62: Trusworthy by Design: Context-Rich AI in Healthcare with Ben Lengerich

SummaryBen Lengerich discusses the importance of context in AI for healthcare, the role of generalized additive models (GAMs), and the challenges of data quality and data compliance. He emphasizes the need for responsible AI practices and highlights the impact of historical data on current medical practices. The discussion also touches on the future of personalized medicine and the necessity of investing in AI to improve healthcare outcomes.Ben Lengerich is an assistant professor of Statistics at the University of Wisconsin–Madison and the founder of Intelligible, where he develops context-adaptive, interpretable AI methods to turn real‑world clinical data into reliable evidence for precision medicine. His research sits at the intersection of machine learning, computational genomics, and medical informatics, with a focus on models that are transparent to clinicians and that account for the specific health context of each patient. Before joining UW–Madison, he was a postdoctoral associate and Alana Fellow at MIT CSAIL and the Broad Institute, advised by Manolis Kellis, after earning his PhD in Computer Science and an MS in Machine Learning from Carnegie Mellon University, where he worked with Eric Xing on methods to uncover patterns in complex biomedical data.  Takeaways:AI systems must understand context in healthcare to be effective.Generalized additive models (GAMs) enhance interpretability in AI.Data quality is paramount for successful AI applications in healthcare.Debugging datasets can uncover systemic issues in healthcare.Surprising insights from predictive modeling can inform better practices.Responsible AI practices are crucial in medical applications.Historical data continues to influence current medical practices.Compliance with regulations is a significant challenge for AI in healthcare.Legacy infrastructure poses barriers to AI implementation.Investing in AI can lead to improved healthcare outcomes and efficiency.Chapters:00:00 Introduction to another AI Snack on AI in Healthcare: Data, Context, Interpretability02:02 Understanding Context in Healthcare AI04:54 Generalized Additive Models Explained07:41 The Importance of Data Quality10:53 Debugging Datasets in Healthcare13:50 Surprising Insights from Predictive Models16:52 Responsible AI in Medicine19:47 Historical Impact on Medical AI22:28 Compliance and Regulations in Medical AI25:50 Bridging Legacy Infrastructure with AI28:03 The Future of AI in Healthcare31:43 AI Literacy for Healthcare Providers34:45 The Case for AI Investment in HealthcareHyperlinks:Ben Lengerich:LinkedIn profileX profileIntelligible websiteAnastassia:Anastassia Lauterbach - LinkedInFirst Public Reading, Romy, Roby and the Secrets of Sleep (1/3)First Public Reading, Romy, Roby and the Secrets of Sleep (2/3)First Public Reading, Romy, Roby and the Secrets of Sleep (3/3)AI Snacks with Romy and Roby@romyandroby“Leading Through Disruption”AI EdutainmentThe AI Imperative BookRomy & Roby BookSubstack