JAMIA

Team uses COVID-19 to test automated acute disease profiling

An automated solution for creating phenotyping algorithms, PheNorm, worked well to identify symptomatic COVID-19 cases in electronic health records, suggesting that automation could speed high-throughput phenotyping of acute disease.

The problem with the problem list

Algorithms to infer missing problems and suggest that they be added to electronic health records improved problem list completeness, with benefits for clinical care, patient comprehension of health conditions and population health.

Machine learning predicts delirium

Using machine learning based on electronic health records of ICU patients predicted new-onset delirium with 82% sensitivity, Vanderbilt researchers found.

PheWAS reveals post-COVID-19 diagnoses

Using a high-throughput informatics technique and electronic health records, Vanderbilt researchers found that COVID-19 survivors had an increased risk for more than 40 new diagnoses.

Impact of digital health interventions

Vanderbilt researchers test and recommend statistical approaches to study the association between engagement with digital health interventions and clinical outcomes.

AI predicts 24-hour hospital discharge

Vanderbilt researchers used a machine learning algorithm and data from more than 26,000 hospital stays to predict who would and would not be discharged over the next 24 hours.