Cardiology Fellow @uw_cvm 🫀@hERGologie lab 🔬Physician-scientist studying ion channels in arrhythmia 🤓
Madison, WI
Joined October 2018
- Tweets5.2K
- Following9.2K
- Followers8.4K
- Likes30.3K
Michael Rockman, MD, PhD retweeted
I started building an #LBBAP simulator for my own learning with @claudeai and got a bit carried away.
Free to use: pacingsimulator.com
Inspired by Intro2echo.com and Angioview.com. Check it out. All feedback welcome!
#EPeeps #LBBAP #cardiotwitter
Michael Rockman, MD, PhD retweeted
PVC localization has been one of the harder things to learn and teach as a cardiology fellow interested in EP.
So I built learnpvcs.com to help connect the 12-lead ECG with 3D cardiac anatomy.
You can explore different PVC origins, rotate and look inside the heart, and work through practice cases.
It's an early educational prototype, and I'd love feedback from fellows, electrophysiologists, and educators.
Which PVC origin do you find hardest to localize?
#CardioTwitter #EPeeps #MedEd @OpenAI @WvuCvFellows @wvudeptofmed
Michael Rockman, MD, PhD retweeted
OpenEvidence, as a physician-facing LLM, is being used by more healthcare professionals month by month. Now, a study analyzes how almost a million US clinicians use it while submitting 271 million AI-enabled clinical queries!
"Early-career physicians (use ratio, 2.54) and those in oncology and hematology (2.00), internal medicine (1.88), and family medicine (1.80) submitted a higher-than-average share of queries; treatment and management accounted for 33.1% of physician queries and diagnosis for 13.9%."
I really liked their concluding remark:
"For much of modern medicine, clinicians have been expected to internalize a body of knowledge through training and maintain it through continuing education. Study findings suggest a growing share of knowledge access is shifting from recall to retrieval."
We cannot expect physicians to know everything, but we can expect them to know where to find the information they need.
The study: jamanetwork.com/journals/jam…
Michael Rockman, MD, PhD retweeted
The website is live at intro2echo.com and the github is at github.com/douyang/echodemo
It was really fun to make the free movement - the video doesn't do it justice, but you can free move both the TTE and TEE probe and it'll generate the 2D image in according to the 3D model.
There's a couple of chapters on common echo concepts and common diseases. Would love feedback!
I'm almost ready to share it, but Claude Code is very good at keeping consistency across views and chapters.
For example, I can specify a heart phenotype, and will make all the images consistent and physiologically plausible.
Even with unlimited time, I don't think I could create such a nice tutorial!
Michael Rockman, MD, PhD retweeted
Watching a whole lot of my Eagles timeline have a whole lot of very strong opinions about a medical condition they’d never heard of until today đź«
Superficial venous thrombosis is not DVT. It’s much lower risk and usually doesn’t require specific treatment. It’s also generally…
Michael Rockman, MD, PhD retweeted
I find that AI assisted science has deeply increased my joy and curiosity, and rather than using this new ability to pump out slop I aspire to publish much less, of a much higher quality. Maybe others will feel the same. 🤷‍♂️
AI is causing academics to have midlife crises
“Should we publish papers?”
I do think there’s a path forward once AI use in research becomes common (soon). Values will shift away from authoring many publications and toward taste/substance of insight, curation, and communication/education. Reviewing and editorial work will be better recognized.
Michael Rockman, MD, PhD retweeted
Who’s using it?
➡️ Early-career physicians: 24% of the workforce, but 62% of physician queries
➡️ NPs & PAs were the most intensive users (~45 queries/month vs 37 for physicians)
➡️ IM, FM & onc/heme used it at ~2x their workforce share
Michael Rockman, MD, PhD retweeted
I feel horrible for the real Dr. Salman, who has this scam artist impersonating her.
Impersonating a real physician with impressive credentials to take advantage of vulnerable patients to sell snake oil AND tell patients to stop their chemotherapy/radiation is diabolical.
unavailable
Michael Rockman, MD, PhD retweeted
As an Editor, in past ~6mo, we are receiving an avalanche of unsolicited review articles (“scoping” “systematic”, narrative, NOS) on topics and ?s NO ONE EVER has asked and clearly with heavy AI lift. Even the cover letters all read the same…alarming and sad. #publishing
đź’ŻThe benefit of an early lab experience is the experience not a publication nor should it be. During my first lab experience, I was jokingly told:
"You could be included in the acknowledgement that this work was completed ❗️in spite of ❗️the assistance of Michael"
Michael Rockman, MD, PhD retweeted
Every time, these kind of "papers" are produced by kids whose parents have access to a lab and are shameless enough to give them publication credit. When I worked in my dad's lab, he was explicitly clear that I would be a footnote at best. All of those "science fair winners" just cribbed someone else's research. It's a huge, huge problem that likely gets solved by the fact that nobody will care about classic college application-maxing within a decade
This quoted post is unavailable.
Michael Rockman, MD, PhD retweeted
Women make up a larger share of programs’ top quartile ranks in the Match than their lower 75% in most specialties (except ObGyn). The difference in medians is widest in fields where women are underrepresented like rad onc (+17.1%), DR (+14%), and IR (+13.8).
Michael Rockman, MD, PhD retweeted
Doing basic science is always a great bet. Now that AI is going to try everything that smells useful, it’s an even better bet (especially in biology, where testing an idea means meeting the real world, and AI can’t just try everything)
Michael Rockman, MD, PhD retweeted
Princeton didn’t consider Anne Carson worthy of a recommendation for tenure, and she couldn’t continue teaching at McGill after its Department of Classics was shut down.
Now, everyone wants to claim her as their own …
