Deep-Reticular Pseudodrusen-Net: A 3-Dimensional Deep Network for Detection of Reticular Pseudodrusen on OCT Scans.

Objective: To propose Deep-RPD-Net, a 3-dimensional deep learning network with semisupervised learning (SSL) for the detection of reticular pseudodrusen (RPD) on spectral-domain OCT scans, explain its decision-making, and compare it with baseline methods. Design: Deep learning model development. Participants: Three hundred fifteen participants from the Age-Related Eye Disease Study 2 Ancillary OCT Study (AREDS2) and 161 participants from the Dark Adaptation in Age-related Macular Degeneration Study (DAAMD). Methods: Two datasets comprising of 1304 (826 labeled) and 1479 (1366 labeled) OCT scans were used to develop and evaluate Deep-RPD-Net and baseline models. The AREDS2 RPD labels were transferred from fundus autofluorescence images, which were captured at the same visit for OCT scans, and DAAMD RPD labels were obtained from the Wisconsin reading center. The datasets were divided into 70%, 10%, and 20% at the participant level for training, validation, and test sets, respectively, for the baseline model. Then, SSL was used with the unlabeled OCT scans to improve the trained model. The performance of Deep-RPD-Net was compared to that of 3 retina specialists on a subset of 50 OCT scans for each dataset. En face and B-scan heatmaps of all networks were visualized and graded on 25 OCT scans with positive labels, using a scale of 1 to 4, to explore the models' decision-making. Main Outcome Measures: Accuracy and area under the receiver-operating characteristic curve (AUROC). Results: Deep-RPD-Net achieved the highest performance metrics, with accuracy and AUROC of 0.81 (95% confidence interval [CI]: 0.76-0.87) and 0.91 (95% CI: 0.86-0.95) on the AREDS2 dataset and 0.80 (95% CI: 0.75-0.84) and 0.86 (95% CI: 0.79-0.91) on the DAAMD dataset. On the subjective test, it achieved accuracy of 0.84 compared with 0.76 for the most accurate retina specialist on the AREDS2 dataset and 0.82 compared with 0.84 on the DAAMD dataset. It also achieved the highest visualization grades, of 3.26 and 3.32 for en face and B-scan heatmaps, respectively. Conclusions: Deep-RPD-Net was able to detect RPD accurately from OCT scans. The visualizations of Deep-RPD-Net were the most explainable to the retina specialist with the highest accuracy. The code and pretrained models are publicly available at https://github.com/ncbi-nlp/Deep-RPD-Net. Financial Disclosures: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.