Practical AI in Medicine with Jason Strelzow
Get the full intelligence
Search transcripts, export clips, track mentions, and explore all topics from “Practical AI in Medicine with Jason Strelzow” inside PodZeus.
The Upper Hand Podcast dives into the transformative role of AI in medicine with Dr. Jason Strelzow, a hand surgeon and self-described 'AI savant' at Washington University. The conversation reveals that AI is no longer a futuristic concept but a present-day tool reshaping clinical workflows—from ambient listening systems that auto-generate medical notes to AI-powered research assistants that automate CV updates and grant writing. Strelzow emphasizes that while AI excels at summarizing, predicting, and organizing information, it lacks human judgment and experience, making it a powerful but secondary tool. The episode confronts pressing ethical dilemmas: liability for AI-generated billing errors, the risk of hallucinated research papers, and the danger of over-reliance on AI in clinical decision-making. Despite these risks, the consensus is clear—AI should augment, not replace, physicians. The most striking insight? AI is already changing how doctors work, but the real challenge lies in mastering its use without surrendering clinical responsibility. Key takeaways include the importance of precise prompting to get accurate medical answers, the growing adoption of ambient listening tools (with 30% usage in orthopedics at WashU), and the urgent need for policy updates to keep pace with AI’s rapid evolution. The hosts also highlight that AI is not a replacement for mentorship—when in doubt, surgeons still call trusted colleagues, not chatbots.
Use specific, context-rich prompts like 'You are an orthopedic researcher. Provide evidence-based treatments for carpal tunnel by surgical and non-surgical methods in the tone of a hand surgery specialist.'
Ambient listening tools like Allie have reduced documentation burden, with 30% adoption in orthopedics at WashU, but still require physician review and editing.
AI can automate repetitive tasks like CV updates, grant writing, and email summarization—freeing up time for higher-value clinical work.
AI hallucinations are real and dangerous: one scientist discovered a paper he never wrote had been cited multiple times in the literature.
Physicians remain legally liable for AI-generated content—AI cannot be held responsible for billing errors, coding mistakes, or medical decisions.
…and 3 more takeaways available in PodZeus
Welcome and Introduction to the Episode
Chuck and Chris welcome listeners to The Upper Hand Podcast, introduce their guest Dr. Jason Strelzow, and set the stage for a deep dive into AI in medicine, emphasizing his expertise and the transformative potential of AI in hand surgery and clinical practice.
Defining AI and Large Language Models
Jason explains AI as a predictive system that interprets human language and generates responses based on vast training data. He distinguishes LLMs (Large Language Models) from traditional search engines like Google, emphasizing that LLMs don’t retrieve data—they predict answers based on patterns in language.
The Power of Prompting in AI
The hosts and guest discuss how crafting precise, context-rich prompts dramatically improves AI output. Jason advises treating AI like a toddler—give it clear, concrete instructions, including role, content, tone, and structure, to get better results.
AI Tools: Claude, ChatGPT, and Open Evidence
Jason shares his preference for Claude due to its advanced features like CoWork, which functions as a virtual co-worker. He also highlights Open Evidence as a powerful medical AI interface, with over 50% of U.S. physicians using it.
Ambient Listening: A Game-Changer in Documentation
“It's a game changer, like indescribably so. I don't go home with notes undictated anymore and my notes are pretty good.”
“Claude is not going to jail for Medicare fraud. It's one of us if that were to happen.”
“The systems are not perfect. Many of them, particularly if you're not fascinated with how to use them, will invent or try to satisfy the question that you've asked it.”
“You're still the PI and you're still responsible. Where are we currently right now with that kind of conversation about ethics and liability?”
Hosts
Guest
Dr. Jason Strelzow
person
Chris
person
Chuck
person
Claude
product
ChatGPT
product
Allie
product
Open Evidence
product
PubMed
organization
Journal of Bone and Joint Surgery
organization
Gemini
product
Get the full intelligence
Search transcripts, export clips, track mentions, and explore all topics from “Practical AI in Medicine with Jason Strelzow” inside PodZeus.
Start discovering podcast insights today
Start with a 7-day trial and explore a growing catalog of popular podcasts. No credit card required.
No credit card required • 7-day trial • Cancel anytime
