practical application. A gradient-based visual explanation provides the explainability of a model of an convolutional neural network (CNN) by indicating important patterns in an input image. Most studies use this explanation to assess CNN’s validity in a qualitative manner. On the other hand, in addition to a model’s explainability, our motivation is to utilize the visual-explanation methods to enhance the classification accuracy of a CNN model. In this study, we propose a weight-analysis-based method to improve the classification accuracy of a trained-CNN model without additional training. The proposed method selects important patterns based on a gradient-based weight analysis of a middle layer in a trained model and suppresses irrelevant patterns in the extracted features for the classification. We applied our analytical method to a convolutional and a global-average-pooling layers in a CNN, which classifies a chest CT volume into COVID-19 typical and non-typical cases. As shown in classification results on 302 testing cases, our method improved the accuracy of the COVID-19 classification.

DOI： 10.1117/12.3006620

　 PREV - NEXT 　

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19