Ultrasound is a widely used imaging modality in clinical practice due to its low cost, portability, and safety. Current research in general AI for healthcare focuses on large language models and general segmentation models, with insufficient attention to solutions addressing both disease prediction and tissue segmentation. In this study, we propose a novel universal framework for ultrasound, namely UniUSNet, which is a promptable framework for ultrasound image classification and segmentation. The universality of this model is derived from its versatility across various aspects. It proficiently manages any ultrasound nature, any anatomical position, any input type and excelling not only in segmentation tasks but also in classification tasks. We introduce a novel module that incorporates this information as a prompt and seamlessly embedding it within the model's learning process. To train and validate our proposed model, we curated a comprehensive ultrasound dataset from publicly accessible sources, encompassing up to 7 distinct anatomical positions with over 9.7K annotations. Experimental results demonstrate that our model achieves performance comparable to state-of-the-art models, and surpasses both a model trained on a single dataset and an ablated version of the network lacking prompt guidance. Additionally, we conducted zero-shot and fine-tuning experiments on new datasets, which proved that our model possesses strong generalization capabilities and can be effectively adapted to new data at low cost through its adapter module. We will continuously expand the dataset and optimize the task specific prompting mechanism towards the universality in medical ultrasound. Model weights, data processing workflows, and code will be open source to the public (https://github.com/Zehui-Lin/UniUSNet).
View on arXiv