Punctate White Matter Lesion Segmentation in Preterm Infants Powered by Counterfactually Generative Learning

TL;DR

This paper introduces DeepPWML, a novel deep learning framework that uses counterfactual reasoning and tissue segmentation to improve the accuracy of punctate white matter lesion segmentation in infant brain MR images, addressing challenges of small size and low contrast.

Contribution

The paper presents a new counterfactual reasoning approach combined with tissue segmentation for precise PWML localization, outperforming existing methods.

Findings

Achieved state-of-the-art segmentation performance on clinical infant MR dataset.

Demonstrated the effectiveness of counterfactual maps in capturing lesion positional information.

Developed a lightweight and easy-to-implement segmentation network.

Abstract

Accurate segmentation of punctate white matter lesions (PWMLs) are fundamental for the timely diagnosis and treatment of related developmental disorders. Automated PWMLs segmentation from infant brain MR images is challenging, considering that the lesions are typically small and low-contrast, and the number of lesions may dramatically change across subjects. Existing learning-based methods directly apply general network architectures to this challenging task, which may fail to capture detailed positional information of PWMLs, potentially leading to severe under-segmentations. In this paper, we propose to leverage the idea of counterfactual reasoning coupled with the auxiliary task of brain tissue segmentation to learn fine-grained positional and morphological representations of PWMLs for accurate localization and segmentation. A simple and easy-to-implement deep-learning framework (i.e., DeepPWML) is accordingly designed. It combines the lesion counterfactual map with the tissue probability map to train a lightweight PWML segmentation network, demonstrating state-of-the-art performance on a real-clinical dataset of infant T1w MR images. The code is available at \href{https://github.com/ladderlab-xjtu/DeepPWML}{https://github.com/ladderlab-xjtu/DeepPWML}.