Development and evaluation of a machine learning model to predict unplanned readmission risk in patients with ulcerative colitis

Key Points

• Strong predictive accuracy of the machine learning model for unplanned readmission risk was observed in ulcerative colitis patients, highlighting its clinical relevance.
• The model demonstrated an accuracy rate that supports clinicians in making more informed decisions regarding patient care and management.
• Analysis of the RF-based model indicates its effectiveness in assessing a 1-year risk period for potential unplanned readmissions in ulcerative colitis patients.
• The web-based risk calculator may enable healthcare professionals to personalize treatment plans based on individual patient readmission risks.

Abstract