Abstract

The translation gap between published research and clinical impact is severe: tens of thousands of AI and statistical papers on medical imaging are published annually, yet fewer than 0.1% reach patients. This talk examines why methodological rigor alone does not drive market adoption. Through case studies of research projects of mammography image-analysis and risk prediction, from uncertainty quantification to domain adaptation, we identify the disconnect between academic validation and clinical requirements. Key insights include the importance of early clinician engagement, the divergence between statistical metrics and clinical outcomes, and the regulatory and validation frameworks necessary for commercialization. We present the complete pathway from research to market exit, including regulatory strategy, prospective validation design, and go-to-market mechanics. The evidence demonstrates that sustainable AI requires integration of scientific rigor, clinical understanding, and business viability from the outset.

Speakers

Shu (Joy) Jiang, PhD, Associate Professor of Surgery, Division of Public Health Sciences at Washington University Medicine

Graham A. Colditz, MD, DrPH, Director, Division of Public Health Sciences, Niess-Gain Professor of Surgery, Associate Director of Prevention and Control, Siteman Cancer Center, Department of Surgery at Washington University Medicine

Moderator

Dr. Jiguo Cao, Canada Research Chair in Data Science; Professor of Statistics and Actuarial Science, Simon Fraser University 

Registration Coming Soon!

About the Speakers

Dr. Shu (Joy) Jiang is an Associate Professor in the Division of Public Health Science and Director of the Epidemiology & Biostatistics Imaging Research Center at Washington University School of Medicine. She is known for her methodological contributions in breast cancer prevention and AI in high-dimensional imaging. Dr. Jiang is also an accomplished entrepreneur; she co-founded Prognosia Inc., an AI-driven cancer risk assessment startup that secured an FDA Breakthrough Device Designation prior to its acquisition by Lunit. Her work has been recognized with the ASA Annie T. Randall Innovator Award, the NCI MERIT Award, and the Forbes 30 Under 30 list, among many others. She serves as an Associate Editor for Biometrics. See Profile

Dr. Graham Colditz is the Niess-Gain Professor at WashU Medicine and Associate Director for prevention and control at Siteman Cancer Center. He is highly cited for his research and has received numerous national awards from AACR, ACS, and ASCO for his contributions to cancer prevention. He is also the Co-Founder of Prognosia Inc, with Shu (Joy) Jiang. He has a long history of translating public health research to prevention and has developed a range of models and online tools integrating epidemiologic data to estimate risk and guide prevention.  With Dr. Jiang he has focused on developing and applying AI technology for breast cancer risk prediction, Prognosia Breast, that secured Breakthrough Device Designation from the FDA.  See Profile

About the Moderator

Dr. Jiguo Cao is a Canada Research Chair in Data Science and a Professor in the Department of Statistics and Actuarial Science at Simon Fraser University in Burnaby, BC, Canada. His research focuses on functional data analysis and machine learning. In 2021, he received the CRM–SSC Award from the Statistical Society of Canada and the Centre de recherches mathématiques in recognition of his outstanding research contributions. See Profile

About the NISS-CANSSI Collaborative Data Science Web Series:

The NISS-CANSSI Collaborative Data Science initiative that the National Institute of Statistical Sciences (NISS) in collaboration with the Canadian Statistical Sciences Institute (CANSSI) brings together experts from various fields to tackle complex data challenges through interdisciplinary teamwork and innovative methodologies.

Goals of the Initiative

The goal is to foster progress in:

• Developing new ideas for experimental and observational data-driven learning and discovery that address key questions at the cutting edge of science and scientific deduction;
• Quantifying and summarizing uncertainty in data-driven theories, as well as complex Data Science models, algorithms, and workflows; and
• Establishing new practices for scientific reproducibility and replicability through Data Science.

Learn more: NISS-CANSSI Collaborative Data Science