High resolution weakly supervised localization architectures for medical images

TL;DR

This paper introduces PYLON, a novel high-resolution weakly-supervised localization architecture for medical images that significantly improves localization accuracy over traditional CAM models, addressing limitations caused by task mismatch and pooling methods.

Contribution

The paper proposes PYLON, a new model that enhances weakly-supervised localization accuracy in medical imaging by overcoming limitations of existing CAM-based methods.

Findings

PYLON achieves 0.62 localization accuracy on NIH Chest X-Ray 14 dataset.

Traditional CAM models achieve around 0.45 accuracy.

Global Average Pooling and Group Normalization negatively impact localization accuracy.

Abstract

In medical imaging, Class-Activation Map (CAM) serves as the main explainability tool by pointing to the region of interest. Since the localization accuracy from CAM is constrained by the resolution of the model's feature map, one may expect that segmentation models, which generally have large feature maps, would produce more accurate CAMs. However, we have found that this is not the case due to task mismatch. While segmentation models are developed for datasets with pixel-level annotation, only image-level annotation is available in most medical imaging datasets. Our experiments suggest that Global Average Pooling (GAP) and Group Normalization are the main culprits that worsen the localization accuracy of CAM. To address this issue, we propose Pyramid Localization Network (PYLON), a model for high-accuracy weakly-supervised localization that achieved 0.62 average point localization accuracy on NIH's Chest X-Ray 14 dataset, compared to 0.45 for a traditional CAM model. Source code and extended results are available at https://github.com/cmb-chula/pylon.