Embedded Files
Multi-Scale Perspectives on AI in Medicine
Multi-Scale Perspectives on AI in Medicine
09:15 - 10:00 Keynote 1
Over the past few years, leaps in AI’s capabilities have inspired society’s imagination and disrupted several industries already. In this talk, I will present multi-scale perspectives how AI is starting to generate impact in the field of medicine. I will present the WYSIWYG framework of forecasting how AI will develop over the coming years and offer suggestions on what skills organizations and individuals can develop to generate values which are still challenging for AI to provide.
About the Speaker
Eric Chang
Executive Director, Taiwan Artificial Intelligence Foundation
Eric Chang
Executive Director, Taiwan Artificial Intelligence Foundation
Eric Chang is an advisor of startup founders in AI + Health space. He is a Venture Partner at W I Harper and an Executive Director of Taiwan Artificial Intelligence Foundation. He was previously the Partner Director of Technology Strategy at Microsoft Research Asia, where he worked on IP portfolio management, starting research themes, and guiding technologies to generate real world impact. Previously, Eric co-founded Microsoft Advanced Technology Center (ATC) in 2003 where he led teams to develop Office, Windows, and other Microsoft products and started a multidisciplinary incubation team.
Prior to joining Microsoft Research, Eric was one of the founding members of the Research group at Nuance Communications, a pioneer in natural speech interface. While at Nuance, Eric led the development of the world’s first deployed Japanese natural language speech recognition system.
Eric graduated from M.I.T. with Ph.D., Master and Bachelor degrees, all in the field of electrical engineering and computer science. Eric has published in the fields of eHealth, speech processing, and machine learning. He is the author of over 15 granted and pending patents.
Identifying Good Biological and Clinical Questions to Challenge AI
Identifying Good Biological and Clinical Questions to Challenge AI
Ie Ming Shih
Professor, School of Medicine, Johns Hopkins University, USA
Ie Ming Shih
Professor, School of Medicine, Johns Hopkins University, USA
10:00 - 10:45 Keynote 2
About the Speaker
Dr. Ie-Ming Shih is the Richard TeLinde Distinguished Professor in the Department of Gynecology and Obstetrics at the Johns Hopkins University School of Medicine. Dr. Shih directs the inter-departmental TeLinde Gynecologic Disease Program and co-directs the Women‘s Malignancy Program at the Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins Medical Institutions. He is the Principal Investigator of the NIH Specialized Program of Research Excellence (SPORE) of Ovarian Cancer focusing on concept-driven clinical trials and early detection of the disease. Dr. Shih's research has been continuously supported by National Institute of Health.
Dr. Shih received his M.D. from Taipei Medical University and earned his Ph.D. from the University of Pennsylvania. Dr. Shih went on completing a clinical fellowship in gynecologic pathology and a research fellowship in cancer genetics at Johns Hopkins before joining the faculty in 2001.
The scientific contributions from his team include 1) molecular characterization of different types of ovarian cancer, 2) elucidation of the origin and evolution trajectory of ovarian precancerous lesions, 3) the discovery of cancer- promoting genes that are amenable for targeting, 4) the etiology of endometriosis and 5) the pathogenesis of gestational trophoblastic neoplasm.
Dr. Shih has published more than 360 original publications and book chapters related to basic and translational cancer research, especially in the study of gynecological malignancies. He has received the 2017 Most Influential Research Paper Award in gynecologic diseases by the Columbia Hospital Foundation, and several research awards from the Tina’s Wish Bozeman Foundation, Gray Foundation, and the Endometriosis Foundation of America to support and recognize his research works.
He sits in several advisory/editorial boards such as the World Health Organization (WHO) Nomenclature Committee for Gynecologic Neoplasms and the NCI Ovarian Task Force of Gynecologic Cancer Steering Committee. Besides his clinical, research and teaching obligations, Dr. Shih is a passionate photographer.
Unsupervised Neural and Neurosymbolic Architectures for Analysis and Interpretation of Medical Imaging
Unsupervised Neural and Neurosymbolic Architectures for Analysis and Interpretation of Medical Imaging
11:00 - 11:40 Keynote 3
Annotation of medical imaging data is notoriously time-consuming, expensive, prone to human biases, and hard to reconcile with the insatiable demands of contemporary machine learning. The models trained on annotated data are often narrow in focusing on features that are specific to a given context (anomaly, pathology, etc.) rather than discovering and capturing more general characteristics of observed structures and processes, which may make them susceptible to deceptive image features and lead to inferior generalization. In this talk, I will argue for stronger involvement of unlabelled data in construction of analytic and diagnostic ML models for medical imaging and review a number of relevant techniques we studied and devised in my research group, including recent neurosymbolic architectures trained via auto-association, in which a generative decoder synthesizes a physically/biologically plausible structural model that explains the observed image. The proposed architectures can learn effectively from small data sets, offer better interpretability than conventional models, and promise better diagnostic accuracy.
About the Speaker
Krzysztof Krawiec
Professor, Institute of Computing Science, Poznan University of Technology
Krzysztof Krawiec
Professor, Institute of Computing Science, Poznan University of Technology
Krzysztof Krawiec is a full professor Professor of Computer Science at Poznan University of Technology, Poland. His primary research areas include neurosymbolic systems, computer vision, medical imaging, and program synthesis. Krzysztof co-authored over 180 publications on the above and related topics, received the Fulbright Senior Advanced Research Award, and was a visiting professor at University of California and Massachusetts Institute of Technology.
He serves as an advisor at the Confederation of Laboratories for Artificial Intelligence in Europe and co-founded the Center for Artificial Intelligence and Machine learning, part of the Horizon 2020 Foundations of Trustworthy AI project funded by the European Commission. He acts also as the Chief AI Advisor at Optopol Technology, a leading manufacturer of ophthalmic imaging systems, and as the CTO of Hylomorph Solutions Ltd.
Page updated
Google Sites
Report abuse