UW-Madison Critical Care Medicine Data Science Lab
@UW_ICU_DataSci
Followers
195
Following
41
Media
8
Statuses
144
Drs. Matthew Churpek, Anoop Mayampurath, and Majid Afshar lead our integrated data science lab, working to improve the care of hospitalized patients.
Madison, WI
Joined September 2020
Congrats to Dr. Churpek and Ms. Spicer on their NEJM pub in collaboration with VUMC & @PCCRG! In a secondary analysis, they used causal ML to predict the individualized effects of ketamine vs etomidate and found no significant HTE.
nejm.org
For critically ill adults undergoing tracheal intubation, observational studies suggest that the use of etomidate to induce anesthesia may increase the risk of death. Whether the use of ketamine ra...
0
0
3
Excited to share our @yourICM publication in collaboration with ANZICS @anzicsctg, showing heterogeneity of treatment effect for early mobilization on mortality. @UWMedicine
link.springer.com
Intensive Care Medicine - Benefit or harm from early mobilisation (EM) in mechanically ventilated patients may vary by individual patient characteristics. We used machine learning to predict...
0
0
2
0
2
4
Today Clinical Implementation of AI-Based Screening for Risk for Opioid Use Disorder in Hospitalized Adults by @Majeans2011 @UW_ICU_DataSci was highlighted as a most impactful paper of 2025 at #AMIA2025 ! https://t.co/p2GkIXTugv
nature.com
Nature Medicine - In a quasi-experimental, pre–post-implementation study involving 51,760 hospitalizations, an electronic health record-based risk prediction model informed clinicians of...
0
4
6
We are excited to share our latest publication, 'Multivariate protein landscape of host response in hospitalised patients with suspected infection in the emergency department,' where we used 29 plasma proteins to map the multivariate host-response to infection. We discovered
nature.com
Nature Communications - Authors measure plasma proteins in a cohort of hospitalised patients presenting to the emergency department with suspected infection, revealing six discrete host response...
0
0
3
0
0
1
Our Nature article "Current and future state of evaluation of LLMs for medical summarization tasks" https://t.co/5xClpaykV4 was spotlighted at @HeyEpic’s UGM 2025 Executive Address! 👏📷Shoutouts to @Majeans2011 of @uw_medicine Dr. Patterson @UWEmerMed, Emma Croxford UW BMI
nature.com
npj Health Systems - Current and future state of evaluation of large language models for medical summarization tasks
1
3
4
Sam Nycklemoe, mentor Dr. Anoop Mayampurath, team published in @AMIAinformatics on our pCART Explainer, a novel algorithm that highlights text within clinical notes to provide medically relevant context about deterioration alerts, improving explainability of the pCART model.
pubmed.ncbi.nlm.nih.gov
We developed pCART Explainer, a novel algorithm that highlights text within clinical notes to provide medically relevant context about deterioration alerts, thereby improving the explainability of...
1
1
6
Come join us in the land of cheese and Epic! Applications are open now for our Clinical Informatics fellowship - work side by side with our team of >20 board certified Informaticists on a range of projects from AI to decision support. https://t.co/1S5MPwkDxe
medicine.wisc.edu
0
1
1
Grad Charlie Kotula, mentor Matt Churpek, & UW+ Loyola+UChicago team developed & compared novel multimodal deep learning models for early detection of deterioration in ward patients, illuminating potential utility of integrating clinical notes in deterioration prediction
jmir.org
Background: Implementing machine learning models to identify clinical deterioration on the wards is associated with decreased morbidity and mortality. However, these models have high false positive...
0
1
2
Grad Sierra Strutz, mentor Anoop Mayampurath, UW + Loyola + UChicago +Northshore team developed a novel hospitalwide XGB model for the early detection of deterioration in children, thereby enabling a unified risk assessment throughout their hospital stays!
pubmed.ncbi.nlm.nih.gov
This retrospective cohort study describes the development of a novel hospitalwide model for continuously predicting the risk of critical events through the entirety of a child's stay. The model...
0
1
1
Pushing the boundaries of AI in medicine — from bedside implementation to continuous evaluation! #XGM2025 #AIinHealthcare #PDSQI9 #LLM #HealthIT
@Majeans2011 @UWInformatics @UWMadison @uw_medicine @UWEmerMed (4/4)
0
0
1
Even better? It’s open-source and automated — so health systems can scale trustworthy evaluation using an LLM-as-a-Judge. 🤖📷 https://t.co/UHiZpqwPN4
@Majeans2011 @UWInformatics @UWMadison
@uw_medicine @UWEmerMed (3/4)
github.com
Contribute to epic-open-source/evaluation-instruments development by creating an account on GitHub.
0
0
0
In less than a year, they’ve published three papers to address a critical gap in evaluating LLMs in healthcare - introducing the PDSQI-9, a validated instrument for LLM summarization. @Majeans2011 @UWInformatics @UWMadison
@uw_medicine @UWEmerMed
https://t.co/xklmgtwrSj (2/4)
git.doit.wisc.edu
GitLab Enterprise Edition
0
0
0
Big shoutout to Dr. Majid Afshar @Majeans2011 of @uw_medicine, Dr. Brian Patterson of @UWEmerMed, and PhD student Emma Croxford of UW BMI for representing @UWMadison at @HeyEpic’s XGM in front of hundreds at the Physician Advisory Council! 👏🎆 @UWInformatics (1/4)
3
2
7