|Ryoma Bise||Last modified date：2021.06.23|
Associate Professor / Data science
Department of Advanced Information Technology
Faculty of Information Science and Electrical Engineering
Department of Advanced Information Technology
Faculty of Information Science and Electrical Engineering
E-Mail *Since the e-mail address is not displayed in Internet Explorer, please use another web browser:Google Chrome, safari.
Reseacher Profiling Tool Kyushu University Pure
Interdisciplinary Information Studies
Country of degree conferring institution (Overseas)
Field of Specialization
Computer Vision, Bio medical image analysis, bio image informatics, pattern recognition
Total Priod of education and research career in the foreign country
Research InterestsMembership in Academic Society
- Computer Vision, Bioimage informatics, Biomedical Image Analysis, Pattern Recognition, Image recoginition, Image processing
keyword : Computer Vision, Bioimage informatics, Biomedical Image Analysis, Pattern Recognition, Image recoginition, Image processing
- In the biological research, cell and molecular level biological phenomena is observed as 2D and 4D(3D+time) images by new imaging technologies to elucidate the principle of biological phenomena. In such research, quantification and data analysis technologies for the biological images captured in a state similar to in vivo are important. However, in vivo, observation targets such as cells and molecules are distributed in a high density. This makes it difficult to quantify the metrics of behaviors and shapes from such images, and this is critical in biological research.
In this research, we are developing multi-object tracking methods for realizing robustness and versatility in high density, and we will apply the tracking methods and data analysis techniques to analyze the behaviors of cells and molecules. This makes the objective data analysis and scale up to big data easy. We believe that this will contribute to innovation in the biological field.
- With the arrival of super-aging society, there is increasing demand for technical support to enable people to continue working while preserving their health and beauty. NII has participated in ImPACT to realize an early diagnosis of disease, and inspection of the internal structure, with advances in photo-acoustic imaging, which performs real-time 3D visualization of changes in properties and functions inside human bodies and substances, non-invasively and non-destructively. The photo-acoustic system is a promising new technology that integrates state-of-the-art laser and ultrasound technologies, where 3D structures of objects can be reconstructed by sensing emitted ultrasound from the objects that absorb near-infrared irradiation. It enables to image the state of the human body and objects whose insides are not visible, non-invasively and non-destructively. In this research, we develop computer-vision technologies to obtain clear images and extract bio-image features to support a diagnosis. For example, we proposed a registration method to generate high-quality 3D volumes in which vessels become clearly visible by aligning shot-volumes that are misaligned by body motions. We are also developing a technology that automatically models vascular structures, which helps in understanding blood vessel conditions strongly related to illnesses.
|1.||Shota Harada, Ryoma Bise, Hideaki Hayashi, Kiyohito Tanaka, and Seiichi Uchida, Soft and Self Constrained Clustering for Group-Based Labeling, Medical Image Analysis, https://doi.org/10.1016/j.media.2021.102097, 2021.05, When using deep neural networks in medical image classification tasks, it is mandatory to prepare a large-scale labeled image set, and this often requires significant effort by medical experts. One strategy to reduce the labeling cost is group-based labeling, where image samples are clustered and then a label is attached to each cluster. The efficiency of this strategy depends on the purity of the clusters. Constrained clustering is an effective way to improve the purity of the clusters if we can give appropriate must-links and cannot-links as constraints. However, for medical image clustering, the conventional constrained clustering methods encounter two issues. The first issue is that constraints are not always appropriate due to the gap between semantic and visual similarities. The second issue is that attaching constraints requires extra effort from medical experts. To deal with the first issue, we propose a novel soft-constrained clustering method, which has the ability to ignore inappropriate constraints. To deal with the second issue, we propose a self-constrained clustering method that utilizes prior knowledge about the target images to set the constraints automatically. Experiments with the endoscopic image datasets demonstrated that the proposed methods give clustering results with higher purity..|
