Curt Langlotz (@curtlanglotz) / X

Curt Langlotz

4,191 posts

Curt Langlotz

@curtlanglotz

Director @StanfordAIMI, senior fellow @stanfordHAI, radiologist, machine learning geek, @RSNA president, @StanfordDBDS alum, author bit.ly/radreportbook.

Menlo Park, CA

profiles.stanford.edu/curtis-langlotz

Joined February 2009

Pinned
Curt Langlotz
@curtlanglotz
Feb 8, 2017
Will #AI replace radiologists? The answer is NO. But rads who use #AI will replace rads who don't @RSNAInformatics @SIIM_Tweets
Curt Langlotz
@curtlanglotz
Jul 20, 2024
Truth: "State of the art LLMs don't accurately diagnose patients (performing worse than physicians), don't follow guidelines, cannot interpret lab results, fail to follow instructions, and are sensitive to both the quantity and order of information." nature.com/articles/s4159…
42K
Curt Langlotz
@curtlanglotz
Jul 1, 2021
Rather than expect perfection from AI, we should assess how AI performs relative to humans, and, even better, how human+AI performs compared to humans alone.
Curt Langlotz
@curtlanglotz
Oct 10, 2021
FYI, @US_FDA cleared AI/ML-Enabled Medical Devices by specialty: Radiology 241 Cardiovascular 41 Hematology 13 Neurology 12 Ophthalmic 6 Chemistry 5 Surgery 5 Microbiology 5 Anesthesia 4 GI-Urology 4 Hospital 3 Dental 1 Ob/Gyn 1 Orthopedic 1 Pathology 1
AI-Enabled Medical Devices
From fda.gov
Curt Langlotz
@curtlanglotz
Oct 24, 2023
Hot off the press, part of @radiology_rsna journal's centennial: The Future of AI and Informatics in Radiology: 10 Predictions pubs.rsna.org/eprint/U4XBY3E…
27K
Most of what we do has no proven downstream outcome effects. IMHO the main reasons the radiology AI companies have not met investor expectations are: 1. high quality clinical training data is a lot harder to come by than radiology AI startups anticipated.
This Post is from an account that no longer exists. Learn more
Curt Langlotz
@curtlanglotz
Oct 5, 2020
Hot off the press: We have developed a self-supervised learning method that is much better for pre-training than ImageNet--reduces labeling needs by an order of magnitude for medical imaging applications: @StanfordAIMI @Radiology_AI
Yuhao Zhang
@yuhaozhangx
Oct 5, 2020
👋 Excited to share our latest work "Contrastive Learning of Medical Visual Representations from Paired Images and Text". We propose a contrastive framework for learning visual representations of medical images from paired textual data. arXiv: arxiv.org/abs/2010.00747 👇 (1/7)
Curt Langlotz
@curtlanglotz
Nov 1, 2019
Hearing about layoffs at some prominent radiology AI companies. Suggests the AI bubble may be deflating...
Some thoughts on the industry approach to AI in radiology (to date). A thread: @stanfordAIMI @StanfordHAI
Curt Langlotz
@curtlanglotz
Aug 1, 2020
Just got my hard copy of @jeremyphoward @GuggerSylvain book today & can’t wait to crack it. (I sat in on the @fastai course earlier this year, but a great way to review.) And I am delighted to share the “blurb box” with the quintessential AI author @NorvigPeter @StanfordAIMI
Curt Langlotz
@curtlanglotz
Jul 19, 2020
I am heartbroken to say goodbye to one of the best individuals I have ever known. He was a brilliant man, not just about science but about people. A wonderful warm and caring person who truly loved those who worked for him and with him. May he rest in peace.
Curt Langlotz
@curtlanglotz
May 15, 2019
My attempt to provide scientific support for the case I have been making in various venues over the last couple of years that AI won’t replace radiologists:
Radiology: Artificial Intelligence
@Radiology_AI
May 15, 2019
"The question of whether Machines Can Think is about as relevant as the question of whether Submarines Can Swim." – Edsger Dijkstra ow.ly/pwcU30oK716 @curtlanglotz @StanfordAIMI #AI #MachineLearning #radiology
Confirming what many believed about deep learning image reconstruction: "Deep learning typically yields unstable methods for image reconstruction from: 1) tiny perturbations, 2) structural changes, and 3) changes in the number of samples." @Radiology_AI pnas.org/content/early/…
Curt Langlotz
@curtlanglotz
Jun 23, 2021
Are you a doc or other health worker who wants to get involved with AI? You may be asking: --What do I need to learn? --Where should I start? --What online courses are available? I looked into these questions during a recent sabbatical: towardsdatascience.com/getting-starte… @Radiology_AI
towardsdatascience.com
Getting Started in Medical AI | Towards Data Science
A Guide to Online Learning Resources for Clinicians