Machine Learning to Predict In-Hospital Mortality Among Patients with Severe Obesity

A recent study featuring PISC Past Trainee Alexis M. Zebrowski, PhD, MPH, evaluates the efficacy of a machine-learning model in predicting in-hospital mortality of patients with severe obesity. The machine-learning model draws data from January 2011 to December 2019 across five hospitals within the Mount Sinai health system. Within the model, 14,078 inpatients with severe obesity (BMI > 40 kg/m2) are evaluated. Data analysis reveals an in-hospital mortality rate of 2.1%. Albumin, acuity level, and chief complaint are the best single predictors of outcome. The team concludes that machine learning models can provide proof of concept performance, effectively predicting mortality rates of patients with severe obesity.

Go to Article

Back to News