Yen-Wei Chen received the B.E. degree in 1985 from Kobe Univ., Kobe, Japan, the M.E. degree in 1987, and the D.E. degree in 1990, both from Osaka Univ., Osaka, Japan. He was a research fellow with the Institute for Laser Technology, Osaka, from 1991 to 1994. From Oct. 1994 to Mar. 2004, he was an associate Professor and a professor with the Department of Electrical and Electronic Engineering, Univ. of the Ryukyus, Okinawa, Japan. He is currently a professor with the college of Information Science and Engineering, Ritsumeikan University, Japan. He is the founder and the first director of Center of Advanced ICT for Medicine and Healthcare, Ritsumeikan University. He was a chair professor with the college of computer technology and science, Zhejiang University, China during 2014-2016. His research interests include medical image analysis, computer vision and computational intelligence. He has published more than 300 research papers in a number of leading journals and leading conferences including IEEE Trans. Image Processing, IEEE Trans. Medical Imaging, CVPR, ICCV, MICCAI. He has received many distinguished awards including ICPR2012 Best Scientific Paper Award, 2014 JAMIT Best Paper Award. He is/was a leader of numerous national and industrial research projects.
Speech Title: "Recent Advances in Medical Image Segmentation Using Deep Learning"
Abstract: Recently, deep learning (DL) plays important roles in many academic and industrial areas especially in computer vision and image recognition. Deep learning uses a neural network with deep structure to build a high-level feature space. It learns data-driven, highly representative, hierarchical image features, which have proven to be superior to conventional hand-crafted low-level features and mid-level features. Deep learning (DL) has also been applied to medical image analysis and achieved state-of-the-art results. Medical images is one of the most important pre-processing steps in computer-aided diagnosis, but it is a challenging task because of limited annotated data, high dimensionality and domain shift between training and test data. In this talk, I will talk about recent advances and solutions for these challenges in medical image segmentation. I will first introduce deep atlas prior, in which we combined semi-supervised deep learning with anatomic atlas as prior information to solve the problem of limited annotated data. Then I will introduce a patch-free 3D medical image segmentation method. As third topic, I will introduce domain adaptation and domain generalization for domain shift problem in medical image segmentation. I will also discuss futures of DL in medical imaging.
C Krishna Mohan received the M.Tech degree from National Institute of Technology Karnataka, Surathkal, India, in 2000 and the Ph.D. degree from Indian Institute of Technology Madras, India, in 2007, all in Computer Science and Engineering. Dr. C. Krishna Mohan is currently a Professor in the Department of Computer Science and Engineering at the Indian Institute of Technology Hyderabad (IIT Hyderabad), India. He is the founding member of IIT Hyderabad working since 2009. He was also the first Head of the Department of Computer Science and Engineering, IIT Hyderabad from May 2010 till October 2014. He worked as Dean Public and Corporate Relations from Jan 2020 till Jan 2023. Before joining IIT Hyderabad, he was a senior faculty member at the National Institute of Technology Karnataka, Surathkal. He held a Visiting Professor's position at Nihon University, Japan. He was a Visiting Researcher at Carnegie Mellon University, Pittsburgh, USA during the summer of 2007. Also, he was a Visiting Researcher at University of California Irvine, California during the summer of 2013. Dr. Krishna Mohan is a Senior Member of IEEE, Member of ACM, AAAI Member, Fellow of Telangana Academy of Sciences (FTAS), Fellow of IETE, Fellow of IEI, and Life Member of ISTE.
Speech Title: "Federated Domain Adaptation in Medical Imaging"
Abstract: Deep learning models have shown their advantage in many different tasks, including medical image analysis. In order to train the deep learning models effectively, it is important to aggregate a significant amount of patient information. And, assembling large medical imaging datasets at a single site is challenging due to the time and cost of acquiring and annotating the images. Although combining medical imaging data from multiple institutions is desirable, the need to ensure patient data privacy makes it difficult to create a centralized database. A viable solution to address data privacy issues is to adopt a decentralized approach using federated learning which helps in developing efficient models for various tasks such as disease classification, segmentation, and anomaly detection. By utilizing federated learning, we can leverage the collective knowledge from diverse healthcare institutions without the need to centrally store or share sensitive patient data. To ensure the efficacy and generalizability of models across these diverse domains, we propose to use domain adaptation approach. By integrating domain adaptation into practical clinical applications, the feature distributions of different domains can be aligned, enabling models to effectively transfer knowledge and achieve yrobust performance in new and unseen environments.
Guohua Cao is an Associate Professor and heads the X-ray Systems Lab at ShanghaiTech University's School of Biomedical Engineering in China. He earned his PhD from Brown University in the United States after completing his undergraduate studies at the University of Science and Technology of China. Prior to joining ShanghaiTech in 2021, he held positions as an Assistant Professor of Physics at the University of North Carolina at Chapel Hill, as well as an Assistant Professor of Biomedical Engineering and Computer Science at Virginia Tech. Dr. Cao's research is centered around biomedical imaging, focusing on developing innovative imaging tools. His team achieved a major breakthrough by creating a carbon nanotube micro-CT that can capture detailed images of a beating mouse heart. He also pioneered a stationary CT architecture that holds potential for stop-action cardiac CT examinations. He has published more than 100 research papers in respected journals and conference proceedings and received several awards including a prestigious NSF CAREER award in 2014.
Speech Title: "Architectures and Algorithms of Ultraportable CT for Point-of-Care Medical Imaging"
Abstract: The rising and aging human population is putting pressure on the traditional hospital-based medical care. With the accelerated digitalization in our society, it is projected that non-acute care particularly the diagnosis and treatment of chronic diseases and conditions will be more shifted to remote care. This paradigm shift demands the growing use of point-of-care medical imaging technologies that can bring imaging to the patient. In this talk, I will focus on our recent works on the architecture design and algorithm development for an ultraportable CT. After introducing some recently proposed CT architectures for compact CT, I will present our analysis on their associated system costs, image formation algorithms, as well as expected image qualities. Potentials of ultraportable CT in future digital medicine will be discussed.