Back to feed
arXiv cs.LG
arXiv cs.LG
7/20/2026
A Transportable Threshold-Based Framework for Interpretable Classification of Medical Data

A Transportable Threshold-Based Framework for Interpretable Classification of Medical Data

Short summary

This paper introduces a fully interpretable, rule-based clinical classification framework using Bernoulli Naïve Bayes with chi-square-guided binarization of continuous variables. Evaluated on diabetes, breast cancer, and heart failure datasets, it achieves AUC scores of 0.800–0.984, comparable to complex black-box models. The framework provides explicit decision rules and calibrated probabilities reproducible with basic arithmetic, supporting trustworthy AI adoption in healthcare.

  • •Interpretable BNB framework with chi-square binarization achieves AUC 0.80–0.98 on three medical benchmarks
  • •Calibrated probabilities and explicit decision rules reproducible without software
  • •Addresses black-box adoption barrier in clinical AI by providing full transparency

Generated with AI, which can make mistakes.

#research-breakthrough#ai-tools
Read full article at arXiv cs.LG

Is this a good recommendation for you?

Comments

Failed to load comments. Please try again.

Explore more

Accelerometry-Derived Digital Biomarkers for Cardiometabolic Risk: A Population-Representative Tabular Benchmark with Uncertainty Quantification
arXiv cs.LG
Who Gets Missed in the Tail? Thresholded Subgroup Underdiagnosis in Long-Tailed Chest X-ray Classification
arXiv cs.LG
Interpretable Language Model for Closed-Loop Type 1 Diabetes Control
arXiv CS.AI
From ML Predictions to Informed Diagnostic Assistance Using the Toulmin Model of Argumentation
arXiv CS.AI