Vision-Language Foundation Models (VLFMs) exhibit remarkable generalization, yet their direct application to medical ultrasound is severely hindered by a profound modality gap. The unique acoustic physics of ultrasound, characterized by speckle noise, shadowing, and heterogeneous textures, often degrades the performance of off-the-shelf VLFMs. To bridge this gap, we propose a novel Hybrid Tuning (HT) strategy for the parameter-efficient adaptation of CLIP-based models to ultrasound analysis. Instead of updating the pre-trained weights, HT freezes the visual backbone and integrates a specialized lightweight adapter. This adapter features a Frequency Filtering module to suppress domain-specific periodic artifacts and a Noise Estimation module to dynamically calibrate feature representations. Extensive evaluations across six multi-center datasets demonstrate that our HT-enhanced models significantly outperform existing state-of-the-art adapters and medical VLFMs in both segmentation and classification tasks. Notably, HT exhibits exceptional data efficiency in few-shot scenarios and robust cross-dataset generalization. Our findings prove that preserving pre-trained semantic priors while explicitly modeling ultrasound-specific noise is key to unlocking foundational intelligence in automated ultrasound diagnosis. The source code is available at https://github.com/jinggqu/NextGen-UIA.
@article{arxiv.2506.08849,
title = {Adapting Vision-Language Foundation Model for Next Generation Medical Ultrasound Image Analysis},
author = {Jingguo Qu and Xinyang Han and Jia Ai and Juan Wu and Tong Zhao and Tonghuan Xiao and Sheng Ning and Yuqi Yang and Jing Qin and Ann Dorothy King and Winnie Chiu-Wing Chu and Jing Cai and Michael Tin-Cheung Ying},
journal= {arXiv preprint arXiv:2506.08849},
year = {2026}
}
Comments
This is the author-submitted LaTeX version with original typesetting. The final published version is available at https://doi.org/10.1016/j.eswa.2026.132560