We live in the age of algorithms. Artificial intelligence and medical imaging communities increasingly use machine learning techniques to interpret large health-related image data. In theory, machine learning should reduce the burden on radiologists and pathologists and would be more cost-effective and provide more consistent image interpretation than humans. But in real world clinical care, medical images alone may not be sufficient in making accurate diagnoses and optimized treatment strategies, while the mathematical models being used today are unregulated and sometimes even uncontestable. In this talk, I will share the construction and augmentation of multimodal data for two of our artificial intelligence apps for breast cancer screening and acute ischemic stroke detection, both clinical procedures incurring high risk of misdiagnosis. Medical care is disease driven. Big data is only as good as the people wielding it. Understanding the context of disease conditions and shortfall of existing clinical procedures will help to define input data and constraints in constructing machine learning models to improve clinical care.
Dr. Wong holds the John S. Dunn, Sr. Distinguished Endowed Chair in Biomedical Engineering; he is a Professor of Radiology, Pathology, Laboratory Medicine, Neurology, and Neurosciences, the Associate Director of Translational Research at Methodist Cancer Center, and Chief of Medical Physics and Chief Research Information Officer at Houston Methodist Hospital. He serves as the Founding Director of the Ting Tsung and Wei Fong Chao Center for BRAIN (Bioinformatics Research and Imaging in the Neurosciences) and Founding Director of the Center for Modeling Cancer Development at Houston Methodist Research Institute. He also holds a dozen of other academic posts across institutions in Texas Medical Center as well as overseas universities and medical schools. Dr. Wong has more than twenty years of research and management experience in industry and academia, including Hewlett-Packard, AT&T Bell Laboratories, the Japanese Fifth Generation Computer Systems Project, Philips Medical Systems and Royal Philips Electronics, Charles Schwab, University of California - San Francisco/Berkeley, Harvard University and Houston Methodist Hospital. He received his senior executive education from the MIT Sloan School of Management, Stanford University Graduate School of Business and Columbia University Graduate School of Business. He holds many patents and has published over 300 peer-reviewed papers and four books. He also serves on and chairs NIH study panels, conference program committees, and the editorial boards of twelve scientific journals.