Predictive models of hospitalization outcomes from the MIMIC-IV dataset. All models were finetuned using our proposed tabular-infused PEFT methods.
Charles Alba
cja5553
AI & ML interests
AI in Public Health, Computational Social Science
Organizations
Predictive models of hospitalization outcomes from the MIMIC-IV dataset. All models were finetuned using our proposed tabular-infused PEFT methods.
Best performing models of paper "Towards Robust Sentiment Analysis of Temporally-Sensitive Policy-Related Online Text"
Models designed to detect spam on Twitter (now known as X).
Predicting multiple surgical complications from pre-operative clinical notes. Finetuned with ~85k clinical notes from Barnes Jewish Hospital.