Beyond Retinal Layers: A Deep Voting Model for Automated Geographic Atrophy Segmentation in SD-OCT Images

Abstract

PurposeTo automatically and accurately segment geographic atrophy (GA) in spectral-domain optical coherence tomography (SD-OCT) images by constructing a voting system with deep neural networks without the use of retinal layer segmentation.MethodsAn automatic GA segmentation method for SD-OCT images based on the deep network was constructed. The structure of the deep network was composed of five layers, including one input layer, three hidden layers, and one output layer. During the training phase, the labeled A-scans with 1024 features were directly fed into the network as the input layer to obtain the deep representations. Then a soft-max classifier was trained to determine the label of each individual pixel. Finally, a voting decision strategy was used to refine the segmentation results among 10 trained models.ResultsTwo image data sets with GA were used to evaluate the model. For the first dataset, our algorithm obtained a mean overlap ratio (OR) 86.94% ± 8.75%, absolute area difference (AAD) 11.49% ± 11.50%, and correlation coefficients (CC) 0.9857; for the second dataset, the mean OR, AAD, and CC of the proposed method were 81.66% ± 10.93%, 8.30% ± 9.09%, and 0.9952, respectively. The proposed algorithm was capable of improving over 5% and 10% segmentation accuracy, respectively, when compared with several state-of-the-art algorithms on two data sets.ConclusionsWithout retinal layer segmentation, the proposed algorithm could produce higher segmentation accuracy and was more stable when compared with state-of-the-art methods that relied on retinal layer segmentation results. Our model may provide reliable GA segmentations from SD-OCT images and be useful in the clinical diagnosis of advanced nonexudative AMD.Translational RelevanceBased on the deep neural networks, this study presents an accurate GA segmentation method for SD-OCT images without using any retinal layer segmentation results, and may contribute to improved understanding of advanced nonexudative AMD.

Automated Quantification of Photoreceptor alteration in macular disease using Optical Coherence Tomography and Deep Learning

“…In particular, Kamnitsas et al used an ensemble of neural networks for brain tumour segmentation that reached the first place in the MICCAI-BRATS contest 35 . Similar ideas were recently explored in other OCT imaging-based tasks such as identifying geographic atrophy lesions 36 or segmenting artefacts 37 . The proposed approach is sufficiently general to incorporate any deep learning model for segmentation.…”

Ensemble Deep Learning for Diabetic Retinopathy Detection Using Optical Coherence Tomography Angiography

“…Incorrect predictions are more heavily emphasized in the training of later networks. The networks are then combined typically by majority voting, which can be used for segmentation networks 18 as well as classification networks. Another method of combining multiple networks is stacking, whereby the networks are combined by a meta-classifier.…”

Automatic quantification of superficial foveal avascular zone in optical coherence tomography angiography implemented with deep learning

“…This allows deep learning to provide highly accurate and objective results and perform even better in generalization with new datasets [14]. Currently, deep learning for the automatic classification and segmentation of OCT images in ophthalmology affords excellent results [15,16]. Furthermore, deep learning is applicable to image B/C adjustments [17,18].…”