Development and Validation of Machine Learning-Based Models for Predicting Postoperative Depression Risk in Patients With Ovarian Cancer
A predictive tool flags ovarian cancer patients at high risk for depression, enabling timely mental health support.
A random forest model outperformed other ML algorithms for predicting postoperative depression in 850 ovarian cancer patients, achieving AUC 0.776 with 13 clinically accessible predictors. The model identifies a targetable high-risk group for perioperative mental health interventions in a population with elevated depression burden.
What the study was
- Study design
- Retrospective ML model development and validation
- Population
- Ovarian cancer surgical patients
- Sample size
- 850
- Category
- Diagnostics
- Maturity
- Validated
- Journal
- Actas Españolas de Psiquiatría
Why it surfaced
Resolved deferred PMID from 2026-04-24. Clinically useful ML prediction model with reasonable performance in n=850; addresses underrecognized mental health burden in ovarian cancer.
A plain-language summary of published research — not medical advice. Talk to a clinician about your own care.