%0 Journal Article
%T A Machine Learning Approach to Predicting Treatment Outcomes in Bipolar Depression with OCD Comorbidity
%A Rocco de Filippis
%A Abdullah Al Foysal
%J Open Access Library Journal
%V 12
%N 2
%P 1-20
%@ 2333-9721
%D 2025
%I Open Access Library
%R 10.4236/oalib.1112894
%X Bipolar depression with comorbid obsessive-compulsive disorder (OCD) presents a significant clinical challenge due to its complex symptomatology, unpredictable treatment responses, and high relapse rates. Traditional ap-proaches to treatment planning lack reliable tools for predicting pa-tient-specific outcomes, leaving clinicians with limited options for personal-izing care. This study leverages advanced machine learning (ML), specifical-ly XGBoost, to develop a predictive framework capable of classifying treat-ment responses while identifying key predictors such as age, clinical scores (HDRS, YBOCS), and treatment characteristics (quetiapine dose). By incor-porating interpretability techniques such as SHAP (SHapley Additive exPla-nations), the model provides transparent insights into how individual fea-tures influence predictions, making the outputs actionable for clinical deci-sion-making. Furthermore, probabilistic predictions are evaluated and cali-brated using isotonic regression to ensure reliability, particularly for high-stakes applications in psychiatry. Through detailed visual analyses, in-cluding confusion matrices, ROC-AUC curves, SHAP plots, and calibration curves, this research bridges the gap between data-driven methodologies and clinical practice, offering a robust framework for advancing personalized treatment strategies in bipolar depression with OCD comorbidity.
%K Bipolar Depression
%K Obsessive-Compulsive Disorder (OCD)
%K Machine  Learning (ML)
%K XGBoost
%K Treatment Response Prediction
%K SHAP (SHapley Ad-ditive exPlanations)
%U http://www.oalib.com/paper/6848756