I’ll be presenting a #ML4H2023 Findings paper I worked on with Thomas Kannampallil, PhD and Seunghwan Kim done @WashUi2db tomorrow. Come visit me at poster session 2 if you’re interested in tabular LMs for Epic EHR audit logs! huggingface.co/papers/2311.06…

New publication out! We used mobile phone data to explore changes in consumption to unhealthy food due to COVID-19: Using Mobile Phone Data to Assess Socio-Economic Disparities in Unhealthy Food Reliance during the COVID-19 Pandemic | Health Data Science spj.science.org/doi/10.34133/h…

just setting up my twttr 👋 Interested in the frontier of #AI and #healthcare? Follow us to keep up with the groundbreaking research here at the @WUSTL Al for Health Institute.

Excited to share new work on surfacing health equity-related biases in LLMs. We design rubrics covering 6 dimensions of bias, release EquityMedQA, a collection of 7 adversarial datasets, and conduct a large-scale human evaluation study with Med-PaLM 2. arxiv.org/abs/2403.12025

Our new work: jamanetwork.com/journals/jaman… showed that for resident physicians, secure messaging use was linked with an increase in telephone use. Results suggest that use of EHR-integrated messaging channels do not replace of existing channels, but add to the communication burden.

New JAMIA paper from my collaborators and I using language models trained on EHR audit log data to measure behavior, action changes, and much more! Go check it out here:

💡Congratulations to Seunghwan (Nigel) Kim, Ben Warner, Daphne Lew, Sunny Lou, MD PhD, and Thomas Kannampallil, PhD on their recent publication in @jamia_journal! Learn more here: academic.oup.com/jamia/advance-… #WashUMed

Glad to announce that our work has been accepted into ACL SRW 2025 ➡️ Online texts are subject to volatile shifts (semantically similar posts months apart can be of opposing sentiments) ➡️ We propose using continuous time series clustering to annotate and fine-tune pre-trained LLMs

A collaboration among Ben Warner and Chenyang Lu WashU McKelvey Engineering along with Sarah England and Peinan Zhao WashU Medicine OBGYN shows sleep patterns in people experiencing pregnancy can effectively predict preterm birth. #WashUNeuroCommunity bit.ly/4ns79bn