DeepBranchTracer is a novel method for reconstructing curvilinear structures from 2D and 3D images. It leverages deep learning to extract both image and geometric features, enabling accurate and continuous reconstruction of line-like objects, such as roads, vessels, and neurons.

It is recommended to use neutube software to visualize images and SWC files. neutube can be downloaded from https://neutracing.com/.

If you found this project useful, you can kindly give us a star ⭐, or cite us in your work 📖:

@inproceedings{liu2024deepbranchtracer, title={DeepBranchTracer: A Generally-Applicable Approach to Curvilinear Structure Reconstruction Using Multi-Feature Learning}, author={Liu, Chao and Zhao, Ting and Zheng, Nenggan}, booktitle={Proceedings of the AAAI Conference on Artificial Intelligence}, volume={38}, number={4}, pages={3548--3557}, year={2024} }