AMA-SAM: Adversarial Multi-Domain Alignment of Segment Anything Model for High-Fidelity Histology Nuclei Segmentation

2025-03-27

AI Summary

To address poor generalization across datasets and insufficient output resolution of the Segment Anything Model (SAM) in nuclear segmentation of histopathological images, this paper proposes a multi-domain collaborative high-fidelity segmentation framework. Methodologically: (1) it pioneers the extension of SAM to multi-domain alignment learning; (2) introduces a Conditional Gradient Reversal Layer (CGRL) to jointly optimize domain-invariant representations and domain-specific discriminative features; and (3) designs a High-Resolution Decoder (HR-Decoder) to directly reconstruct fine-grained nuclear boundaries. The framework integrates adversarial domain alignment, cross-domain self-supervised fine-tuning, and feature reconstruction. Extensive experiments on multiple public histopathology benchmarks demonstrate substantial improvements over state-of-the-art methods, effectively mitigating domain shift while enhancing segmentation accuracy and boundary fidelity.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Segmentation
• Application Domains: Humanities & Computational Social Science

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

Accurate segmentation of cell nuclei in histopathology images is essential for numerous biomedical research and clinical applications. However, existing cell nucleus segmentation methods only consider a single dataset (i.e., primary domain), while neglecting to leverage supplementary data from diverse sources (i.e., auxiliary domains) to reduce overfitting and enhance the performance. Although incorporating multiple datasets could alleviate overfitting, it often exacerbates performance drops caused by domain shifts. In this work, we introduce Adversarial Multi-domain Alignment of Segment Anything Model (AMA-SAM) that extends the Segment Anything Model (SAM) to overcome these obstacles through two key innovations. First, we propose a Conditional Gradient Reversal Layer (CGRL), a multi-domain alignment module that harmonizes features from diverse domains to promote domain-invariant representation learning while preserving crucial discriminative features for the primary dataset. Second, we address SAM's inherent low-resolution output by designing a High-Resolution Decoder (HR-Decoder), which directly produces fine-grained segmentation maps in order to capture intricate nuclei boundaries in high-resolution histology images. To the best of our knowledge, this is the first attempt to adapt SAM for multi-dataset learning with application to histology nuclei segmentation. We validate our method on several publicly available datasets, demonstrating consistent and significant improvements over state-of-the-art approaches.

Problem

• Enhancing nuclei segmentation accuracy in histopathology images
• Reducing domain shift impact in multi-dataset learning
• Improving SAM's resolution for fine-grained nuclei boundaries

Innovation

• Conditional Gradient Reversal Layer for domain alignment
• High-Resolution Decoder for fine-grained segmentation
• Adapting SAM for multi-dataset nuclei segmentation