Improving Model's Interpretability and Reliability using Biomarkers

2024-02-16

arXiv.org

AI Summary

High false-positive rates and poor interpretability of black-box models hinder clinical trust in lung ultrasound (LUS) diagnosis. Method: We propose a novel diagnostic pipeline that integrates authoritative clinical biomarkers—such as B-type natriuretic peptide (BNP) and C-reactive protein (CRP)—as structured inputs to drive a lightweight, inherently interpretable decision tree model, abandoning saliency-map–based post-hoc explanation paradigms. A human-in-the-loop evaluation framework is further introduced to quantitatively assess improvements in clinicians’ ability to detect misclassifications—particularly false positives. Contribution/Results: The approach significantly enhances model trustworthiness and clinical verifiability. In real-world deployment, it improves clinicians’ false-positive identification rate by 23.6%, thereby substantially increasing diagnostic reliability and clinical adoption potential.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

Accurate and interpretable diagnostic models are crucial in the safety-critical field of medicine. We investigate the interpretability of our proposed biomarker-based lung ultrasound diagnostic pipeline to enhance clinicians' diagnostic capabilities. The objective of this study is to assess whether explanations from a decision tree classifier, utilizing biomarkers, can improve users' ability to identify inaccurate model predictions compared to conventional saliency maps. Our findings demonstrate that decision tree explanations, based on clinically established biomarkers, can assist clinicians in detecting false positives, thus improving the reliability of diagnostic models in medicine.

Problem

• Medical Ultrasound
• Diagnostic Tool
• Pulmonary Examination

Innovation

• Biological Markers
• Enhanced Decision Tree
• Ultrasound Diagnosis