
Sign up to save your podcasts
Or


In this week's podcast episode - I summarize thinking and learning so far on how to evaluate an ai tool in the healthcare space. I talk about using AI to do things that AI does well and humans do poorly.
In this week's Health Data Ethics podcast episode, I talk about healthcare-specific methods of evaluating LLM output - a recent paper on HumanELY, a web tool for evaluating LLM output across five separate axes, is a great lens for thinking about your ai tools. I also discuss a recent evaluation of GPT-4 Vision in which our AI friend ends up with the right answer to a medical case but can't quite tell us why.
In this episode I review The Code Breaker, a biography of Jennifer Doudna by Walter Isaacson. I talk about previous moral panics about new technology (recombinant DNA!) and compare with artificial intelligence today.
In this episode of the Health Data Ethics podcast I dive into a recent paper in Lancet Digital Health about GPT-4 perpetuating racial and gender biases when asked to diagnose patients. After a victory lap, we settle in to talk about what this means for the use of LLMs as a diagnostic tool and what healthcare teams might want to consider if they implement this technology.
In this episode I begin laying out a framework for how to evaluate AI tools in the healthcare environment, starting with short vs long term impact, the value of a Super Bowl commercial, and figuring out tasks that AIs are good at.
in this episode I talk about my day job as an IT Business Relationship Manager, and give a hypothetical example of a project that benefits from having a person to speak both "clinician" and "IT" languages.
I'm back with an episode in which I discuss patient attitudes towards human scribes, revisit some 2018 predictions about digital scribes, and make a new recommendation when your healthcare IT group pilots a new digital documentation tool: Publish about it!
In this episode I talk about "objective" data and training AI from a patient perspective as well from a healthcare data perspective.
In my podcast episode this week I talk about a recent article on "skill rot", misattributed Cicero quotes, the total amount of skill being conserved, and more.
In this episode I get a little teary about the Mars Rover while talking through some thoughts on what AI does best and what humans do best.
From the publisher's feed
Health tech conversations, from a healthcare IT professional. We're going to talk about medical innovation, technology, and the ethical and operational considerations for health systems. In other…