BackMix: Mitigating Shortcut Learning in Echocardiography with Minimal Supervision

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bransby, Kit Mills, Beqiri, Arian, Kim, Woo-Jin Cho, Oliveira, Jorge, Chartsias, Agisilaos, Gomez, Alberto
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913406529830912
author Bransby, Kit Mills
Beqiri, Arian
Kim, Woo-Jin Cho
Oliveira, Jorge
Chartsias, Agisilaos
Gomez, Alberto
author_facet Bransby, Kit Mills
Beqiri, Arian
Kim, Woo-Jin Cho
Oliveira, Jorge
Chartsias, Agisilaos
Gomez, Alberto
contents Neural networks can learn spurious correlations that lead to the correct prediction in a validation set, but generalise poorly because the predictions are right for the wrong reason. This undesired learning of naive shortcuts (Clever Hans effect) can happen for example in echocardiogram view classification when background cues (e.g. metadata) are biased towards a class and the model learns to focus on those background features instead of on the image content. We propose a simple, yet effective random background augmentation method called BackMix, which samples random backgrounds from other examples in the training set. By enforcing the background to be uncorrelated with the outcome, the model learns to focus on the data within the ultrasound sector and becomes invariant to the regions outside this. We extend our method in a semi-supervised setting, finding that the positive effects of BackMix are maintained with as few as 5% of segmentation labels. A loss weighting mechanism, wBackMix, is also proposed to increase the contribution of the augmented examples. We validate our method on both in-distribution and out-of-distribution datasets, demonstrating significant improvements in classification accuracy, region focus and generalisability. Our source code is available at: https://github.com/kitbransby/BackMix
format Preprint
id arxiv_https___arxiv_org_abs_2406_19148
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle BackMix: Mitigating Shortcut Learning in Echocardiography with Minimal Supervision
Bransby, Kit Mills
Beqiri, Arian
Kim, Woo-Jin Cho
Oliveira, Jorge
Chartsias, Agisilaos
Gomez, Alberto
Computer Vision and Pattern Recognition
Artificial Intelligence
Neural networks can learn spurious correlations that lead to the correct prediction in a validation set, but generalise poorly because the predictions are right for the wrong reason. This undesired learning of naive shortcuts (Clever Hans effect) can happen for example in echocardiogram view classification when background cues (e.g. metadata) are biased towards a class and the model learns to focus on those background features instead of on the image content. We propose a simple, yet effective random background augmentation method called BackMix, which samples random backgrounds from other examples in the training set. By enforcing the background to be uncorrelated with the outcome, the model learns to focus on the data within the ultrasound sector and becomes invariant to the regions outside this. We extend our method in a semi-supervised setting, finding that the positive effects of BackMix are maintained with as few as 5% of segmentation labels. A loss weighting mechanism, wBackMix, is also proposed to increase the contribution of the augmented examples. We validate our method on both in-distribution and out-of-distribution datasets, demonstrating significant improvements in classification accuracy, region focus and generalisability. Our source code is available at: https://github.com/kitbransby/BackMix
title BackMix: Mitigating Shortcut Learning in Echocardiography with Minimal Supervision
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2406.19148