predictive analytics

Predicting postop opioid use

VUMC study finds that preoperative patient characteristics can predict daily opioid use at six months after surgery, suggesting opportunities for development of electronic risk-stratification algorithms.

Metabolic signatures of Type 2 diabetes risk

Vanderbilt epidemiologists found 32 blood metabolites associated with obesity and showed that adding these to traditional disease prediction models improves accuracy of determining Type 2 diabetes risk.

Rheumatoid arthritis and heart disease: a common path

An increase in certain antibodies in patients with rheumatoid arthritis can serve as a predictive biomarker for cardiovascular disease, Vanderbilt researchers have discovered.

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.

Artificial intelligence predicts opioid overdose in Tennessee

Researchers at Vanderbilt and the Tennessee Department of Health have developed 30-day predictive models for fatal and non-fatal opioid-related overdose among patients receiving opioid prescriptions in the state.

From left, Dan Roden, MD, Ayesha Muhammad, Jonathan Mosley, MD, PhD, and Sara Van Driest, MD, PhD, found that a genome-wide approach can improve the prediction of drug responses.

For more precise drug treatments, ‘squeeze’ the genome: study finds

Large-scale studies will be required to identify the complexity of genetic variations that affect how patients respond to a given drug and whether they will have side effects, according to researchers at Vanderbilt University Medical Center.