Development and Validation of an Interpretable Machine Learning Model for Predicting Adverse Clinical Outcomes in Placenta Accreta Spectrum: A Multicenter Study.
Authors
Affiliations (5)
Affiliations (5)
- Department of Radiology, The Third Affiliated Hospital of Shenzhen University (Luohu Hospital Group), Shenzhen 518000, China (H.L., Y.Y., L.W., X.C., K.W., H.L.).
- Department of Radiology, The Second Affiliated Hospital of Soochow University, Suzhou 215004, China (Y.Z.).
- Department of Urology, Southern University of Science and Technology Hospital, Shenzhen 518000, China (H.M.).
- Shantou University Medical College, Shantou University, Shantou 515000, China (W.L., H.Z., J.H.).
- Department of Radiology, The Third Affiliated Hospital of Shenzhen University (Luohu Hospital Group), Shenzhen 518000, China (H.L., Y.Y., L.W., X.C., K.W., H.L.). Electronic address: [email protected].
Abstract
Placenta accreta spectrum (PAS) is a serious perinatal complication. Accurate preoperative identification of patients at high risk for adverse clinical outcomes is essential for developing personalized treatment strategies. This study aimed to develop and validate a high-performance, interpretable machine learning model that integrates MRI morphological indicators and clinical features to predict adverse outcomes in PAS, and to build an online prediction tool to enhance its clinical applicability. This retrospective study included 125 clinically confirmed PAS patients from two centers, categorized into high-risk (intraoperative blood loss over 1500 mL or requiring hysterectomy) and low-risk groups. Data from Center 1 were used for model development, and data from Center 2 served as the external validation set. Five MRI morphological indicators and six clinical features were extracted as model inputs. Three machine learning classifiers-AdaBoost, TabPFN, and CatBoost-were trained and evaluated on both internal testing and external validation cohorts. SHAP analysis was used to interpret model decision-making, and the optimal model was deployed via a Streamlit-based web platform. The CatBoost model achieved the best performance, with AUROCs of 0.90 (95% CI: 0.73-0.99) and 0.84 (95% CI: 0.70-0.97) in the internal testing and external validation sets, respectively. Calibration curves indicated strong agreement between predicted and actual risks. SHAP analysis revealed that "Cervical canal length" and "Gestational age" contributed negatively to high-risk predictions, while "Prior C-sections number", "Placental abnormal vasculature area", and Parturition were positively associated. The final online tool allows real-time risk prediction and visualization of individualized force plots and is freely accessible to clinicians and patients. This study successfully developed an interpretable and practical machine learning model for predicting adverse clinical outcomes in PAS. The accompanying online tool may support clinical decision-making and improve individualized management for PAS patients.