Estimation of current and post-treatment retinal function in chronic central serous chorioretinopathy using artificial intelligence

Abstract

Refined understanding of the association of retinal microstructure with current and future (post-treatment) function in chronic central serous chorioretinopathy (cCSC) may help to identify patients that would benefit most from treatment. In this post-hoc analysis of data from the prospective, randomized PLACE trial (NCT01797861), we aimed to determine the accuracy of AI-based inference of retinal function from retinal morphology in cCSC. Longitudinal spectral-domain optical coherence tomography (SD-OCT) data from 57 eyes of 57 patients from baseline, week 6–8 and month 7–8 post-treatment were segmented using deep-learning software. Fundus-controlled perimetry data were aligned to the SD-OCT data to extract layer thickness and reflectivity values for each test point. Point-wise retinal sensitivity could be inferred with a (leave-one-out) cross-validated mean absolute error (MAE) [95% CI] of 2.93 dB [2.40–3.46] (scenario 1) using random forest regression. With addition of patient-specific baseline data (scenario 2), retinal sensitivity at remaining follow-up visits was estimated even more accurately with a MAE of 1.07 dB [1.06–1.08]. In scenario 3, month 7–8 post-treatment retinal sensitivity was predicted from baseline SD-OCT data with a MAE of 3.38 dB [2.82–3.94]. Our study shows that localized retinal sensitivity can be inferred from retinal structure in cCSC using machine-learning. Especially, prediction of month 7–8 post-treatment sensitivity with consideration of the treatment as explanatory variable constitutes an important step toward personalized treatment decisions in cCSC.

Predicting persistent central serous chorioretinopathy using multiple optical coherence tomographic images by deep learning

“…Diagnosis, classification, and prognosis prediction of a disease using DL has become an active field of imaging research in ophthalmology, especially the retina 6 , 8 – 13 . In CSC, diagnosis and classification of the disease, detection of subretinal fluid, and prediction of post-therapeutic visual acuity have been investigated using DL 9 , 10 , 12 – 15 . Most previous studies have used fundus photography or OCT B-scans alone for development of DL models.…”

“…OCT volume scans were used to estimate post-treatment retinal function in CSC patients in a study by Pfau et al . 15 . Their study showed that localized retinal sensitivity after treatment can be inferred from the thicknesses of the retinal layers in CSC eyes using machine learning (random forest after feature extraction).…”

Central serous chorioretinopathy: A review

“…A recent study by Xu et al 162 reported the application of AI to predict visual acuity and post‐therapeutic OCT of patients with CSC from a base dataset including clinical data from patient's medical records (e.g., age, sex, history of steroid use) and measured features from multimodal imaging (FFA, ICGA and OCT). A post‐hoc analysis of the PLACE trial data by Pfau et al 163 indicated that retinal structure in chronic CSC may be used to infer retinal sensitivity using machine‐learning. In particular, the prediction of post‐treatment retinal sensitivity at the 7‐ to 8‐month follow‐up by taking into account the type of treatment as an explanatory variable could contribute to future individualisation of treatment options.…”

An overview of artificial intelligence in diabetic retinopathy and other ocular diseases

“…A deep neural network-based AI model has been applied for epidermal membrane (ERM) detection based on color fundus photographs ( 80 , 81 ). A random forest-based regression model was used to infer local retinal sensitivity from the retinal structure and the model was applied to the CSC patients for personalized treatment ( 81 ).…”

“…A deep neural network-based AI model has been applied for epidermal membrane (ERM) detection based on color fundus photographs ( 80 , 81 ). A random forest-based regression model was used to infer local retinal sensitivity from the retinal structure and the model was applied to the CSC patients for personalized treatment ( 81 ). Yoon et al ( 82 ) used convolutional neural networks and achieved performance of 93.8, 90.0, 99.1, and 98.9% in accuracy, sensitivity, specificity, and AUC for the diagnosis of CSC.…”