FourierNet: Shape-Preserving Network for OCT Segmentation

This code is the implementation of a shape-preserving network, called FourierNet, that we proposed for Henle’s fiber layer (HFL) segmentation in optical coherence tomography images. FourierNet is a new cascaded network design that puts forward the idea of benefiting the shape prior of HFL in the network training.

The source codes are provided here.

NOTE: The following source codes are provided for research purposes only. The authors have no responsibility for any consequences of use of these source codes. If you use any part of these codes, please cite the following paper.

• S. Cansiz, C. Kesim, S.N. Bektas, Z. Kulali, M. Hasanreisoglu, and C. Gunduz-Demir, “FourierNet: Shape-preserving network for Henle’s fiber layer segmentation in optical coherence tomography images,” IEEE Journal of Biomedical and Health Informatics, 27(2):1036-1047, 2023.

Please contact Selahattin Cansiz for further questions.

Source code

The provided zip file contains three files to make the necessary function calls:

• deepModels.py: It includes the network architectures for the FourierNet model.

• trainTestModels.py: It contains function calls to start training and testing processes. The argument should be set to “tr” or “ts” in the main function to call the code for training and testing, respectively. It also includes codes to load the image dataset and sets the parameters used in the network architecture.

• calculateFourierDescriptors.py: It calculates the Fourier descriptor maps from the manual annotations.