UniBrain: A Unified Model for End-to-End Brain Imaging Analysis
UniBrain-Assistant is an open-source, browser-based platform that integrates end-to-end deep learning into a fully conversational workflow for structural brain MRI analysis. We integrate UniBrain into an AI agent for fully automated sMRI preprocessing, analysis, and visualization, achieving a “one-click, done quick” protocol. It functions like a chatbot to invoke expert knowledge to facilitate interactive pipelines.
Visit our UniBrain-Assistant GitHub repository at https://github.com/soz223/UniBrainAssistant.
Recent Papers
Here's a list of publications from our research group related to UniBrain:
| Year | Title | Venue | Code | Paper |
|---|---|---|---|---|
| 2025 | End-to-End Deep Learning for Structural Brain Imaging: A Unified Framework | arXiv preprint arXiv:2502.18523 | GitHub | arXiv |
These publications offer a range of approaches and tools for those interested in Brain Network Analysis.
Contribution
We welcome any form of contribution from the community. Please feel free to email us.
Citation
Please cite our paper if you find our platform useful for your work:
@article{su2025end,
title={End-to-End Deep Learning for Structural Brain Imaging: A Unified Framework},
author={Su, Yao and Han, Keqi and Zeng, Mingjie and Sun, Lichao and Zhan, Liang and Yang, Carl and He, Lifang and Kong, Xiangnan},
journal={arXiv preprint arXiv:2502.18523},
year={2025}
}