AI Meets MedTech
The Best Podcasts About AI in Healthcare and Medical Technology (And What Sets Them Apart)
The conversation around artificial intelligence in medicine has grown loud. Finding a podcast on the subject now takes about thirty seconds.
Finding one that actually challenges its guests is harder. It interrogates real-world AI deployment and treats clinicians as the primary audience.
A clear-eyed roundup of the best AI-in-healthcare podcasts, comparing depth of clinical scrutiny, and making the case for practitioner-first coverage. It categorizes podcasts by approach and argues why depth of clinical scrutiny matters more than prestige.
Why the Podcast Landscape for Healthcare AI Looks the Way It Does
Most digital health podcasts launched during a period when the dominant question was “will AI change medicine?” The answer is now plainly yes, and the interesting questions have moved on. Practitioners want to know how a radiology AI tool performs on patient populations that were underrepresented in its training data. Physicians want to understand what happens to liability when a clinical decision support system contributes to a misdiagnosis. Ethicists and clinicians alike want someone to press past the product announcement and ask about failure modes.
Podcasts built to answer the first wave of questions are not always well-suited to the second wave. That gap matters, and it is worth naming when comparing shows.
Established Voices in the AI-in-Healthcare Podcast Space
NEJM AI Grand Rounds brings genuine academic rigor to the format. Produced in association with the New England Journal of Medicine, it draws on peer-reviewed research and tends to be methodologically serious. Its strength is precision. Its limitation, at times, is that the institutional frame can soften the harder policy and equity questions in favor of scientific ones.
The Radiology AI Podcast serves a specific clinical community well. Radiologists navigating FDA-cleared AI tools, workflow integration, and imaging-model performance will find relevant, technically grounded content here. The focus is appropriately narrow, which is a feature for its core audience and a limitation for generalist listeners.
AI Rounds positions itself closer to clinical practice and tends to foreground physician perspective on AI tools in active use. It is a useful show for practitioners trying to understand what medical decision support looks like on the ground rather than in a controlled study.
Health Innovation Matters and the Stanford Medicine podcast (in its AI-adjacent episodes) represent the institutional health system perspective. Both are well-produced and draw credible guests. They are strongest when covering research pipelines and weakest when questions of AI liability in healthcare, health equity and AI, or vendor accountability come up. These are topics where institutional caution often shapes what gets said on the record.
What the Best Shows Do Differently
Additionally, shows that endure over time share a few structural habits. They treat clinical AI implementation as a complex, ongoing process rather than a launch event. They seek to understand evolving workflows, not only initial outcomes. They ask guests about what went wrong, not just what worked.
Moreover, they bring in perspectives from outside the vendor ecosystem, including frontline clinicians, bioethicists, regulators, and patients. In this context, A clear-eyed roundup of the best AI-in-healthcare podcasts, comparing depth of clinical scrutiny, and making the case for practitioner-first coverage, can illuminate how practitioners should engage with these narratives.
Production quality matters less than most podcast directories imply. A show recorded in a hospital break room with a practitioner who is willing to say “this tool caused a problem we did not anticipate” is more valuable to a working clinician than a polished studio conversation that stays inside the approved talking points. Candid healthcare conversations are, in practice, rare. Shows that create the conditions for them are doing something genuinely useful.
Healthcare AI ethics, specifically how bias gets introduced, how accountability gets assigned, and how health equity concerns get weighted in deployment decisions, is the topic that most clearly separates shows willing to ask hard questions from shows that prefer comfortable ones.
Where AI Meets MedTech Fits in This Landscape
AI Meets MedTech was built with a specific orientation: artificial intelligence in medicine is worth examining seriously, which means examining it critically. The show takes the clinical perspective as its starting point rather than the technology vendor’s perspective or the investor’s perspective.
The kinds of questions the show pursues include what real-world AI deployment actually costs a health system in terms of integration time, staff retraining, and workflow disruption. They include who bears responsibility when an AI-assisted diagnosis is wrong. They include whether the patient populations most likely to be harmed by a poorly validated model are the same populations that were underrepresented in the training data.
The show brings together an unusually diverse set of voices: founders building AI-powered diagnostics and care platforms, clinicians who left practice to solve problems they lived firsthand, software engineers navigating the gap between what AI can do and what healthcare will accept, attorneys working out who’s liable when algorithms get it wrong, and operators scaling technology across some of the world’s largest health systems.
What sets AI Meets MedTech apart isn’t the guest list- it’s the questions. Every conversation goes beyond the pitch deck into the harder territory: What does it actually take for a physician to trust an algorithm? Who benefits when AI gets it right and who pays when it doesn’t? Is the industry moving fast enough, or are people paying the price while we debate governance frameworks?
Host Anna Nesterova spent 13+ years on the commercial side of MedTech before launching the show. That background shapes every conversation: she’s less interested in how a technology works in a lab, and more interested in who buys it, who trusts it, who gets left behind, and what happens when it gets something wrong.
The show is for anyone with a stake in the future of healthcare, whether you’re building AI tools, investing in health tech, working inside a hospital system, or simply someone who will one day sit across from a doctor whose decisions are shaped by an algorithm. That’s all of us.
How to Choose the Right Podcast for Your Needs
If you are a radiologist wanting to stay current on imaging AI, the Radiology AI Podcast is a logical starting point. If you want peer-reviewed research discussed with academic rigor, NEJM AI Grand Rounds is worth your time. If you want to understand what clinical AI implementation looks like from the perspective of the people doing it, with questions about ethics, equity, and accountability included rather than avoided, AI Meets MedTech is the show that consistently occupies that space.
