BackMix: Mitigating Shortcut Learning in Echocardiography with Minimal Supervision

TL;DR

BackMix is a simple data augmentation technique that reduces shortcut learning in echocardiography classification by randomizing backgrounds, improving model focus, accuracy, and generalization, especially with limited labels.

Contribution

The paper introduces BackMix, a novel background augmentation method that mitigates shortcut learning in echocardiogram classification and extends effectively to semi-supervised settings.

Findings

BackMix improves classification accuracy on in- and out-of-distribution datasets.

It enhances model focus on relevant ultrasound regions.

The method remains effective with as little as 5% segmentation labels.

Abstract

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