Cardiac Digital Twins: Inside the Research Lab

What if doctors could test heart treatments on a computational replica of your cardiovascular system before touching you? 

In episode 902, we go inside NTT Research's Medical & Health Informatics Lab to explore the science of cardiac digital twins: personalized software models that simulate how individual patients respond to drugs and therapies.

Lab Director Joe Alexander, an M.D. cardiologist and Ph.D. biomedical engineer, explains how his team is building toward autonomous systems that could one day deliver heart failure treatment without human intervention. 

We discuss the gap between today's trial-and-error cardiology and true precision medicine, why mechanistic models matter more than black-box AI for clinical trust, and what must go right for this technology to reach patients.

In this conversation, you will learn:

How digital twin technology is evolving from industrial applications to healthcare and life sciencesWhy mechanistic models that explain causation may earn more trust than black-box AI that only predictsThe challenges of building autonomous systems for high-stakes, safety-critical decisionsWhat it takes to validate AI-driven systems in regulated industries like healthcareHow interdisciplinary teams combine engineering, medicine, and data science to tackle complex problemsThe timeline realities of deep R&D — and how to measure progress when commercialization is years away

Join us to ask your questions directly to Dr. Alexander and participate in this conversation about the future of AI in healthcare!

Key TakeawaysAutomate Precision to Outperform Human Standards 

The lab develops autonomous, closed-loop systems to manage acute heart failure with greater precision than manual intervention. This technology simultaneously adjusts multiple drug inputs to reduce myocardial oxygen consumption while maintaining stable perfusion.

Feedback loops immediately correct discrepancies between projected models and patient responses, optimizing recovery paths. This automation reduces variability in care and seeks to go beyond the limitations of human specialists in dynamic, high-pressure settings.

Deploy Causal Models Over Black-Box Algorithms

Strategies for complex environments should emphasize mechanistic models that explain cause-and-effect rather than depend solely on correlation-based AI. Dr. Alexander’s team builds cardiovascular digital twins using electrical analog frameworks to replicate specific physiological functions.

This approach mimics predictive maintenance in aviation by developing a mathematical model to monitor individual patient responses. Specific physiological rules enable transparent validation of medical decisions, unlike the opaque deep learning methods often used in standard AI applications.

Implement Graduated Autonomy for Risk Mitigation

High-stakes autonomous systems require a phased "human-in-the-loop" approach to ensure safety and regulatory compliance. Dr. Alexander sometimes describes the technology as a clinical co-pilot to assist physician decision-making before moving toward full automation. 

This graduated approach bridges the gap between proof-of-concept animal trials and clinical application. Successful implementation ultimately democratizes access to specialized care and promotes health equity in resource-limited settings.

Episode Participants

Joe Alexander, M.D., Ph.D., is Director of the MEI Lab at NTT Research. His background is in both engineering and medicine. After graduating with a degree in Chemical Engineering from Auburn University, he studied medicine as a fellow of the Medical Scientist Training Program at The Johns Hopkins University Medical School where he received both M.D. and Ph.D. degrees.

Michael Krigsman is a globally recognized analyst, strategic advisor, and industry commentator known for his deep business transformation, innovation, and leadership expertise. He has presented at industry events worldwide and written extensively on the reasons for IT failures. His work has been referenced in the media over 1,000 times and in more than 50 books and journal articles; his commentary on technology trends and business strategy reaches a global audience.