The field of artificial intelligence in medicine is serious enough to deserve serious examination. The best medtech practitioners working in this space deserve a podcast that treats their professional judgment as the frame for evaluating AI tools, not an afterthought to the technology narrative. That is what AI Meets MedTech is built to be. New episodes are available wherever you listen to podcasts, and you can learn more about the show and its archive by visiting the AI Meets MedTech website.
Women Leading AI in MedTech: Real Implementation, Real Stakes
The conversation about women in AI and medtech is often reduced to a list of names and a round of applause. That framing is both insufficient and counterproductive. The practitioners, researchers, and executives shaping this field deserve to be examined on the merits of their work, the friction they navigate, and the questions their decisions force into the open. This piece is an attempt at that.
The Research Layer: Where the Hard Problems Live
The foundational research shaping clinical AI today owes a significant debt to women who were asking the hard questions before the term “AI in medtech” entered mainstream vocabulary.
Fei-Fei Li, co-director of the Stanford Human-Centered AI Institute, has consistently argued that AI development without diverse input produces tools that encode existing inequities into clinical workflows. Her work at Stanford University AI in medicine is not merely theoretical; it pushes developers and health systems alike to interrogate whose data trained the model and whose population will bear the cost if it fails.
Regina Barzilay at MIT has focused specifically on cancer diagnostics and drug discovery, areas where the gap between a promising model and a clinically deployable tool is wide and consequential. Her work demonstrates that getting a model to perform well on benchmark data is a different problem entirely from getting it to perform reliably in a live clinical environment with real heterogeneity in patient populations and imaging equipment.
Daphne Koller, founder of insitro, brought machine learning to the drug discovery pipeline at a time when the industry was still skeptical that the approach could survive contact with the biological complexity of disease. The practical wager her work represents, that large-scale biological data and modern ML can shorten and de-risk development timelines, is now widely accepted, though the proof points are still accumulating.
Suchi Saria, whose work on predictive models for sepsis detection has been implemented in real hospital environments, represents the translation layer between research and care delivery. Her work through Bayesian Health is a practical case study in what it takes to move a model from a paper into a clinical alert that a nurse will trust and act on.
The Implementation Layer: Where Theory Meets Resistance
Implementation is where ideas encounter institutional inertia, budget constraints, liability concerns, and clinician skepticism. The women working in this layer are solving problems that no benchmark dataset can simulate.
Alicia Chong Rodriguez, co-founder of Bloomer Tech, has focused on wearable cardiac monitoring for women, a population historically underrepresented in cardiovascular research. Her work addresses an equity gap that is both a clinical problem and a data problem: models trained predominantly on male physiology will underperform for women, and the downstream consequences are clinical, not just statistical.
Oriana Papin-Zoghbi, CEO of AOA Dx, is working on early ovarian cancer detection, a diagnostic challenge where late-stage detection remains the norm and the cost of a false negative is severe. The implementation challenge here is not only technical; it involves navigating regulatory pathways, reimbursement structures, and physician adoption simultaneously.
At Sutter Health, practitioners working alongside AI tools are encountering a set of questions that do not have clean answers yet: when an algorithm flags a patient as high risk and the clinician disagrees, who carries the liability? How is that disagreement documented? These are not edge cases. They are daily realities in health systems that have moved past pilot programs into production deployment.
Dominik Demolle, Dornoosh Zonoobi, Jasmine Balloch, Beth McCombs, and Julia Zarb represent a cohort of practitioners and builders who are doing this work across imaging, diagnostics, and care coordination. Their contributions are often less visible in the press, but they are the people making deployment decisions, writing validation protocols, and sitting in the room when a model’s recommendation is questioned by a surgeon.
Meena Mallipeddi and Michelle Wu are among those working at the intersection of clinical AI product development and real-world evidence generation, where the gap between what a model promises and what it delivers in practice gets measured and, where necessary, corrected.
Joanna Shields, whose background spans technology policy and governance, has pushed the conversation toward accountability structures. The question of who is responsible when an AI-assisted diagnosis is wrong is not just a legal question; it is an organizational design question, and it needs to be answered before systems scale.
Quote by Joanne Skiles, PHD, AI and Cloud Advisor:
I If we live in a world where AI stays a little bit of a magic black box, I do not see the adoption of it really going forward, because I think what makes a good doctor is being a skeptic and looking for answers and truth. I do think AI needs to be a little bit more transparent in the decisions it is making. “
The Signal from Nyquist AI and the Broader Pattern
Nyquist AI, working on clinical language models and ambient documentation, represents a category of medtech AI where the product is embedded in the clinical encounter itself, not reviewing data after the fact. The practitioners involved in building and validating these tools are navigating a different kind of trust problem: not just whether the model is accurate, but whether the clinician’s interaction with it changes the nature of the encounter.
The pattern across all of these figures and organizations is consistent. The most substantive contributions are happening where someone is willing to sit with the difficulty of a real implementation problem rather than optimizing for a demo. That requires a specific kind of intellectual honesty about what the technology can and cannot do.
What This Means for the Field
The women named in this piece are not a monolith, and the point is not to celebrate them as a category. The point is that their work surfaces specific, tractable problems that the field needs to reckon with: equity in training data, liability in deployment, trust in clinical adoption, and the gap between benchmark performance and real-world reliability.
AI Meets MedTech exists to go past the press release and into the practical. The field advances when practitioners are asked, on the record, what is actually hard about this. Joanne Skiles, PHD, Software Engineer, AI and Cloud Advisor, the episode title: The High Stakes of AI in Medicine is one example of the conversations the platform is built around.
If you work in this space and want to engage with these questions seriously, the show is the right place to start. Listen, subscribe, and if you have something substantive to contribute, get in touch.
Hello world!
Welcome to WordPress. This is your first post. Edit or delete it, then start writing!