GS-EMA: Integrating Gradient Surgery Exponential Moving Average with Boundary-Aware Contrastive Learning for Enhanced Domain Generalization in Aneurysm Segmentation

TL;DR

This paper introduces GS-EMA, a novel method combining gradient surgery exponential moving average with boundary-aware contrastive learning, to improve domain generalization and segmentation accuracy of cerebral aneurysms across diverse medical datasets.

Contribution

The paper presents a new domain generalization approach that enhances aneurysm segmentation by learning domain-invariant features using GS-EMA and BACL techniques.

Findings

Improved segmentation robustness across different datasets.

Reduced over-segmentation and better aneurysm structure capture.

Enhanced ability to generalize to unseen domains.

Abstract

The automated segmentation of cerebral aneurysms is pivotal for accurate diagnosis and treatment planning. Confronted with significant domain shifts and class imbalance in 3D Rotational Angiography (3DRA) data from various medical institutions, the task becomes challenging. These shifts include differences in image appearance, intensity distribution, resolution, and aneurysm size, all of which complicate the segmentation process. To tackle these issues, we propose a novel domain generalization strategy that employs gradient surgery exponential moving average (GS-EMA) optimization technique coupled with boundary-aware contrastive learning (BACL). Our approach is distinct in its ability to adapt to new, unseen domains by learning domain-invariant features, thereby improving the robustness and accuracy of aneurysm segmentation across diverse clinical datasets. The results demonstrate that our proposed approach can extract more domain-invariant features, minimizing over-segmentation and capturing more complete aneurysm structures.