Teaching residents about #sglt2 turned into a history lesson. Which fruit you all think led to the discovery of sglt2 inhibitors ? #MedTwitter #NephTwitter #nephJC #CardioTwitter #AcademicTwitter #diabetes #heartfailure Thread ⬇️

🤖 AI is reshaping hospital medicine. JHM’s Editor-in-Chief introduces a special issue on AI’s promise + pitfalls in care 🔗: https://t.co/8jJjn6EsBY ✍️:@SamirShahMD @WrayCharlie @Zahirkanjee @AndrewOlsonMD 🖼️: #VisualAbstract by @taybro7 #MoreThanAJournal #ArtificialIntelligence

January 2019-2023 is not the Covid-19 today. These findings are not contextually relevant.

Interesting paper: When any use of AI is mentioned, participants perceived physicians as less competent, trustworthy & empathetic, & were less willing to book an appointment with them Link: https://t.co/RsIcfXCQZa

FDA Approves Polypill Widaplik for Hypertension The U.S. Food and Drug Administration has approved #Widaplik (telmisartan, amlodipine, and indapamide) for the treatment of hypertension in adults #CVPrev #polypill #hypertension

There's a lot to be said on this newest preprint showing that reliance on LLMs in writing tasks leads to a 'consistent underperformance' in neural and behavioral levels with potential long-term consequences. Interestingly, I think this ties to similar but (early)

Presented at #CCR25: A trial involving children with severe acute malnutrition and gastroenteritis showed no evidence of a difference in mortality at 96 hours between oral and intravenous rehydration strategies. Full GASTROSAM trial results: https://t.co/sbUAoVegCJ

Improving the use of GDMT in practice is incredibly challenging Despite hope and prior claims that low-intensity electronic letter nudges to patients or clinicians are effective… No benefit found in this factorial RCT The NUDGE-CKD Trial | Circulation

BREAKING: Apple just proved AI "reasoning" models like Claude, DeepSeek-R1, and o3-mini don't actually reason at all. They just memorize patterns really well. Here's what Apple discovered: (hint: we're not as close to AGI as the hype suggests)

Huge update to our preprint today on the superhuman performance of reasoning models in medical diagnosis! TL;DR – they don't just surpass humans in meaningful benchmarks, but in actual medical care from unstructured clinical data: A 🧵⬇️: https://t.co/7fYxNfebiF

How well do FDA-cleared AI ECG tools do in the real world? The latest issue of @NEJM_AI has an important new study evaluating calibration of an FDA-cleared tool for identification of hypertrophic cardiomyopathy with disappointing real world performance but a pathway for

TAPER-V RCT🔥 14d vanco→taper (125 mg q12h ×7d,then q24h ×7d) vs placebo for initial CDI. Although not stat significant, vancomycin taper showed a trend toward reduced C. difficile recurrence by Day 56. Also reduced need for fidaxomicin, potentially lowering cost #ESCMIDGlobal

It is important to be deliberate and comprehensive in the application of Generative AI to health care. All innovations in health care require balancing new approaches with appropriate guardrails. https://t.co/kRAJa6cTLC

I encourage every doctor to try out MedArena -- a Turing test of AI models on the tasks that YOU think are important!

Low-dose semaglutide reduced alcohol craving and consumption in adults with alcohol use disorder. https://t.co/jatA52boPk

Treatment of gestational diabetes with metformin and additional glyburide, if needed, was not noninferior to insulin in preventing infants from being born large for gestational age. https://t.co/24flAv3iJU

A randomized trial of GPT-4 vs 92 physicians with or without this #AI LLM for performance on patient care tasks. AI improved physician performance, on par with AI alone (based on 5 clinical vignettes) https://t.co/c7b82kQLi8 @NatureMedicine @AdamRodmanMD @jonc101x

Air pollution exposure and #cardiometabolic risk https://t.co/EKFS235vLt #AirPollution #T2D

Redefining obesity. It is time. 📏BMI is NOT a direct measure of fat 📏BMI does NOT establish the distribution of fat 📏BMI cannot determine with excess fat is a health problem @TheLancet Lancet Commision 📎 https://t.co/Wds0FJIbp9

Which revascularization strategy is best for patients with severe LV systolic dysfunction? Read the long-term mortality outcomes after CABG vs. PCI in severe ischaemic cardiomyopathy in #EHJ. https://t.co/wfp1SkSZOr #revascularization #cardiotwitter @ESC_Journals @escardio

The conventional way of classifying metastatic cancers according to their organ of origin is denying people access to drugs that could help them https://t.co/MVI56KfrM6