Building Medical Self-Diagnosis Chatbot—AI in Assisting Patients’ Medical Decisions
Tuesday, November 14, 2017
Artisan Court at the Brickyard (BYAC) 150, Tempe campus [map]
During this seminar, we discuss the technical details of building a self-diagnosis chatbot, and describe a set of deep learning, Bayesian network and statistical approaches that help build the “explainable and easy-to-debug inference and reasoning engine.” In addition, we describe semi-automated approaches using deep learning that enrich the medical knowledge base as well as NLP techniques to understand user’s daily “medical jargons.” At the end, we briefly discuss multi-modal approaches that automatically generate reports from medical images using a CNN, RNN and GAN network. A live demo will be given.
About the Speaker
Wei Fan is the currently the executive director of Tencent Medical AI Lab in Palo Alto, CA, Shenzhen and Beijing, China. Previously, he was senior director of Baidu Big Data Lab in Sunnyvale, CA; Associate Director Huawei Noah’s Ark Lab in Hong Kong; and Research Staff Member in IBM T.J. Watson Research. His current interests are using artificial intelligence to transform traditional medicine. He received his doctorate in Computer Science from Columbia University in 2001.