$μ$Match: Foundation Models for Semi-supervised Learning and Domain Adaptation in EM

2026-06-19

AI Summary

This work addresses the longstanding challenge in electron microscopy (EM) image segmentation, where task diversity and scarce annotations have heavily relied on labor-intensive manual labeling, hindering efficient ultrastructural analysis. To overcome this limitation, we propose μMatch, a novel framework that systematically integrates multiple vision foundation models—including SAM, SAM2, μSAM, and DINOv2/v3—into EM multi-structure segmentation. By synergistically combining student-teacher semi-supervised learning with domain adaptation strategies, μMatch substantially reduces dependence on fully annotated data. Extensive experiments demonstrate that our method consistently outperforms strong baselines across several challenging EM segmentation tasks while significantly decreasing the required annotation effort, thereby advancing EM analysis toward scalable and efficient automation.

Abstract

Vision foundation models have substantially advanced computer vision, enabling state-of-the-art performance in zero- and few-shot settings. They have been successfully applied to biomedical imaging tasks ranging from organ segmentation in computed tomography to cell segmentation in light microscopy. Electron microscopy (EM) is a central modality for analyzing cellular ultrastructure due to its nanometer-scale resolution. However, the application of foundation models in EM has so far been limited to specific organelles, such as mitochondria, largely due to the diversity of segmentation tasks and the scarcity of comprehensively annotated data. As a result, EM segmentation still predominantly relies on supervised learning, requiring extensive manual annotation and limiting ultrastructural analysis. To address this gap, we propose $μ$Match, a framework for semi-supervised learning and domain adaptation that leverages foundation models. We implement state-of-the-art student-teacher-based methods and evaluate multiple foundation models (SAM, SAM2, $μ$SAM, DINOv2/v3) on challenging EM tasks, including mitochondrion, nucleus, and neurite segmentation. Our results demonstrate consistent improvements over strong baselines and highlight a path toward substantially reducing the annotation effort in EM.

Problem

• electron microscopy
• segmentation
• annotation scarcity
• semi-supervised learning
• domain adaptation

Innovation

• foundation models
• semi-supervised learning
• domain adaptation
• electron microscopy
• μMatch