Subject-Specific Lesion Generation and Pseudo-Healthy Synthesis for Multiple Sclerosis Brain Images

TL;DR

This paper introduces a novel generative approach for creating realistic lesion and pseudo-healthy brain images in MS MRI, enhancing disease understanding and segmentation accuracy through synthetic data augmentation.

Contribution

The method uniquely models local lesion characteristics to generate both synthetic lesions on healthy images and pseudo-healthy images from pathological ones, improving data augmentation for segmentation.

Findings

Generated highly realistic pseudo-healthy and pseudo-pathological images.

Data augmentation with synthetic images improves segmentation performance.

Outperforms traditional and recent lesion-aware augmentation methods.

Abstract

Understanding the intensity characteristics of brain lesions is key for defining image-based biomarkers in neurological studies and for predicting disease burden and outcome. In this work, we present a novel foreground-based generative method for modelling the local lesion characteristics that can both generate synthetic lesions on healthy images and synthesize subject-specific pseudo-healthy images from pathological images. Furthermore, the proposed method can be used as a data augmentation module to generate synthetic images for training brain image segmentation networks. Experiments on multiple sclerosis (MS) brain images acquired on magnetic resonance imaging (MRI) demonstrate that the proposed method can generate highly realistic pseudo-healthy and pseudo-pathological brain images. Data augmentation using the synthetic images improves the brain image segmentation performance compared to traditional data augmentation methods as well as a recent lesion-aware data augmentation technique, CarveMix. The code will be released at https://github.com/dogabasaran/lesion-synthesis.