Machine Learning-Based Automated Detection and Quantification of Geographic Atrophy and Hypertransmission Defects Using Spectral Domain Optical Coherence Tomography

Abstract

The current study describes the development and assessment of innovative, machine learning (ML)-based approaches for automated detection and pixel-accurate measurements of regions with geographic atrophy (GA) in late-stage age-related macular degeneration (AMD) using optical coherence tomography systems. 900 OCT volumes, 100266 B-scans, and en face OCT images from 341 non-exudative AMD patients with or without GA were included in this study from both Cirrus (Zeiss) and Spectralis (Heidelberg) OCT systems. B-scan and en face level ground truth GA masks were created on OCT B-scan where the segmented ellipsoid zone (EZ) line, retinal pigment epithelium (RPE) line, and bruchs membrane (BM) line overlapped. Two deep learning-based approaches, B-scan level and en face level, were trained. The OCT B-scan model had detection accuracy of 91% and GA area measurement accuracy of 94%. The en face OCT model had detection accuracy of 82% and GA area measurement accuracy of 96% with primary target of hypertransmission on en face OCT. Accuracy was good for both devices tested (92–97%). Automated lesion size stratification for CAM cRORA definition of 250um minimum lesion size was feasible. High-performance models for automatic detection and segmentation of GA area were achieved using OCT systems and deep learning. The automatic measurements showed high correlation with the ground truth. The en face model excelled at identification of hypertransmission defects. The models performance generalized well across device types tested. Future development will include integration of both models to enhance feature detection across GA lesions as well as isolating hypertransmission defects without GA for pre-GA biomarker extraction.

Automated Identification and Segmentation of Ellipsoid Zone At-Risk Using Deep Learning on SD-OCT for Predicting Progression in Dry AMD

“…A DL model was trained using a previously described UNet architecture with approximately 20 million parameters and 41 layers [ 27 ]. The architecture used images resized to 256 × 256 pixel as training inputs, kernel width of 5, early training stopping after 7 epochs without validation improvement, a batch size of 40, and samples per epoch of 200.…”

Relationship Between Hypertransmission Defect Size and Progression in Eyes with Intermediate Age-related Macular Degeneration

“…In addition, our analyses were limited to a single OCT device, and it is not clear how our results would apply to other devices. However, previous reports have shown that iRORA/cRORA lesions may be detected reliably on both Cirrus and Spectralis images [32], and that hyperTDs can be detected on Spectralis or Swept-source OCT en face images as well [20,33,34]. We did not have corresponding visual function data to describe the functional de cit associated with the hyperTD lesions.…”

Identifying geographic atrophy

“…They have been reported to achieve high performance in the quantitative analysis of disease including AMD classification, segmentation of retinal layers, assessing for progression, response to treatment in clinical trials, or predicting visual function [45 ▪▪ ,46]. Deep learning models have been formulated to identify both iRORA and cRORA lesions within an OCT B-scan volume, achieving similar sensitivity compared with human graders [47,48]. Deep learning algorithms using en face SS-OCT images have also been identified as accurate and reproducible for the assessment of GA growth over time [49].…”