A Deep Learning System for Fully Automated Retinal Vessel Measurement in High Throughput Image Analysis

Abstract

MotivationRetinal microvasculature is a unique window for predicting and monitoring major cardiovascular diseases, but high throughput tools based on deep learning for in-detail retinal vessel analysis are lacking. As such, we aim to develop and validate an artificial intelligence system (Retina-based Microvascular Health Assessment System, RMHAS) for fully automated vessel segmentation and quantification of the retinal microvasculature.ResultsRMHAS achieved good segmentation accuracy across datasets with diverse eye conditions and image resolutions, having AUCs of 0.91, 0.88, 0.95, 0.93, 0.97, 0.95, 0.94 for artery segmentation and 0.92, 0.90, 0.96, 0.95, 0.97, 0.95, 0.96 for vein segmentation on the AV-WIDE, AVRDB, HRF, IOSTAR, LES-AV, RITE, and our internal datasets. Agreement and repeatability analysis supported the robustness of the algorithm. For vessel analysis in quantity, less than 2 s were needed to complete all required analysis.

Integrating oculomics with genomics reveals imaging biomarkers for preventive and personalized prediction of arterial aneurysms

“…One of the concerns is that MFS patients also suffer from high myopia [ 49 ]; this might influence the morphology of the retinal vessels and induce a spurious association between RVF and MFS [ 33 ]. Hence, in this study, apart from clinical risk factors, refractive error was also carefully adjusted in the model to reduce potential bias caused by myopic status.…”

“…Only the retinal fundus images from the right eye were used for analysis. A machine learning system, referred to as the Retinal-based Microvascular Health Assessment System (RMHAS), was previously developed and validated to automatically and quickly extract and quantify retinal microvascular features [ 33 ]. For each image, pan-retinal vessel geometric parameters, such as calibre, complexity, length, tortuosity and branching angle were quantified.…”

“…Compared to clinical phenotypic data, oculomics derived from retinal fundus image is considered more convenient and efficient, non-invasive, with lower costs (for all resources), and more importantly, reflects systemic vascular properties. Previously, we developed a deep learning algorithm, the Retinal-based Microvascular Health Assessment System (RMHAS) [ 33 ] that can generate data on the omics scale from RVFs, thus providing more comprehensive oculomics of RVFs. In this study, personalized genetic risk of aneurysm was integrated with the oculomics by phenome-wide association analysis (PheWAS) to identify biological meaningful aneurysm-RVF biomarkers.…”

Burden, determinants, consequences and care of multimorbidity in rural and urbanising Telangana, India: protocol for a mixed-methods study within the APCAPS cohort

“…Data quality reports that describe missing data, variable distributions and questionnaire completion times are assessed by the research team fortnightly. On an ongoing basis, quality of echocardiograms, ECG traces and CIMT images are examined by an experienced cardiologist and the retinal images with an automated tool 75. The protocol is repeated in a random 5% of participants 2–4 weeks following initial data collection, intraclass correlation coefficients (ICCs) are estimated for key continuous variables (eg, anthropometrics) as the study is ongoing, and refresher training conducted if ICCs indicate decreasing reliability.…”

Using Artificial Intelligence to Analyse the Retinal Vascular Network: The Future of Cardiovascular Risk Assessment Based on Oculomics? A Narrative Review

“…To reduce the involvement of the operators and move from semi-automated to fully automated image processing, several DL-based algorithms have emerged [79][80][81] for either segmenting biomarkers (veins, fovea) or grading the image quality [82]. Since most of the information is supplied by the retinal vessels in oculomics, automated vessel segmentation was the main focus of the research community, with most publications concentrating on this area as compared with other retinal biomarkers (optic disk, fovea [83,84]).…”