Oncdoc

Dr. Urun This is not really a model competition. It is a constraint system. H&E AI can approximate what assays like Oncotype DX are doing. That is clear. Morphology carries real signal. But oncology decisions are not made on approximation alone. They are made on validated, defensible assays that are embedded in guidelines and can be justified clinically and legally. So the field actually sits across three layers: H&E AI as the efficient, low cost signal layer Oncotype DX as the incumbent, guideline backed decision layer New multi-omic approaches like Caris MI Clarity, which combine pathology with DNA and RNA to provide deeper biological resolution, but are still earning decision level trust The key point is that better prediction alone does not drive adoption. Accountability does. That is why H&E will not immediately replace Oncotype, even if it gets close. And it is why newer approaches like MI Clarity by Caris have to prove not just that they are more informative, but that decisions based on them are safe and defensible. The transition will be layered. Efficiency expands first, but the decision layer only shifts when guidelines and trust follow.

@dvasishtha Get your data right .. HCA expanding residency slots by 7000 and making money exploiting tax payers .. NPs keep switching specialities , stay in same urban areas and drive up the cost of care with unnecessary referrals..

Just a reminder that the healthcare costs that are hitting most people’s pockets are not from independent physicians. It’s from insurers and healthcare systems buying up physicians who can’t make it in today’s economic environment. Physicians are down 35+% in the last decade as far as pay. Insurers are posting record profits.

@VincentRK @charanproducer @PSampathkumarMD @rameshlaus @ThisIsDSP @nramind My late grand mother’s favorite singer. She was in tears when he passed away. May they both rest in peace.

@charanproducer @PSampathkumarMD @rameshlaus @ThisIsDSP @nramind

The Voice that touched a billion people for 5 decades. Listen to a tiny sample of his virtuosity in this medley that I compiled as we remember him on his birthday. We miss him. #SPB #SPBLivesOn

@TiansterZhang @DrRanaMcKay @ASCO @OncoAlert @DrChoueiri @bergsa83 @AarmstrongDuke @Daniel_J_George @ERPlimackMD @neerajaiims @PGrivasMDPhD @montypal @crisbergerot @DrYukselUrun Thanks for posting. Any reason why the combo seems less effective for >20 Mets? Seems like that should be the population that derives more benefit.

@DrRanaMcKay presents #COMRADE trial of olaparib-radium vs radium in mCRPC #prostatecancer. Thought provoking difference between cohorts even unselected population! @ASCO #ASCO25 @OncoAlert

@BertrandTOMBAL @EORTC @Silke_Gillessen @SoaresAndrey @GETUG_Unicancer @cancertrials_ie @sogug1 Thank you.

@ArunAR101 @EORTC @Silke_Gillessen @SoaresAndrey @GETUG_Unicancer @cancertrials_ie @sogug1 annalsofoncology.org/article/S0923-…

In PEACE-3 (EORTC 1333), the addition of 6 cycles of Ra223 to enzalutamide improves PSA response time and rates (≥90%), ALP reduction time (≥30%), and ALP normalization at 6 and 12 months. @EORTC @Silke_Gillessen @SoaresAndrey @GETUG_Unicancer @cancertrials_ie @sogug1

@DrDiGiorgio But won’t the pendulum swing the other way eventually?

The only reason private equity is gaining a foothold in medicine is because hospital systems already rigged the game. Pile on financial pressure (facility fees that favor hospitals, 340B profits hospitals pocket while independent docs get nothing) and you corner physicians into one choice: sell out or quit.

@BrentAWilliams2 Can you educate me why health care spending leveled off? Not being a smart ass I’m genuinely asking.

This is how the physicians in Congress have failed us! Greg has become a congressman who happens to be a physician instead of a physician who happens to be a congressman. We have to live with terrible health policies, laws, and regulations; we don't need your insults as well! @noahkaufmanmd @BrentAWilliams2 @txsportsdoc @notaproviderMD @DrDiGiorgio @DrJMarine @mass_marion

Scientists “ we can convert between artificial numerical representations”. Social media “Plato was right”…

@VixCentral @colin_fraser Of course LLM’s understanding of meaning can mapped and compared to other llm s and be deciphered without going to the source…they are based on the same mathematical and scientific principles…

@colin_fraser Since the evidence was overwhelming for the existence of Universal Grammar this is evidence that LLMs are more similar thinkers to us than we thought. What would be really interesting is to ask an LLM to invent a language that is not compatible with the human UG.

Idk about Plato but total Chomsky vindication

@emollick So wait- computer languages / programs created with mathematical and statistical principles map to the same mathematical and statistical principles no matter what computer language and program is used? Are we sure that’s unexpected?

Huh. Looks like Plato was right. A new paper shows all language models converge on the same "universal geometry" of meaning. Researchers can translate between ANY model's embeddings without seeing the original text. Implications for philosophy and vector databases alike.

@MichaelMindrum He wouldn’t know, he’s a chiropractor

Sometimes I think the American healthcare system would completely collapse if residents didn’t exist as cheap labor. Imagine if residents were paid the same wage as APPs. Most will argue residents are paid trainees but hospitals don’t decrease APPs wages when they’re learning on the job

@sweatystartup You don’t have to play this game with your kids because you’re rich. Didn’t you go to Cornell? Seems like that worked out.

@DouglasYaoDY @ArtirKel You're conflating logical elegance with conceptual depth. Arguably the emergent nature of contingent rules in biology makes it intellectually harder than a field based on axiomatic rules.

@DutchRojas @JeffryGerberMD 25x liability as well….

@JeffryGerberMD If there are 25 of you, 24x7, you own it.

AI Will Clone the Best Physicians, Starting with Primary Care Steve Jobs once said, “A lot of times, people don’t know what they want until you show it to them.” AI in medicine is about to do exactly that, except instead of showing people what they want, it will clone the physicians they trust. AI won’t replace doctors. It will replicate the best ones, making their judgment, decision-making, and clinical intuition available at scale. It won’t just read guidelines or automate paperwork. It will be trained on the best physicians, the ones who see things others miss, who diagnose faster, who treat smarter. And it starts with primary care. Primary care is broken. Not because of bad physicians, but because the system forces them to operate at the speed of billing, not the speed of medicine. AI flips that equation. It gives every patient access to the equivalent of the best primary care doctor, 24/7, without burnout, insurance barriers, or scheduling delays. This isn’t ChatGPT with a stethoscope. This is pattern recognition, clinical reasoning, and real-world expertise, captured, trained, and deployed at scale. It’s a physician’s brain, unchained from the constraints of time. And for independent physicians? AI will be your digital twin. Your expertise, distilled and extended, serving patients while you sleep. Your best version, never overbooked, never rushed, always thinking. This is coming. The only question is: will it be built for physicians or against you? The future of medicine is intelligence, not bureaucracy. It’s scale without sacrifice. It’s AI, trained by the best, making the best available to everyone. The clones are coming. Be ready. #healthcare

@drmoneymatters @drgurner McKinsey has a group of consultants who are working on that report…

@drgurner When will America automate McKinsey?

Well, this is something no one is really preparing for. The estimations might be right, or might be wrong... But either way - shouldn't media be talking about, people reskilling, or at least have any conversations about it? Not a reason to panic, but kind of huge to ignore.

@drmoneymatters the doctor who made the comments sounds frustrated. Just last week I saw a WSJ/Bloomberg article on jobless ivy league MBA…

This 💩 pisses me off. Do we need additional training? sure, but then why don’t MBAs need medical degrees to run healthcare orgs? No one answers that question. Why an MD alone is ‘practically worthless’ for physician leaders today beckersphysicianleadership.com/leadership/why…