UniUltra: Interactive Parameter-Efficient SAM2 for Universal Ultrasound Segmentation

2025-11-19

AI Summary

To address the performance degradation of SAM2 in ultrasound image segmentation—caused by cross-modal discrepancies—and its impracticality for resource-constrained clinical deployment, this paper proposes a parameter-efficient adaptation framework. Our method introduces (1) a Context-Edge Hybrid Adapter (CH-Adapter) to enhance fine-grained anatomical structure perception, and (2) Deeply Supervised Knowledge Distillation (DSKD) to simultaneously achieve model compression and performance preservation. By fine-tuning only 8.91% of SAM2’s parameters, our adapted model reduces total parameters by 94.08% and significantly lowers inference overhead. Evaluated on a multi-center, multi-organ ultrasound dataset, the proposed approach outperforms existing state-of-the-art methods, demonstrating strong cross-modal generalization and practical clinical deployability.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

The Segment Anything Model 2 (SAM2) demonstrates remarkable universal segmentation capabilities on natural images. However, its performance on ultrasound images is significantly degraded due to domain disparities. This limitation raises two critical challenges: how to efficiently adapt SAM2 to ultrasound imaging while maintaining parameter efficiency, and how to deploy the adapted model effectively in resource-constrained clinical environments. To address these issues, we propose UniUltra for universal ultrasound segmentation. Specifically, we first introduce a novel context-edge hybrid adapter (CH-Adapter) that enhances fine-grained perception across diverse ultrasound imaging modalities while achieving parameter-efficient fine-tuning. To further improve clinical applicability, we develop a deep-supervised knowledge distillation (DSKD) technique that transfers knowledge from the large image encoder of the fine-tuned SAM2 to a super lightweight encoder, substantially reducing computational requirements without compromising performance. Extensive experiments demonstrate that UniUltra outperforms state-of-the-arts with superior generalization capabilities. Notably, our framework achieves competitive performance using only 8.91% of SAM2's parameters during fine-tuning, and the final compressed model reduces the parameter count by 94.08% compared to the original SAM2, making it highly suitable for practical clinical deployment. The source code is available at https://github.com/xq141839/UniUltra.

Problem

• Adapting SAM2 to ultrasound images efficiently despite domain disparities
• Maintaining parameter efficiency while enhancing ultrasound segmentation performance
• Deploying adapted models effectively in resource-constrained clinical environments

Innovation

• Context-edge hybrid adapter enhances ultrasound fine-grained perception
• Deep-supervised knowledge distillation reduces computational requirements
• Parameter-efficient fine-tuning achieves competitive performance with fewer parameters