Joint Score-Threshold Optimization for Interpretable Risk Assessment Under Partial Supervision
This work addresses incremental improvements in interpretable risk assessment for healthcare applications, focusing on handling partial supervision and asymmetric costs in clinical workflows.
The authors tackled the problem of optimizing risk assessment tools in healthcare under partial supervision and asymmetric misclassification costs by proposing a mixed-integer programming framework that jointly optimizes scoring weights and category thresholds, achieving practical deployability with governance constraints.
Risk assessment tools in healthcare commonly employ point-based scoring systems that map patients to ordinal risk categories via thresholds. While electronic health record (EHR) data presents opportunities for data-driven optimization of these tools, two fundamental challenges impede standard supervised learning: (1) partial supervision arising from intervention-censored outcomes, where only extreme categories can be reliably labeled, and (2) asymmetric misclassification costs that increase with ordinal distance. We propose a mixed-integer programming (MIP) framework that jointly optimizes scoring weights and category thresholds under these constraints. Our approach handles partial supervision through per-instance feasible label sets, incorporates asymmetric distance-aware objectives, and prevents middle-category collapse via minimum threshold gaps. We further develop a CSO relaxation using softplus losses that preserves the ordinal structure while enabling efficient optimization. The framework supports governance constraints including sign restrictions, sparsity, and minimal modifications to incumbent tools, ensuring practical deployability in clinical workflows.