Iterative Attention Mining for Weakly Supervised Thoracic Disease Pattern Localization in Chest X-Rays
This work addresses the challenge of creating better data sources for computer-aided diagnosis in medical imaging, though it is incremental as it builds on existing weakly supervised localization methods.
The paper tackled the problem of localizing thoracic diseases in chest X-rays using only image-level labels, proposing an iterative attention mining strategy with knowledge preservation and multi-scale aggregation to improve localization accuracy, and demonstrated superior performance over a CAM-based approach on the ChestX-ray14 dataset.
Given image labels as the only supervisory signal, we focus on harvesting, or mining, thoracic disease localizations from chest X-ray images. Harvesting such localizations from existing datasets allows for the creation of improved data sources for computer-aided diagnosis and retrospective analyses. We train a convolutional neural network (CNN) for image classification and propose an attention mining (AM) strategy to improve the model's sensitivity or saliency to disease patterns. The intuition of AM is that once the most salient disease area is blocked or hidden from the CNN model, it will pay attention to alternative image regions, while still attempting to make correct predictions. However, the model requires to be properly constrained during AM, otherwise, it may overfit to uncorrelated image parts and forget the valuable knowledge that it has learned from the original image classification task. To alleviate such side effects, we then design a knowledge preservation (KP) loss, which minimizes the discrepancy between responses for X-ray images from the original and the updated networks. Furthermore, we modify the CNN model to include multi-scale aggregation (MSA), improving its localization ability on small-scale disease findings, e.g., lung nodules. We experimentally validate our method on the publicly-available ChestX-ray14 dataset, outperforming a class activation map (CAM)-based approach, and demonstrating the value of our novel framework for mining disease locations.