Steph Cabral
@stephcabral_
PGY-3 @BIDMC_IM | Interests in Pulm/Critical Care, AI, POCUS, and Rock Climbing.
ID:1312839098961465348
04-10-2020 19:36:59
67 Tweets
224 Followers
302 Following
Our new study in JAMA Internal Medicine looking at the reasoning abilities of GPT-4 compared with human physicians just came out.
Big picture: AI displays (much) better reasoning than humans, makes diagnoses similarly, but hallucinates considerably more.
A 🧵to put in context ⬇️
Super excited to share our passion project comparing human and LLM reasoning.
An important finding: GPT generally wrote MORE, didn’t cut corners like humans thus scores could be ⬆️
Adam Rodman Raja-Elie Abdulnour Steph Cabral Byron Crowe Philip Wilson
jamanetwork.com/journals/jamai…
We know LLMs can ace multiple-choice exams. Taking us deeper, an important new study led by Steph Cabral and Adam Rodman conducts a nuanced evaluation of the clinical reasoning abilities of GPT-4 wrt physicians. Guess who wins?
Need more of these!
jamanetwork.com/journals/jamai…
On pulm consults with Harvard Pulmonary & Critical Care Fellowship BIDMC IM Residency, and consulted on a case of platypnea that is still escaping us. In the spirit of a Staci Saunders intern report today on effective med-ed, here's my general approach to the condition and attempt at an infographic!
#MedTwitter
We are absolutely delighted to welcome this amazing class of new fellows to the MGH/BIDMC Pulmonary and Critical Care Medicine Fellowship in July 2024!
MGH Division of Pulmonary & Critical Care Medicine
BIDMC Pulmonary, Critical Care, and Sleep Medicine
We are VERY excited to be kicking off this year's Harvard Medical School Combined Grand Rounds series with A Turing Test for Clinical Reasoning: Large Language Models and the Future of Diagnosis!
(link here: healthcare.partners.org/streaming/Live…)
Medical educators have a crucial role in shaping the inevitable integration of AI into the healthcare.
Adam Rodman and I wrote about it in NEJM ⬇️
nejm.org/doi/full/10.10…