Simultaneous Alignment and Surface Regression Using Hybrid 2D-3D Networks for 3DCoherent Layer Segmentation of Retina OCT Images

Code for paper "Simultaneous Alignment and Surface Regression Using Hybrid 2D-3D Networks for 3DCoherent Layer Segmentation of Retina OCT Images."

The codes are implemented in PyTorch 1.4 and trained on NVIDIA 2080Ti GPUs.

Parts of codes are borrowed from https://github.com/Hui-Xie/DeepLearningSeg.

The preprocess code: https://github.com/YufanHe/oct_preprocess

datasets

1. public A2A OCT dataset: https://people.duke.edu/~sf59/RPEDC_Ophth_2013_dataset.htm
2. public JHU OCT dataset: http://iacl.ece.jhu.edu/index.php/Resources

Contact

If you have any questions, please do not hesitate to contact liuhong@stu.xmu.edu.cn