EP13 Dong Yao, MD and Shivam Vedak, MD Stanford (Part 2)
Engineering the Future — AI Documentation, Bias, and Clinical Workflows (Part 2)
Guests: Dr. Dong Yao and Dr. Shivam Vedak, Emergency Physicians and Clinical Informaticians, Stanford University
Release Date: 6/5/25
Episode Summary:
In Part 2 of our conversation with Stanford's Dr. Dong Yao and Dr. Shivam Vedak, we dive deeper into the real-world deployment of AI tools across hospital settings. This episode explores how large language models are transforming charting, documentation, and even helping identify bias in clinical workflows.
DJ and Ethan explore recent research on disparities in cardiac testing and how national-scale EHR data platforms like Epic Cosmos are enabling new insights into equity, care variability, and system-level improvement. Dong and Shivam then return to reflect on where AI can realistically deliver value today — and what still needs work.
From automation bias and trust in AI assistants to strategies for responsible deployment and monitoring, this episode balances practical applications with thoughtful foresight about how to keep clinicians in the loop and patients at the center.
In This Episode, We Cover:
🧾 The promise and pitfalls of AI-generated clinical documentation.
🩺 How trust and user phenotype influence adoption of ambient AI tools.
🧠 The role of education in preparing clinicians to evaluate and use AI outputs critically.
⚖️ Disparities in ED cardiac testing and the insights from massive EHR datasets.
🔄 Model bias, fairness frameworks, and system-level monitoring strategies.
🔧 Why summarization is harder than it looks — and what might come next.
Referenced Study:
Taylor RA, et al. Disparities in Troponin and ECG Testing Among ED Patients With Chest Pain. American Journal of Emergency Medicine, 2024.
Takeaway Quote:
“Automation bias is real — and we have to build systems to recognize when it’s happening and how to catch it.” — Dong Yao, MD
Connect with the Guests:
Follow STAT AI:
🌐 Website: stataipodcast.com
🐦 Twitter: @STATaipodcast
🎧 Listen: Apple Podcasts | Spotify | Prime Music
