A Deep Learning System for Fully Automated Retinal Vessel Measurement in High Throughput Image Analysis
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-bas...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-03-01
|
| Series: | Frontiers in Cardiovascular Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fcvm.2022.823436/full |
| _version_ | 1818484735702204416 |
|---|---|
| author | Danli Shi Zhihong Lin Wei Wang Zachary Tan Xianwen Shang Xueli Zhang Wei Meng Zongyuan Ge Mingguang He Mingguang He Mingguang He |
| author_facet | Danli Shi Zhihong Lin Wei Wang Zachary Tan Xianwen Shang Xueli Zhang Wei Meng Zongyuan Ge Mingguang He Mingguang He Mingguang He |
| author_sort | Danli Shi |
| collection | DOAJ |
| description | 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. |
| first_indexed | 2024-12-10T15:58:46Z |
| format | Article |
| id | doaj.art-35757379e8e1484dabf6115943e7d954 |
| institution | Directory Open Access Journal |
| issn | 2297-055X |
| language | English |
| last_indexed | 2024-12-10T15:58:46Z |
| publishDate | 2022-03-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Cardiovascular Medicine |
| spelling | doaj.art-35757379e8e1484dabf6115943e7d9542022-12-22T01:42:31ZengFrontiers Media S.A.Frontiers in Cardiovascular Medicine2297-055X2022-03-01910.3389/fcvm.2022.823436823436A Deep Learning System for Fully Automated Retinal Vessel Measurement in High Throughput Image AnalysisDanli Shi0Zhihong Lin1Wei Wang2Zachary Tan3Xianwen Shang4Xueli Zhang5Wei Meng6Zongyuan Ge7Mingguang He8Mingguang He9Mingguang He10State Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangzhou, ChinaFaculty of Engineering, Monash University, Melbourne, VIC, AustraliaState Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangzhou, ChinaCentre for Eye Research Australia, East Melbourne, VIC, AustraliaDepartment of Ophthalmology, Guangdong Provincial People's Hospital, Guangdong Eye Institute, Guangdong Academy of Medical Sciences, Guangzhou, ChinaDepartment of Ophthalmology, Guangdong Provincial People's Hospital, Guangdong Eye Institute, Guangdong Academy of Medical Sciences, Guangzhou, ChinaGuangzhou Vision Tech Medical Technology Co., Ltd., Guangzhou, ChinaResearch Center and Faculty of Engineering, Monash University, Melbourne, VIC, AustraliaState Key Laboratory of Ophthalmology, Zhongshan Ophthalmic Center, Sun Yat-sen University, Guangzhou, ChinaCentre for Eye Research Australia, East Melbourne, VIC, AustraliaDepartment of Ophthalmology, Guangdong Provincial People's Hospital, Guangdong Eye Institute, Guangdong Academy of Medical Sciences, Guangzhou, ChinaMotivationRetinal 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.https://www.frontiersin.org/articles/10.3389/fcvm.2022.823436/fullartificial intelligenceautomated analysishierarchical vessel morphologycardiovascular diseaseepidemiology |
| spellingShingle | Danli Shi Zhihong Lin Wei Wang Zachary Tan Xianwen Shang Xueli Zhang Wei Meng Zongyuan Ge Mingguang He Mingguang He Mingguang He A Deep Learning System for Fully Automated Retinal Vessel Measurement in High Throughput Image Analysis Frontiers in Cardiovascular Medicine artificial intelligence automated analysis hierarchical vessel morphology cardiovascular disease epidemiology |
| title | A Deep Learning System for Fully Automated Retinal Vessel Measurement in High Throughput Image Analysis |
| title_full | A Deep Learning System for Fully Automated Retinal Vessel Measurement in High Throughput Image Analysis |
| title_fullStr | A Deep Learning System for Fully Automated Retinal Vessel Measurement in High Throughput Image Analysis |
| title_full_unstemmed | A Deep Learning System for Fully Automated Retinal Vessel Measurement in High Throughput Image Analysis |
| title_short | A Deep Learning System for Fully Automated Retinal Vessel Measurement in High Throughput Image Analysis |
| title_sort | deep learning system for fully automated retinal vessel measurement in high throughput image analysis |
| topic | artificial intelligence automated analysis hierarchical vessel morphology cardiovascular disease epidemiology |
| url | https://www.frontiersin.org/articles/10.3389/fcvm.2022.823436/full |
| work_keys_str_mv | AT danlishi adeeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT zhihonglin adeeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT weiwang adeeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT zacharytan adeeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT xianwenshang adeeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT xuelizhang adeeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT weimeng adeeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT zongyuange adeeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT mingguanghe adeeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT mingguanghe adeeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT mingguanghe adeeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT danlishi deeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT zhihonglin deeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT weiwang deeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT zacharytan deeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT xianwenshang deeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT xuelizhang deeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT weimeng deeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT zongyuange deeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT mingguanghe deeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT mingguanghe deeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis AT mingguanghe deeplearningsystemforfullyautomatedretinalvesselmeasurementinhighthroughputimageanalysis |