UniUSNet: A Promptable Framework for Universal Ultrasound Disease Prediction and Tissue Segmentation

Zehui Lin, Zhuoneng Zhang, Xindi Hu, Zhifan Gao, Xin Yang +3 more
6/3/2024
cs.CV

Official code: Zehui-Lin/UniUSNet (16 stars, last commit 2025-06-25).

Abstract

Ultrasound is widely used in clinical practice due to its affordability, portability, and safety. However, current AI research often overlooks combined disease prediction and tissue segmentation. We propose UniUSNet, a universal framework for ultrasound image classification and segmentation. This model handles various ultrasound types, anatomical positions, and input formats, excelling in both segmentation and classification tasks. Trained on a comprehensive dataset with over 9.7K annotations from 7 distinct anatomical positions, our model matches state-of-the-art performance and surpasses single-dataset and ablated models. Zero-shot and fine-tuning experiments show strong generalization and adaptability with minimal fine-tuning. We plan to expand our dataset and refine the prompting mechanism, with model weights and code available at (https://github.com/Zehui-Lin/UniUSNet).

View on arXivView PDF

Code Implementations(1)

Zehui-Lin/UniUSNetOfficial100%

[BIBM‘24] This repository is the official implementation for "UniUSNet: A Promptable Framework for Universal Ultrasound Disease Prediction and Tissue Segmentation".

164PythonJun 20, 20241 years ago

Similar Papers

UltraUPConvNet: A UPerNet- and ConvNeXt-Based Multi-Task Network for Ultrasound Tissue Segmentation and Disease Prediction

96% match1 repo42025

Grounding DINO-US-SAM: Text-Prompted Multi-Organ Segmentation in Ultrasound with LoRA-Tuned Vision-Language Models

90% match2025

A Fully Open and Generalizable Foundation Model for Ultrasound Clinical Applications

88% match2025

SAS: Segment Anything Small for Ultrasound -- A Non-Generative Data Augmentation Technique for Robust Deep Learning in Ultrasound Imaging

88% match2025

UMind-VL: A Generalist Ultrasound Vision-Language Model for Unified Grounded Perception and Comprehensive Interpretation

87% match2025

UniUltra: Interactive Parameter-Efficient SAM2 for Universal Ultrasound Segmentation

87% match1 repo242025

Cite this paper

@article{lin2024uniusnet,
  title  = {UniUSNet: A Promptable Framework for Universal Ultrasound Disease Prediction and Tissue Segmentation},
  author = {Zehui Lin and Zhuoneng Zhang and Xindi Hu and Zhifan Gao and Xin Yang and Yue Sun and Dong Ni and Tao Tan},
  year   = {2024},
  eprint = {2406.01154},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2406.01154v3}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2406.01154.svg)](https://codeofpaper.com/paper/2406.01154)

Discussion