Torn ACL confirmed. ACL reconstruction ✅ Get well soon champ. See you in 9-12 months.

Ulberg undergoes surgery within week of injury rules out minor injuries. No official announcement yet but… valgus load and an appearance on video of knee/leg snapping or shifting after planting on it is a common textbook mechanism for ACL tear. The valgus moment can also stress the menisci and MCL so a combined inj... Show more

RFK jr expects the Food and Drug Administration to change the status of "about 14" peptides. Some on the list have orthopedic implications such as the touted BPC157 and TB500. Taken together sometimes called the “Wolverine Stack”. Has anyone tried using these peptides and if so what were your results?

It’s time… #ortuesdays Late 60yo female. 2 years of shoulder pain… Failed attempts at conservative PT, NSAIDS, injections… MRI small full rotator cuff tear, partial proximal biceps tear, anterior labrum tear… Implants ✅ OR ready ✅

Finished some additional work on OrthoIQ platform over the weekend. Sharing some more screenshots. 👇 Shown here a patella pain consult: Mid-consultation, the Research Agent pulls live PubMed citations relevant to your case (work in progress). No potential for hallucinated references though. Actual indexed literat... Show more

This is what a multidisciplinary AI consult on the platform actually looks like. Five specialists. Independent assessments. Per-agent confidence indicators. Not a chatbot response. A structured brief. [High/Medium/Low confidence bands] OrthoIQ. orthoiq.io

OrthoIQ is work in progress and here are the latest updates. I’ve posted some recently about limitations in AI healthcare. One voice. No accountability. No outcomes tracking. I’ve been slowly working on building an answer, hopefully. ➡️ OrthoIQ. I’ve noticed as a practicing orthopedic surgeon, the ever-increasing ... Show more

Does a decaf espresso shot ever make sense? Witnessed at my local morning coffee haunt and just trying to understand…🧐🤔

A system that can’t measure whether it helped you can’t improve. It can only get more confident in being wrong. The bar for “good enough” in consumer health AI is way too low right now.

One of the most frustrating things about AI health tools right now: Who is tracking whether they actually work? With an increasing number of users turning to LLMs for health questions there is no real follow up. There is the thumbs up 👍/down 👎 a user can click after a ChatGPT response but… not sure what that gets... Show more

Prediction markets force participants to put stakes on outcomes. The result is that bad forecasters lose and good ones win. Over time, the system gets more accurate because accuracy is incentivized. Prediction markets in finance. ✅ Prediction markets in sports. ✅ What would it look like if AI medical agents had to be... Show more

ChatGPT health is one voice. One model. One perspective with no stakes. That’s not how good medicine works.

Here’s what ChatGPT can’t tell you: Whether it’s right. It has no mechanism to be accountable for the answer. It gives you the same confidence whether it’s 90% sure or 40% sure. And tomorrow it won’t remember if it was wrong.