Gary Weissman
@garyweissman
Followers
1K
Following
9K
Media
80
Statuses
2K
clinical informatics, AI/ML, critical care and pulmonary medicine, health policy, health services research; husband and dad, human. @[email protected]
Philadelphia, PA
Joined June 2010
I've been fortunate to review a number of papers on clinical prediction models this past year and I've noticed a number of frequent errors. This thread is mostly for the clinician-investigator considering building and publishing a predictive model of some kind. 3 major points. /1
13
208
495
Primary care provider preferences regarding artificial intelligence in point-of-care cancer screening ft. Vinayak Ahluwalia (@Penn_MSHP), @marilyn_schap (@PennDGIM), @garyweissman (@PennPulmonary) & @ravi_b_parikh (@EmoryUniversity) @PennLDI
https://t.co/uIv9pJETno
1
2
3
#AI for #cancer #screening are proliferating. But does it align with what #PCPs find useful? We explore in our new @HAClab_ paper in @socmdm @MedDecMak Policy & Practice led by Vinny Ahluwalia w/ @marilyn_schap @garyweissman -->
1
4
7
A major challenge in AI health care regulation is balancing patient safety with innovation. LDI Senior Fellows @daniel_s_herman and @garyweissman propose adapting CLIA, a long-standing clinical lab oversight program, to regulate AI. Learn more:
ldi.upenn.edu
Artificial intelligence – already a booming sector in health care — will need more than just smart innovation to succeed.
2
1
1
CLIF makes its print debut in @yourICM with @_jcrojas, @WF_Parker, @Nick_Ingraham, @ChadHoc, @KaveriChhikara, @CatGaoHow, @bhavani_md, @Vaishvik_, @garyweissman, @kevingbuell, @XSusanHan, @jaykoyner, @yuanhypnosluo, and many other friends whose handles I don't know (link next):
1
9
15
Looking for a clinical research coordinator job? Interested in understanding if clinical AI systems are effective, safe, and equitable? Come work with our lab to deploy and evaluate AI decision support systems in real clinical practice! https://t.co/quezHipw4X
1
3
8
Very grateful to @pennaitech for the support, and to co-authors @LauraZwaan81 and Sigall Bell from whom I learned a ton in writing this!
A new paper by our Year 1 Awardee @garyweissman and his team" "Diagnostic scope: the AI can’t see what the mind doesn’t know" - Read on https://t.co/n0JUU0vKwQ
@degruyter_brill
0
0
3
Very excited to work with @RajeevAlur @AI4Code @RICEric22 @garyweissman @ravi_b_parikh @SameedKhatana Deo Rajat and Payal Shah to advance trustworthy AI for clinical decision support in this @ARPA_H project.
Penn researchers have received a $7-million, four-year award from the Advanced Research Projects Agency for Health (@ARPA_H) to develop AI systems that support personalized medical treatment of #BreastCancer & other diseases. @PennEngineers @PennMedicine
0
5
21
Great talk by @garyweissman at #AMIA2024 @AMIAinformatics - the time where broad spectrum abx should be started is not the same time as identified by sepsis labels!
0
1
13
Race & ethnicity & clinician linguistic expressions of doubt in hospital admission notes ft. @courtneley, Xinwei Chen (@PennDGIM), @jayaaysola (@PennCHEA), Eden Addisu (@PennLDI), Ari Klein, @davy_weiss (@UPennDBEI) & @garyweissman (@PennPulmonary) https://t.co/bciH9ifviw
0
4
3
For more on this, join us on October 10 for a day-long conference on GenAI in health care:
ldi.upenn.edu
Rapidly emerging applications of large language models can improve health care decision making–and outcomes–for providers and for patients.
1
2
3
🚨 Just released! 🚨 Excited to share our new preprint on the Common Longitudinal ICU Data Format (CLIF)! This open-source format standardizes ICU data for federated research, advancing AI and critical care research. Check it out: ( https://t.co/f9c2LEpLQX)
#ICU #DataScience #AI
medrxiv.org
Background Critical illness, or acute organ failure requiring life support, threatens over five million American lives annually. Electronic health record (EHR) data are a source of granular informa...
