An open API service indexing awesome lists of open source software.

https://github.com/Adnan-CAS/AtrousLora

AtrousLora: AI-powered Vascular Segmentation Module
https://github.com/Adnan-CAS/AtrousLora

Last synced: 3 months ago
JSON representation

AtrousLora: AI-powered Vascular Segmentation Module

Awesome Lists containing this project

README

        

# AtrousLora: AI-powered Vascular Segmentation Module
AtrousLora is an advanced deep learning module designed to enhance the SAM (Segment Anything Model) for vascular segmentation in medical imaging. By integrating AtrousLora with SAM, this module improves segmentation performance, providing robust and reliable results essential for medical professionals in tasks such as diagnostic imaging and surgical planning.

Built with flexibility and performance in mind, AtrousLora utilizes cutting-edge transformer-based architectures and innovative methods like LoRA (Low-Rank Adaptation) and Atrous Attention to optimize segmentation. It is tailored to work efficiently even in complex cases, where manual segmentation is time-consuming and prone to human error. AtrousLora aims to reduce the workload of healthcare professionals, enabling more efficient and accurate decisions in patient care. Vascular segmentation is particularly helpful for treatment planning and disease diagnosis, assisting clinicians in visualizing and analyzing blood vessels to improve clinical outcomes.

This module is designed to be seamlessly integrated with SAM, offering an easy-to-use framework for delivering high-quality vascular segmentation across various imaging modalities, making it a versatile tool for a wide range of healthcare applications.

# References
If you find this repository useful, please consider citing the following paper:

[VesselSAM: Leveraging SAM for Aortic Vessel Segmentation with LoRA and Atrous Attention](https://arxiv.org/abs/2502.18185)

```bibtex
@article{adnan2025vesselsam,
title={VesselSAM: Leveraging SAM for Aortic Vessel Segmentation with LoRA and Atrous Attention},
author={Adnan Iltaf, Rayan Merghani Ahmed, Bin Li and Shoujun Zhou},
journal={arXiv preprint arXiv:2502.18185},
year={2025}
}
```

**Note:** The rest of the details and the code will be released soon.