Adam Rodman(@AdamRodmanMD) 's Twitter Profileg
Adam Rodman

@AdamRodmanMD

General internist, educator, historian, author, co-director @iMEDEducation @BIDMC_IM, host of #histmed podcast @BedsideRounds @ACPInternists. He/him/his. 🖖🚲

ID:124172052

linkhttp://www.bedsiderounds.org calendar_today18-03-2010 13:58:15

14,0K Tweets

16,8K Followers

1,4K Following

Follow People
Jonathan H Chen MD PhD(@jonc101x) 's Twitter Profile Photo

Lots of hype around ' vs. Doctors', but nobody is asking the more relevant question: What happens if you combine + Doctors?

Thread compiled by lead author Ethan Goh.
Bryan Bunning Elaine Khoong Rob Gallo Arnie Milstein Damon Centola

lnkd.in/eEAwp6rb

account_circle
Isaac Kohane(@zakkohane) 's Twitter Profile Photo

What AI implementors should know about healthcare before trying to 'fix' it zaklab.org/blog/the-medic… Deliberately opinionated. Comments welcome.

What AI implementors should know about healthcare before trying to 'fix' it zaklab.org/blog/the-medic… Deliberately opinionated. Comments welcome.
account_circle
World of Engineering(@engineers_feed) 's Twitter Profile Photo

The term 'patch', meaning a software update, originates from the days when computer code was written on physical cards with holes punched in them. If there was a change in the code, you would 'patch' that section of the card with a piece of tape, covering it, and changing it.

The term 'patch', meaning a software update, originates from the days when computer code was written on physical cards with holes punched in them. If there was a change in the code, you would 'patch' that section of the card with a piece of tape, covering it, and changing it.
account_circle
Ethan Mollick(@emollick) 's Twitter Profile Photo

Before PowerPoint, DuPont had a Chart Room. Made in 1919, it was considered so vital that its existence was secret for 40 years

Charts would zoom towards the board on monorails. The original version was a set of fixed slides that executives in wheelie chairs would be pushed past

Before PowerPoint, DuPont had a Chart Room. Made in 1919, it was considered so vital that its existence was secret for 40 years Charts would zoom towards the board on monorails. The original version was a set of fixed slides that executives in wheelie chairs would be pushed past
account_circle
Journal of Hospital Medicine(@JHospMedicine) 's Twitter Profile Photo

How should medical educators react to recent CMS changes removing review of systems from billing reqs?
Adam Rodman Gurpreet Dhaliwal: An excellent opportunity to refocus on patient-centered history-taking and professionalism.

bit.ly/45UBMw3
: Catie Glatz, MD

How should medical educators react to recent CMS changes removing review of systems from billing reqs? @AdamRodmanMD @Gurpreet2015: An excellent opportunity to refocus on patient-centered history-taking and professionalism. bit.ly/45UBMw3 #VisualAbstract: @CatieGlatz
account_circle
Paul Sax(@PaulSaxMD) 's Twitter Profile Photo

How about a randomized clinical trial? Two hospitals:

Sending out these 'hospital and ER are full, discharge people faster' emails
vs.
Investing resources to make more clinical services/tests available 7 days/week instead of 5

Endpoint: Change in length of stay

account_circle
Geoff Stetson, MD(@GStetsonMD) 's Twitter Profile Photo

HE DID IT!!!

Thanks to the incomparable Greg Ow, our is here!!!

If you have , you can access our virtual mentor - chat.openai.com/g/g-JS1hSZarK-…

If you do not subscribe to , we are working on making it more available. Stay tuned!

HE DID IT!!! Thanks to the incomparable Greg Ow, our #MedEdMENTOR #VirtualMentor is here!!! If you have #ChatGPTplus, you can access our virtual mentor - chat.openai.com/g/g-JS1hSZarK-… If you do not subscribe to #ChatGPTplus, we are working on making it more available. Stay tuned!
account_circle
Arjun (Raj) Manrai(@arjunmanrai) 's Twitter Profile Photo

In our new preprint, we evaluated GPT-4V on 934 challenging NEJM medical image cases and 69 clinicopathological conferences. GPT-4V outperformed human respondents overall and across difficulty levels, skin tones, and image types except radiology, where it matched humans. GPT-4V…

In our new preprint, we evaluated GPT-4V on 934 challenging @NEJM medical image cases and 69 clinicopathological conferences. GPT-4V outperformed human respondents overall and across difficulty levels, skin tones, and image types except radiology, where it matched humans. GPT-4V…
account_circle
Richard Van Noorden(@Richvn) 's Twitter Profile Photo

How large is science's fake-paper problem?
Massive, finds a new ML text-analysis by Adam Day. It suggests paper mills rise to 1.5-2% of all papers; 3% in biomed.
That's 70,000 papers last year alone, >400,000 overall. As a 'conservative estimate'.
nature.com/articles/d4158…

How large is science's fake-paper problem? Massive, finds a new ML text-analysis by @ClearSkiesAdam. It suggests paper mills rise to 1.5-2% of all papers; 3% in biomed. That's 70,000 papers last year alone, >400,000 overall. As a 'conservative estimate'. nature.com/articles/d4158…
account_circle
Physicians' Gallery(@RCPEHeritage) 's Twitter Profile Photo

X-ray by Dawson Turner, 1896. A primitive x-ray installation, the first in Edinburgh, was set up in Turner's house at George Square. As a result of his work he had several of his fingers amputated and lost an eye

X-ray by Dawson Turner, 1896. A primitive x-ray installation, the first in Edinburgh, was set up in Turner's house at George Square. As a result of his work he had several of his fingers amputated and lost an eye
account_circle
David R. MacIver(@DRMacIver) 's Twitter Profile Photo

Huh.

This is a genuine actual output from GPT-4's DALL-E plugin. This is the first time I've seen an image generator quite this pixel-perfect just straight up copy something from its training set.

Huh. This is a genuine actual output from GPT-4's DALL-E plugin. This is the first time I've seen an image generator quite this pixel-perfect just straight up copy something from its training set.
account_circle
Ethan Mollick(@emollick) 's Twitter Profile Photo

A paper that really illustrates both the unexpected power, and unexpected risks, that come from LLMs.

Given text of anonymous posts on Reddit, GPT-4 can infer things like income, gender & location with 85%+ accuracy at 1% of the cost required by humans. arxiv.org/pdf/2310.07298…

A paper that really illustrates both the unexpected power, and unexpected risks, that come from LLMs. Given text of anonymous posts on Reddit, GPT-4 can infer things like income, gender & location with 85%+ accuracy at 1% of the cost required by humans. arxiv.org/pdf/2310.07298…
account_circle
Tom Mullaney(@tsmullaney) 's Twitter Profile Photo

This is a fascinating post, even beyond the reasons one might expect! Jumping right in, the Chinese characters 中國航空公司 appear pixelated, almost like low-resolution bitmap fonts... Cool, huh? But wait there's more... /1 Thread cc yifanxxx

account_circle
Ethan Mollick(@emollick) 's Twitter Profile Photo

Think you can detect AI writing? Experts can't.

This paper had editors from top linguistics journals try to do just that and 'reviewers were largely unsuccessful in identifying AI versus human writing, with an overall positive identification rate of only 38.9%.' HT Anna Mills, annamillsoer.bsky.social, she/her

Think you can detect AI writing? Experts can't. This paper had editors from top linguistics journals try to do just that and 'reviewers were largely unsuccessful in identifying AI versus human writing, with an overall positive identification rate of only 38.9%.' HT @EnglishOER
account_circle