Constructing and Evaluating an Explainable Model for COVID-19 Diagnosis from Chest X-rays

TL;DR

This paper develops an explainable AI model for COVID-19 diagnosis from chest X-rays, combining neural networks and symbolic decision trees to provide clinically relevant visual and textual explanations.

Contribution

It introduces a hybrid approach that extracts domain-specific features with neural networks and constructs a symbolic decision tree for interpretable diagnosis, enhancing clinical usefulness.

Findings

Neural models effectively identify key radiological features.

Textual explanations based on clinical features are potentially useful.

Visual explanations require clinical relevance to be effective.

Abstract

In this paper, our focus is on constructing models to assist a clinician in the diagnosis of COVID-19 patients in situations where it is easier and cheaper to obtain X-ray data than to obtain high-quality images like those from CT scans. Deep neural networks have repeatedly been shown to be capable of constructing highly predictive models for disease detection directly from image data. However, their use in assisting clinicians has repeatedly hit a stumbling block due to their black-box nature. Some of this difficulty can be alleviated if predictions were accompanied by explanations expressed in clinically relevant terms. In this paper, deep neural networks are used to extract domain-specific features(morphological features like ground-glass opacity and disease indications like pneumonia) directly from the image data. Predictions about these features are then used to construct a symbolic model (a decision tree) for the diagnosis of COVID-19 from chest X-rays, accompanied with two kinds of explanations: visual (saliency maps, derived from the neural stage), and textual (logical descriptions, derived from the symbolic stage). A radiologist rates the usefulness of the visual and textual explanations. Our results demonstrate that neural models can be employed usefully in identifying domain-specific features from low-level image data; that textual explanations in terms of clinically relevant features may be useful; and that visual explanations will need to be clinically meaningful to be useful.