0
7
18
Many thanks to the authors including @garyweissman of @PennLDI who provides further context and links to a github page for code and data on this excellent thread (3/3) 👇 https://t.co/qiYo0IfgSE
Everyone is excited about AI in healthcare. But is it safe? Is it effective? Is it equitable? Well, it's hard to say. Heres the @PennLDI blog post trying to answer these questions about FDA authorized clinical decision support devices in critical care https://t.co/dTBAC2ncGu 🧵
0
2
3
An editorial makes some specific recomendations including that the "FDA should require rigorous preapproval studies of validity, safety, and efficacy, as well as postmarketing surveillance of clinical utility" (2/3) 👉 https://t.co/3nwxeeDBgG
1
4
6
“Of 521 authorizations in the FDA AI/ML database, we identified 10 that might inform care for patients with critical illness, and 𝐧𝐨 𝐬𝐭𝐮𝐝𝐢𝐞𝐬 𝐞𝐱𝐚𝐦𝐢𝐧𝐢𝐧𝐠 𝐭𝐡𝐞 𝐜𝐥𝐢𝐧𝐢𝐜𝐚𝐥 𝐢𝐦𝐩𝐚𝐜𝐭 for these device authorizations.” (1/3) 👉 https://t.co/uSgYmdzOhE
3
18
27
📢 ANNOUNCEMENT | @theNAMedicine has selected 11 Scholars in Diagnostic Excellence in partnership with @CMSSmed for its 2024-2025 cohort. More: https://t.co/z3S8vraUwD
@carlberdahl @atgcbag @DrAGonzMD @batzionH @NataliaKhalaf @m_e_nielsen @vschaye @ShaziaMSiddique @garyweissman
0
4
5
We are looking for postdocs to collaborate with Penn Medicine faculty for an exciting new project on Trustworthy AI for clinical decision making
1
12
27
Exciting opportunity for postdoctoral candidates interested in pursuing a career in AI for Medicine! You will get to work with a multi-disciplinary team including myself, @RajeevAlur, @RICEric22, @ravi_b_parikh, @DrQiLong, @SameedKhatana, @garyweissman, and many others.
We are looking for postdocs to collaborate with Penn Medicine faculty for an exciting new project on Trustworthy AI for clinical decision making
0
5
12
@EricTopol @JAMANetworkOpen @atulbutte We commend @cykwilliams @atulbutte and team on this new work. In this @JAMANetworkOpen editorial we offer some thoughts on the study's promising findings and important caveats for advancing this important area of work @AriBFriedman @garyweissman
jamanetwork.com
Emergency departments (EDs) are experiencing historic crowding. A lack of patient guidance for where to seek care for acute illness further exacerbates ED strain. Consequently, patients often arrive...
0
5
6
Important message below but infuriating #AI field has to 'discover' this rather than being guided by long-standing stats literature eg, https://t.co/Z6gwMhdeVO by @GSCollins says: 'Categorising continuous predictors is ... inefficient & should not be used in model development'
onlinelibrary.wiley.com
Continuous predictors are routinely encountered when developing a prognostic model. Investigators, who are often non-statisticians, must decide how to handle continuous predictors in their models....
AI models perform better when trained on underlying measurements rather than the associated diagnoses, because dichotomizing data loses information that is useful in training AI models. Read the Case Study by @amey_vrudhula, BS, et al.: https://t.co/2ItDoArgtG
12
55
226
The elephant in the room as we deploy AI in clinical settings: The Limits of Clinician Vigilance as an AI Safety Bulwark https://t.co/sasZxjXytD Thanks to my coauthors @Bob_Wachter and Don Redelmeier @UCSF @UCSF_DOCIT @doctorscompany @MooreFound
jamanetwork.com
This Viewpoint examines the potential problems of clinician reliance on the use of artificial intelligence (AI) in health care and offers suggestions on how AI could be designed to promote clinician...
0
20
54