Diagnostic Rule Extraction Using Neural Networks

TL;DR

This paper presents a method to extract understandable diagnostic rules from neural networks trained on incomplete clinical data, aligning machine decisions with doctor’s conclusions.

Contribution

It introduces a technique to convert neural network models into logical formulas and diagnostic tables for better interpretability in medical diagnosis.

Findings

Neural networks correctly classified all test instances.

Extracted rules align with doctor’s diagnoses.

Rules are represented as logical formulas and diagnostic tables.

Abstract

The neural networks have trained on incomplete sets that a doctor could collect. Trained neural networks have correctly classified all the presented instances. The number of intervals entered for encoding the quantitative variables is equal two. The number of features as well as the number of neurons and layers in trained neural networks was minimal. Trained neural networks are adequately represented as a set of logical formulas that more comprehensible and easy-to-understand. These formulas are as the syndrome-complexes, which may be easily tabulated and represented as a diagnostic table that the doctors usually use. Decision rules provide the evaluations of their confidence in which interested a doctor. Conducted clinical researches have shown that iagnostic decisions produced by symbolic rules have coincided with the doctor's conclusions.