A multidimensional clinical prediction model for early screening of recurrent spontaneous abortion: integrating coagulation, immune, and endocrine markers - PubMed
4 hours ago
- #machine learning
- #clinical prediction model
- #recurrent spontaneous abortion
- Developed a multidimensional clinical prediction model for early screening of recurrent spontaneous abortion (RSA).
- Integrated coagulation, immune, and endocrine markers for accurate RSA risk prediction.
- Used a Transformer-based tabular model (TabPFN) which outperformed other machine learning algorithms with an ROC-AUC of 0.927.
- Identified six key biomarkers for a parsimonious model: anti-phosphatidylserine/prothrombin antibodies (aPS/PT), protein C (PC), antinuclear antibodies (ANA), antithrombin III (AT-III), thrombin time (TT), and body mass index (BMI).
- Highlighted thrombo-immune dysregulation as a central mechanism in RSA through SHAP analysis.
- Proposed a cost-effective and scalable screening strategy suitable for resource-limited settings.