Adversarial regression training for visualizing the progression of chronic obstructive pulmonary disease with chest x-rays

TL;DR

This paper introduces VR-GAN, a novel adversarial training method that visualizes disease progression in chest X-rays for COPD, enhancing interpretability of regression models in medical imaging.

Contribution

It presents a new generative adversarial network approach for visualizing disease severity progression in regression tasks, specifically applied to COPD chest X-ray analysis.

Findings

Outperforms classification-based visualization techniques.

Produces realistic disease progression images.

Correlates generated maps with actual disease changes.

Abstract

Knowledge of what spatial elements of medical images deep learning methods use as evidence is important for model interpretability, trustiness, and validation. There is a lack of such techniques for models in regression tasks. We propose a method, called visualization for regression with a generative adversarial network (VR-GAN), for formulating adversarial training specifically for datasets containing regression target values characterizing disease severity. We use a conditional generative adversarial network where the generator attempts to learn to shift the output of a regressor through creating disease effect maps that are added to the original images. Meanwhile, the regressor is trained to predict the original regression value for the modified images. A model trained with this technique learns to provide visualization for how the image would appear at different stages of the disease. We analyze our method in a dataset of chest x-rays associated with pulmonary function tests, used for diagnosing chronic obstructive pulmonary disease (COPD). For validation, we compute the difference of two registered x-rays of the same patient at different time points and correlate it to the generated disease effect map. The proposed method outperforms a technique based on classification and provides realistic-looking images, making modifications to images following what radiologists usually observe for this disease. Implementation code is available at https://github.com/ricbl/vrgan.