Biostatisticians Use Machine Learning Approach to Improve Risk Prediction for Recurrent Health Events

Researchers at the U-M School of Public Health have developed a machine learning method that more accurately predicts recurring health events—even with incomplete patient records. Led by biostatistics doctoral student Abigail Loe, the algorithm outperforms traditional models in identifying flare-up risks, such as those in COPD patients. The team plans to apply their approach to other conditions involving recurring events, aiming to improve outcomes through more targeted interventions.