High resolution weakly supervised localization architectures for medical images

10/22/2020
by   Konpat Preechakul, et al.
0

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.

READ FULL TEXT
research
07/01/2020

Weakly-Supervised Segmentation for Disease Localization in Chest X-Ray Images

Deep Convolutional Neural Networks have proven effective in solving the ...
research
12/23/2021

Learning Hierarchical Attention for Weakly-supervised Chest X-Ray Abnormality Localization and Diagnosis

We consider the problem of abnormality localization for clinical applica...
research
11/08/2017

SIMILARnet: Simultaneous Intelligent Localization and Recognition Network

Global Average Pooling (GAP) [4] has been used previously to generate cl...
research
05/29/2020

Weakly Supervised Lesion Localization With Probabilistic-CAM Pooling

Localizing thoracic diseases on chest X-ray plays a critical role in cli...
research
10/02/2020

Tubular Shape Aware Data Generation for Semantic Segmentation in Medical Imaging

Chest X-ray is one of the most widespread examinations of the human body...
research
03/21/2018

Weakly Supervised Medical Diagnosis and Localization from Multiple Resolutions

Diagnostic imaging often requires the simultaneous identification of a m...
research
04/14/2022

LEFM-Nets: Learnable Explicit Feature Map Deep Networks for Segmentation of Histopathological Images of Frozen Sections

Accurate segmentation of medical images is essential for diagnosis and t...

Please sign up or login with your details

Forgot password? Click here to reset