LesionMix: A Lesion-Level Data Augmentation Method for Medical Image Segmentation

TL;DR

LesionMix is a novel lesion-aware data augmentation technique for medical image segmentation that enhances lesion diversity at the lesion level, leading to improved segmentation performance across multiple datasets.

Contribution

Introduces LesionMix, a lesion-level augmentation method that improves diversity and segmentation accuracy by focusing on lesion-specific features.

Findings

Outperforms recent Mix-based augmentation methods

Effective across multiple modalities and datasets

Enhances lesion shape, location, and intensity diversity

Abstract

Data augmentation has become a de facto component of deep learning-based medical image segmentation methods. Most data augmentation techniques used in medical imaging focus on spatial and intensity transformations to improve the diversity of training images. They are often designed at the image level, augmenting the full image, and do not pay attention to specific abnormalities within the image. Here, we present LesionMix, a novel and simple lesion-aware data augmentation method. It performs augmentation at the lesion level, increasing the diversity of lesion shape, location, intensity and load distribution, and allowing both lesion populating and inpainting. Experiments on different modalities and different lesion datasets, including four brain MR lesion datasets and one liver CT lesion dataset, demonstrate that LesionMix achieves promising performance in lesion image segmentation, outperforming several recent Mix-based data augmentation methods. The code will be released at https://github.com/dogabasaran/lesionmix.