Responsible AI in Healthcare With: Dr. Nigam Shah

Dr. Nigam Shah, Co-founder of Atropos Health and Chief Data Scientist at Stanford Healthcare, shares his framework for assessing AI’s role in improving patient care. He describes the 'firm assessment' process, which evaluates expected benefits, feasibility, and incentives before deploying models. Shah also critiques current AI-in-healthcare research for lacking real-world EHR data and focusing solely on accuracy metrics, advocating instead for task-specific evaluation and uncertainty measures. He emphasizes that impactful AI requires starting with the right clinical question, selecting datasets fit for that question, and integrating evaluations into the organization’s responsible AI lifecycle.

About the Guest

Dr. Nigam Shah is Co-founder of Atropos Health and Chief Data Scientist at Stanford Healthcare, specializing in real-world data and AI-driven clinical insights. Learn more at: https://www.linkedin.com/in/nigam/

Notable Quote

"You have to score the value of a dataset in light of the question you have."

Key Takeaways

  • Impact measurement starts with defining the right clinical question.
  • Responsible AI lifecycle integrates assessment into organizational workflows.
  • Data fitness depends on the specific patient care question being asked.

Transcript Summary

How do you evaluate AI’s impact on patient care?

At Stanford, we use a 'firm assessment'—a structured process for defining expected outcomes, assessing incentives, and simulating impact before deployment, embedded in a responsible AI lifecycle.

What are the current benchmarking gaps?

Most LLM studies in healthcare lack real EHR data and focus on accuracy without uncertainty metrics. We need clear tasks, relevant data, and appropriate evaluation methods.

How should data be used?

Start with the question, then score datasets for fitness to that question to avoid wasted resources.

About the Series

AI and Healthcare—with Mika Newton and Dr. Sanjay Juneja is an engaging interview series featuring world-renowned leaders shaping the intersection of artificial intelligence and medicine.

Dr. Sanjay Juneja, a hematologist and medical oncologist widely recognized as “TheOncDoc,” is a trailblazer in healthcare innovation and a rising authority on the transformative role of AI in medicine.

Mika Newton is an expert in healthcare data management, with a focus on data completeness and universality. Mika is on the editorial board of AI in Precision Oncology and is no stranger to bringing transformative technologies to market and fostering innovation.

Get new episode alerts via email

By clicking the Submit button, you agree to xCures's Terms of Service and Privacy Policy.