Diagnosis of central serous chorioretinopathy by deep learning analysis of en face images of choroidal vasculature

Purpose To classify central serous chorioretinopathy (CSC) by deep learning (DL) analyses of en face images of choroidal vasculature obtained by optical coherence tomography (OCT) and to analyze the regions of interest for DL from heatmaps. Methods One-hundred eyes were studied; 53 eyes with CSC and 47 normal eyes. Volume scans of 12×12 mm square were obtained at the same time as the OCT angiographic (OCTA) scans (Plex Elite 9000 Swept-Source OCT®, Zeiss). High-quality en face images of the choroidal vasculature of the segmentation slab of one-half of the subfoveal choroidal thickness were created for the analyses. The entire 100 en face images were divided into 80 for training (100 times) and 20 for validation. The Neural Network Console (NNC) developed by Sony and the Keras-Tensorflow backend and developed by Google were used as the software for the classification with 16 layers of convolutional neural networks. The active region of the heatmap based on the feature quantity extracted by DL was also evaluated as the percentages with gradient-weighted class activation mapping implementation in Keras. Results In the 20 eyes used for validation including 8 eyes with CSC, the accuracy rate of the validation was 100% (20/20) for NNC and 95% (19/20) for Keras. This difference was not significant ( P =0.33). The mean active region in the heatmap image was 12.5% in CSC eyes which was significantly lower than the 79.8% in normal eyes ( P <0.01). Conclusions CSC can be automatically classified with high accuracy from en face images of the choroidal vasculature by DLs with different programs, convolutional layer structures, and small data sets. Heatmap analyses showed that DL focused on the area occupied by the choroidal vessels and their uniformity. We conclude that DL can help in the diagnosis of CSC.