Frequency guided vision model improves ultrasound image analysis tasks
FreqDINO++: A Frequency-Guided Multi-Task Routing Vision Foundation Model for Universal Ultrasound Analysis
Computer Vision and Pattern Recognition
Summary
Ultrasound scans help doctors check for things like cancer and unborn baby health, but analyzing these images involves many different tasks. The researchers designed a new computer model called FreqDINO++ that helps the same system handle multiple analysis tasks together, like finding lesions and deciding if they are harmful. It uses special techniques to look at image details across different scales and combines information to improve accuracy. The model works better than previous ones across many different ultrasound tests and types of data.
What this means in practice
- •For medical imaging software teams: Develop ultrasound diagnostic tools that perform multiple tasks like lesion detection and classification within a single efficient model.
- •For radiology workflow engineers: Integrate advanced multi-task ultrasound analysis models to improve cancer screening and prenatal diagnosis accuracy and efficiency.
Authors
Qing Xu, Yixuan Zhang, Yue Li, Xiangjian He, Qian Zhang, Mainul Haque, Rong Qu, Wenting Duan, Jieyun Bai, Zhen Chen
Abstract
Ultrasound image analysis plays a crucial role in cancer screening and prenatal diagnosis, yet comprehensive assessment requires jointly addressing tasks such as lesion segmentation and benign-malignant classification. While recent vision foundation models have shown remarkable universal representations, unlocking their potential for ultrasound is bottlenecked by the considerable domain gap from natural images. Existing methods typically fine-tune heavy vision encoders for isolated tasks, incurring substantial computational overhead while overlooking the underlying commonalities across heterogeneous tasks. In this work, we propose FreqDINO++, a frequency-guided multi-task routing vision foundation model for universal ultrasound analysis. We first introduce a Multi-task Routing Adapter (MR-Adapter) to support parameter-efficient integration of task-common and task-specific knowledge, a Frequency-aware Feature Enhancer (F$^2$-Enhancer) is then designed to capture the rich multi-scale frequency characteristics of ultrasound images, and a Task-aligned Collaborative Decoder (TC-Decoder) is devised to promote collaboration between dense and global prediction tasks through global-local token interaction. Extensive experiments on large-scale multi-task and external single-task ultrasound benchmarks demonstrate that FreqDINO++ consistently outperforms strong baselines and recent foundation models across 27 diverse clinical task scenarios, while also showing promising generalization to unseen data. The code is at https://github.com/MingLang-FD/FreqDINO-Plus.