Chris Williams
@cykwilliams
Followers
84
Following
119
Media
1
Statuses
39
Postdoc @UCSF_BCHSI Butte lab. MD alum @Cambridge_Uni. Previously AFP @CUH_NHS.
San Francisco
Joined January 2021
Such a pleasure to be live on the @abc7newsbayarea evening news discussing how AI might one day be used in #ER triage! Check out the replay here if you missed it (from 1.55): https://t.co/R389b2SPrR
@UCSFHospitals @UCSF_DOCIT @atulbutte
abc7news.com
The Bay Area's source for breaking news and live streaming video online. Covering San Francisco, Oakland and San Jose and all of the greater Bay Area.
0
7
15
Thrilled to announce our latest paper 'Physician- and Large Language Model–Generated Hospital Discharge Summaries', out today in @JAMAInternalMed!
LLM-generated discharge summaries were of comparable quality to those generated by physicians, though they contained more errors, and both types had low overall harmfulness scores. https://t.co/uhSRZsrCVw
0
3
6
@cykwilliams presenting our results on use of GPT to find out why patients are not getting follow up colonoscopies after a positive FIT test #quality #AI with @UrmimalaSarkar @j_r_a_m funded by The Doctors Company @UCSF_DOCIT
Join @cykwilliams at 4:45 p.m. as he presents the use of GPT-4 to determine reasons for missed follow-up colonoscopies. Learn about its high accuracy and the potential for LLMs to drive healthcare quality improvements. Work supported by the #ADAPT center: https://t.co/4VBjEnpLpW
0
3
6
Standing room only and multiple hallway spillovers in these LLM talks! #AMIA2024 @AMIAinformatics with talks on #Gatortron, generating DC summaries @cykwilliams @UCSFHospitals, HTN NLP @JianchengYe, extracting #SDOH by Tim @IAIM_NU @yuanhypnosluo 🎉
0
3
13
Excited to be at #AMIA2024 in SF this week, giving two talks (both today!): ➡️ Evaluating Large Language Models for Drafting ED Discharge Summaries ➡️ Utilizing GPT-4 to determine reasons for missed follow-up colonoscopy following abnormal non-invasive colorectal cancer screening
0
4
11
#AI is helpful in an ER setting but shouldn't be blindly trusted, UCSF’s Dr. @cykwilliams tells @kron4news. It can help answer exam questions & draft notes—"but it’s not currently designed for situations that call for multiple considerations," he explains. https://t.co/plGZu3nQEr
0
3
3
1
0
1
Thrilled to announce our latest paper 'Evaluating the use of large language models to provide clinical recommendations in the Emergency Department' published today in Nature Communications! https://t.co/1MmGIoD1t6
nature.com
Nature Communications - The emergence of large language models has the potential to transform healthcare. Here, the authors show that, when providing clinical recommendations, these models perform...
1
8
17
Excited to announce our latest paper "Enhancing emergency department charting: Using Generative Pre-trained Transformer-4 (GPT-4) to identify laceration repairs"! Congrats to @karanbains and the rest of the team! @atulbutte @AaronKornblith
https://t.co/yXyum9xjng
onlinelibrary.wiley.com
Click on the article title to read more.
0
4
13
UCSF's Town Hall this Friday, 7/12 will include Bakar postdoc @cykwilliams "How AI Can Help Emergency Department Prioritization" - new time at 12-1pm. Tune in! https://t.co/Lzha0DS7tu His recent work: https://t.co/0UEza1dTlf
0
6
7
Excited to announce our latest paper "Application of the Sepsis-3 criteria to describe sepsis epidemiology in the Amsterdam UMCdb intensive care dataset" published today! @tedinburgh_ @drPaulElbers @AriErcole
https://t.co/Jyph6fTblp
0
2
7
Thanks to @KCBSRadio for having me on air!
#AI can analyze and categorize a patient’s symptoms—possibly helping w/ future patient triage in the ER. Out of 2 patients, a study found the model can identify which condition was more serious 89% of the time, UCSF’s Christopher Williams says. 🔊 :
0
0
7
Excited to be at @AMIAinformatics #CIC24 in Minneapolis this week alongside an incredible @UCSFHospitals @UCSF_DOCIT team, giving a talk on how #LLMs can be used to identify barriers to optimal patient care #qualityimprovement Join us at 9.30 am CT Wednesday in Marquette 2&3!
0
2
12
#AI can help triage patients in emergency departments, a study in @JAMANetworkOpen finds. But lead author Dr. Christopher Williams cautions: “First we need to know if it works & understand how it works, & then be careful & deliberate in how it is applied."
ucsf.edu
With further validation and clinical trials, the use of artificial intelligence in emergency departments could one day help prioritize patients based on the urgency of their treatment, and help with...
0
3
6
In one of the first studies to test whether AI can help triage real-world ER patients, new UCSF research suggests AI could one day help doctors make one of the most critical decisions in medicine: who to give urgent medical care to first https://t.co/I83Q6MlhvB via @sfchronicle
sfchronicle.com
UCSF research suggests AI could one day help doctors make one of the most critical...
1
2
3
@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
Tune into @abc7newsbayarea live at 5.30 pm PT tonight where I'll be discussing our findings https://t.co/sdTqJumisW
@atulbutte
abc7news.com
Watch live streaming video on ABC7news.com and stay up-to-date with the latest KGO news broadcasts as well as live breaking news whenever it happens.
Thrilled to announce our latest paper "Use of a Large Language Model to Assess Clinical Acuity of Adults in the Emergency Department" published today in @JAMANetworkOpen @TravisZack2 @bmeow19 @madhumitasushil Michelle Wang @AaronKornblith @atulbutte
0
1
13
Out today in @JAMANetworkOpen! Cross-sectional study w/ 10 000 pairs of ED visits, @OpenAI GPT accurately identified the patient w/ higher acuity when given pairs of deidentified presenting histories. Work by @cykwilliams @TravisZack2 @bmeow19 @madhumitasushil Michelle Wang
LLMs may be able to accurately identify higher acuity patient presentation when given pairs of presenting histories extracted from patients’ first ED documentation. https://t.co/Yl4cSNoBBH
@atulbutte
4
24
78
The potential for LLM #AI to do accurate triage in the emergency department for patient acuity @JAMANetworkOpen @atulbutte
https://t.co/y0YiOBXLB2
11
51
164