|2.||H. Tokunaga, B.K. Iwana, Y. Teramoto, A. Yoshizawa, and R. Bise, Negative Pseudo Labeling using Class Proportion for Semantic Segmentation in Pathology, 16th European Conference on Computer Vision (ECCV2020), 2020, 2020.08, We propose a weakly-supervised cell tracking method that can train a convolutional neural network (CNN) by using only the annotation of "cell detection" (i.e., the coordinates of cell positions) without association information, in which cell positions can be easily obtained by nuclear staining. First, we train a co-detection CNN that detects cells in successive frames by using weak-labels. Our key assumption is that the co-detection CNN implicitly learns association in addition to detection. To obtain the association information, we propose a backward-and-forward propagation method that analyzes the correspondence of cell positions in the detection maps output of the co-detection CNN. Experiments demonstrated that the proposed method can match positions by analyzing the co-detection CNN. Even though the method uses only weak supervision, the performance of our method was almost the same as the state-of-the-art supervised method..|
|3.||K. Nishimura, J. Hayashida, C. Wang, D.F.E. Ker, and R. Bise, Weakly-Supervised Cell Tracking via Backward-and-Forward Propagation, 16th European Conference on Computer Vision (ECCV2020), 2020, https://doi.org/10.1007/978-3-030-58610-2_7, pp.104-pp.121, 2020.08, [URL], We propose a weakly-supervised cell tracking method that can train a convolutional neural network (CNN) by using only the annotation of “cell detection” (i.e., the coordinates of cell positions) without association information, in which cell positions can be easily obtained by nuclear staining. First, we train co-detection CNN that detects cells in successive frames by using weak-labels. Our key assumption is that co-detection CNN implicitly learns association in addition to detection. To obtain the association, we propose a backward-and-forward propagation method that analyzes the correspondence of cell positions in the outputs of co-detection CNN. Experiments demonstrated that the proposed method can associate cells by analyzing co-detection CNN. Even though the method uses only weak supervision, the performance of our method was almost the same as the state-of-the-art supervised method. Code is publicly available in https://github.com/naivete5656/WSCTBFP..|
|4.||J. Hayashida, K. Nishimura and R. Bise, MPM: Joint Representation of Motion and Position Map for Cell Tracking, IEEE CVPR2020, https://doi.org/10.1109/CVPR42600.2020.00388, 3822-3831, 2020.06, Conventional cell tracking methods detect multiple cells in each frame (detection) and then associate the detection results in successive time-frames (association). Most cell tracking methods perform the association task independently from the detection task. However, there is no guarantee of preserving coherence between these tasks, and lack of coherence may adversely affect tracking performance. In this paper, we propose the Motion and Position Map (MPM) that jointly represents both detection and association for not only migration but also cell division. It guarantees coherence such that if a cell is detected, the corresponding motion flow can always be obtained. It is a simple but powerful method for multi-object tracking in dense environments. We compared the proposed method with current tracking methods under various conditions in real biological images and found that it outperformed the state-of-the-art (+5.2\% improvement compared to the second-best)..|
|5.||H. Tokunaga, Y. Teramoto, A. Yoshizawa, R. Bise, Adaptive Weighting Multi-Field-of-View CNN for Semantic Segmentation in Pathology, IEEE CVPR2019, 10.1109/CVPR.2019.01288, 2019.06.|
|6.||Ryoma Bise, Yoichi Sato, Cell Detection From Redundant Candidate Regions Under Nonoverlapping Constraints, IEEE Transactions on Medical Imaging, 10.1109/TMI.2015.2391095, 34, 7, 1417-1427, 2015.07, Cell detection in microscopy images is essential for automated cell behavior analysis including cell shape analysis and cell tracking. Robust cell detection in high-density and low-contrast images is still challenging since cells often touch and partially overlap, forming a cell cluster with blurry intercellular boundaries. In such cases, current methods tend to detect multiple cells as a cluster. If the control parameters are adjusted to separate the touching cells, other problems often occur: a single cell may be segmented into several regions, and cells in low-intensity regions may not be detected. To solve these problems, we first detect redundant candidate regions, which include many false positives but in turn very few false negatives, by allowing candidate regions to overlap with each other. Next, the score for how likely the candidate region contains the main part of a single cell is computed for each cell candidate using supervised learning. Then we select an optimal set of cell regions from the redundant regions under nonoverlapping constraints, where each selected region looks like a single cell and the selected regions do not overlap. We formulate this problem of optimal region selection as a binary linear programming problem under nonoverlapping constraints. We demonstrated the effectiveness of our method for several types of cells in microscopy images. Our method performed better than five representative methods, achieving an F-measure of over 0.9 for all data sets. Experimental application of the proposed method to 3-D images demonstrated that also works well for 3-D cell detection..|
- Information Processing Society of Japan