# InfoMask: Masked Variational Latent Representation to Localize Chest Disease

## TL;DR

This paper introduces InfoMask, a novel unsupervised method that improves localization of chest diseases in X-ray images by filtering irrelevant background signals using a learned masking mechanism, without requiring detailed annotations.

## Contribution

It proposes a learned spatial masking mechanism that enhances disease localization accuracy by reducing background noise in attention maps, without relying on pixel-level annotations.

## Key findings

- Effective localization of pneumonia in chest X-rays without pixel annotations
- Outperforms existing weakly supervised localization methods
- Reduces noise in attention maps for better disease region identification

## Abstract

The scarcity of richly annotated medical images is limiting supervised deep learning based solutions to medical image analysis tasks, such as localizing discriminatory radiomic disease signatures. Therefore, it is desirable to leverage unsupervised and weakly supervised models. Most recent weakly supervised localization methods apply attention maps or region proposals in a multiple instance learning formulation. While attention maps can be noisy, leading to erroneously highlighted regions, it is not simple to decide on an optimal window/bag size for multiple instance learning approaches. In this paper, we propose a learned spatial masking mechanism to filter out irrelevant background signals from attention maps. The proposed method minimizes mutual information between a masked variational representation and the input while maximizing the information between the masked representation and class labels. This results in more accurate localization of discriminatory regions. We tested the proposed model on the ChestX-ray8 dataset to localize pneumonia from chest X-ray images without using any pixel-level or bounding-box annotations.