Human-centered XAI for Burn Depth Characterization

TL;DR

This paper introduces a human-centered, explainable AI framework for burn depth classification using ultrasound images, which improves accuracy by integrating expert knowledge and texture features.

Contribution

It presents a novel human-in-the-loop framework utilizing LIME explanations to enhance burn ultrasound classifiers with texture features, validated on real porcine data.

Findings

Texture features improve classifier accuracy.

Accuracy increased from ~88% to ~94%.

Expert knowledge integration validates model improvements.

Abstract

Approximately 1.25 million people in the United States are treated each year for burn injuries. Precise burn injury classification is an important aspect of the medical AI field. In this work, we propose an explainable human-in-the-loop framework for improving burn ultrasound classification models. Our framework leverages an explanation system based on the LIME classification explainer to corroborate and integrate a burn expert's knowledge -- suggesting new features and ensuring the validity of the model. Using this framework, we discover that B-mode ultrasound classifiers can be enhanced by supplying textural features. More specifically, we confirm that texture features based on the Gray Level Co-occurance Matrix (GLCM) of ultrasound frames can increase the accuracy of transfer learned burn depth classifiers. We test our hypothesis on real data from porcine subjects. We show improvements in the accuracy of burn depth classification -- from ~88% to ~94% -- once modified according to our framework.