Dennis Wei, Senior Research Scientist at IBM specializing in human-centered trustworthy AI, speaks with Pitt’s HexAI podcast host, Jordan Gass-Pooré, about his work focusing on trustworthy machine learning, including interpretability of machine learning models, algorithmic fairness, robustness, causal inference and graphical models.

Concentrating on explainable AI, they speak in depth about the explainability of Large Language Models (LLMs), the field of in-context explainability and IBM’s new In-Context Explainability 360 (ICX360) toolkit. They explore research project ideas for students and touch on the personalization of explainability outputs for different users and on leveraging explainability to help guide and optimize LLM reasoning. They also discuss IBM’s interest in collaborating with university labs around explainable AI in healthcare and on related work at IBM looking at the steerability of LLMs and combining explainability and steerability to evaluate model modifications.

This episode provides a deep dive into explainable AI, exploring how the field's cutting-edge research is contributing to more trustworthy applications of AI in healthcare. The discussion also highlights emerging research directions ideal for stimulating new academic projects and university-industry collaborations.

Guest profile: https://research.ibm.com/people/dennis-wei

ICX360 Toolkit: https://github.com/IBM/ICX360

Jason Moore, Chair of the Department of Computational Biomedicine and Director of the Center for Artificial Intelligence Research and Education (CAIRE) at Cedars-Sinai Medical Center in Los Angeles, CA, speaks with Pitt’s HexAI podcast host, Jordan Gass-Pooré, about his work, the strategic investments his center is making in technology and specialized human expertise to support advanced AI research and about the incorporation of AI and AI agents into precision health.

They speak in depth about the recent and rapid emergence of agentic AI, which is expected to have a significant impact on healthcare and how his team’s work is advancing the field. They also touch on vetting, deploying, and monitoring AI models for clinical use; explainable AI, trust, and transparency; using AI chatbots to improve the patient experience; the importance of building effective collaborations between industry and academia; and Cedar-Sinai’s new PhD program in Health AI.

Peter Maurer, Assistant Professor of Molecular Engineering at the University of Chicago Pritzker School of Molecular Engineering, speaks with Pitt’s HexAI podcast host, Jordan

Gass-Pooré, about the future impact of quantum sensing on biomedical research and diagnostics.

Peter's research lab leverages the extreme environmental sensitivity of quantum systems to develop powerful sensors suitable for cutting-edge biological research that are optically addressable and can operate under ambient conditions. He outlines both near-term and future applications of powerful quantum sensors in pathology and laboratory medicine. He provides a key example of how these sensors could enable a new type of nanoscale NMR spectroscopy, capable of measuring magnetic fields from biomolecules to non-invasively probe their chemical information and signaling pathways. In the near future, he points to diagnostic tools, currently being developed by companies, that use the unique optical signatures of quantum sensors for highly sensitive, background-free protein detection in small volumes. For the long term, he envisions the technology as a "field opener" for studying protein aggregation in neurodegenerative diseases like Alzheimer's and Parkinson’s.

Peter outlines how AI can be applied to analyze complex data from sensors that respond to multiple environmental factors and highlights the challenge of bringing together experts from quantum technology, biophysics, and medicine who can "talk each other's language.” He also touches on how the use of synthetic data in quantum sensing is a "completely under-appreciated" area with the potential to analyze complex environmental properties that would otherwise be missed by looking at single types of measurements. To advance the field from academic proofs-of-concept to clinical tools, he stresses the need for collaboration with academic and industry partners who can help engineer robust, "turnkey" systems that can be widely tested and used.

The University of Pittsburgh Health and Explainable AI podcast is a collaborative initiative between the Health and Explainable AI (HexAI) Research Laboratory in the Department of Health Information Management at the School of Health and Rehabilitation Sciences, and the Computational Pathology and AI Center of Excellence (CPACE), at the University of Pittsburgh School of Medicine.

Hosted by Jordan Gass-Pooré, a health and science reporter, this podcast series explores the transformative integration of responsible and explainable artificial intelligence into health informatics, clinical decision-making, and computational medicine. From reshaping diagnostic accuracy to enhancing patient care pathways, we'll highlight how AI is creating new bridges between researchers, clinicians, and healthcare innovators.

Led by Ahmad P. Tafti, Hooman Rashidi and Liron Pantanowitz, the HexAI podcast is committed to democratizing knowledge around ethical, explainable, and clinically relevant AI. Through insightful conversations with domain experts, AI practitioners and students will spotlight the latest breakthroughs, discuss real-world applications, and unpack the challenges and opportunities that lie ahead in responsible AI in healthcare. So whether you're a student, practitioner, researcher, or policymaker, this is your gateway to the future of AI-powered healthcare