Aligning Evaluation with Clinical Priorities: Calibration, Label Shift, and Error Costs

Gerardo Flores, Alyssa Hasegawa Smith, Julia Fukuyama, Ashia C. Wilson

Keywords: binary classification, calibration, label shift, cost-sensitive evaluation, proper scoring rules

TL;DR: Clipped Log Scores measures miscalibration in units of outcomes, allowing for bounded but uncertain label shift as well as asymmetric costs.

Abstract:

Machine learning-based decision support systems are increasingly deployed in clinical settings, where probabilistic scoring functions are used to inform and prioritize patient management decisions.
However, widely used scoring rules, such as accuracy and AUC-ROC, fail to adequately reflect key clinical priorities, including calibration, robustness to distributional shifts, and sensitivity to asymmetric error costs.
In this work, we propose a principled yet practical evaluation framework for selecting calibrated thresholded classifiers that explicitly accounts for uncertainty in class prevalences and domain-specific cost asymmetries.
Building on the theory of proper scoring rules, particularly the Schervish representation, we derive an adjusted variant of cross-entropy (log score) that averages cost-weighted performance over clinically relevant ranges of class balance.
The resulting evaluation is simple to apply, sensitive to clinical deployment conditions, and designed to prioritize models that are both calibrated and robust to real-world variations.

Primary Area: Evaluation (e.g., methodology, meta studies, replicability and validity, human-in-the-loop)

Submission Number: 18357