Automated Cone Cell Identification on Adaptive Optics Scanning Laser Ophthalmoscope Images Based on TV-L1 Optical Flow Registration and <i>K</i>-Means Clustering

Automated Cone Cell Identification on Adaptive Optics Scanning Laser Ophthalmoscope Images Based on TV-L1 Optical Flow Registration and <i>K</i>-Means Clustering

Cone cell identification is essential for diagnosing and studying eye diseases. In this paper, we propose an automated cone cell identification method that involves TV-L1 optical flow estimation and <i>K</i>-means clustering. The proposed algorithm consists of the following steps: image...

Full description

Bibliographic Details
Main Authors: Yiwei Chen, Yi He, Jing Wang, Wanyue Li, Lina Xing, Xin Zhang, Guohua Shi
Format: Article
Language:English
Published: MDPI AG 2021-03-01
Series:Applied Sciences
Subjects:
adaptive optics scanning laser ophthalmoscope
image registration
image segmentation
retinal imaging
Online Access:https://www.mdpi.com/2076-3417/11/5/2259
id doaj-3e157a802d7941b985285f823425c9fb
record_format Article
spelling doaj-3e157a802d7941b985285f823425c9fb2021-03-05T00:01:44ZengMDPI AGApplied Sciences2076-34172021-03-01112259225910.3390/app11052259Automated Cone Cell Identification on Adaptive Optics Scanning Laser Ophthalmoscope Images Based on TV-L1 Optical Flow Registration and <i>K</i>-Means ClusteringYiwei Chen0Yi He1Jing Wang2Wanyue Li3Lina Xing4Xin Zhang5Guohua Shi6Jiangsu Key Laboratory of Medical Optics, Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou 215163, ChinaJiangsu Key Laboratory of Medical Optics, Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou 215163, ChinaJiangsu Key Laboratory of Medical Optics, Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou 215163, ChinaJiangsu Key Laboratory of Medical Optics, Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou 215163, ChinaJiangsu Key Laboratory of Medical Optics, Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou 215163, ChinaJiangsu Key Laboratory of Medical Optics, Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou 215163, ChinaJiangsu Key Laboratory of Medical Optics, Suzhou Institute of Biomedical Engineering and Technology, Chinese Academy of Sciences, Suzhou 215163, ChinaCone cell identification is essential for diagnosing and studying eye diseases. In this paper, we propose an automated cone cell identification method that involves TV-L1 optical flow estimation and <i>K</i>-means clustering. The proposed algorithm consists of the following steps: image denoising based on TV-L1 optical flow registration, bias field correction, cone cell identification based on <i>K</i>-means clustering, duplicate identification removal, identification based on threshold segmentation, and merging of closed identified cone cells. Compared with manually labelled ground-truth images, the proposed method shows high effectiveness with precision, recall, and F1 scores of 93.10%, 94.97%, and 94.03%, respectively. The method performance is further evaluated on adaptive optics scanning laser ophthalmoscope images obtained from a healthy subject with low cone cell density and subjects with either diabetic retinopathy or acute zonal occult outer retinopathy. The evaluation results demonstrate that the proposed method can accurately identify cone cells in subjects with healthy retinas and retinal diseases.https://www.mdpi.com/2076-3417/11/5/2259adaptive optics scanning laser ophthalmoscopeimage registrationimage segmentationretinal imaging
collection DOAJ
language English
format Article
sources DOAJ
author Yiwei Chen
Yi He
Jing Wang
Wanyue Li
Lina Xing
Xin Zhang
Guohua Shi
spellingShingle Yiwei Chen
Yi He
Jing Wang
Wanyue Li
Lina Xing
Xin Zhang
Guohua Shi
Automated Cone Cell Identification on Adaptive Optics Scanning Laser Ophthalmoscope Images Based on TV-L1 Optical Flow Registration and <i>K</i>-Means Clustering
Applied Sciences
adaptive optics scanning laser ophthalmoscope
image registration
image segmentation
retinal imaging
author_facet Yiwei Chen
Yi He
Jing Wang
Wanyue Li
Lina Xing
Xin Zhang
Guohua Shi
author_sort Yiwei Chen
title Automated Cone Cell Identification on Adaptive Optics Scanning Laser Ophthalmoscope Images Based on TV-L1 Optical Flow Registration and <i>K</i>-Means Clustering
title_short Automated Cone Cell Identification on Adaptive Optics Scanning Laser Ophthalmoscope Images Based on TV-L1 Optical Flow Registration and <i>K</i>-Means Clustering
title_full Automated Cone Cell Identification on Adaptive Optics Scanning Laser Ophthalmoscope Images Based on TV-L1 Optical Flow Registration and <i>K</i>-Means Clustering
title_fullStr Automated Cone Cell Identification on Adaptive Optics Scanning Laser Ophthalmoscope Images Based on TV-L1 Optical Flow Registration and <i>K</i>-Means Clustering
title_full_unstemmed Automated Cone Cell Identification on Adaptive Optics Scanning Laser Ophthalmoscope Images Based on TV-L1 Optical Flow Registration and <i>K</i>-Means Clustering
title_sort automated cone cell identification on adaptive optics scanning laser ophthalmoscope images based on tv-l1 optical flow registration and <i>k</i>-means clustering
publisher MDPI AG
series Applied Sciences
issn 2076-3417
publishDate 2021-03-01
description Cone cell identification is essential for diagnosing and studying eye diseases. In this paper, we propose an automated cone cell identification method that involves TV-L1 optical flow estimation and <i>K</i>-means clustering. The proposed algorithm consists of the following steps: image denoising based on TV-L1 optical flow registration, bias field correction, cone cell identification based on <i>K</i>-means clustering, duplicate identification removal, identification based on threshold segmentation, and merging of closed identified cone cells. Compared with manually labelled ground-truth images, the proposed method shows high effectiveness with precision, recall, and F1 scores of 93.10%, 94.97%, and 94.03%, respectively. The method performance is further evaluated on adaptive optics scanning laser ophthalmoscope images obtained from a healthy subject with low cone cell density and subjects with either diabetic retinopathy or acute zonal occult outer retinopathy. The evaluation results demonstrate that the proposed method can accurately identify cone cells in subjects with healthy retinas and retinal diseases.
topic adaptive optics scanning laser ophthalmoscope
image registration
image segmentation
retinal imaging
url https://www.mdpi.com/2076-3417/11/5/2259
work_keys_str_mv AT yiweichen automatedconecellidentificationonadaptiveopticsscanninglaserophthalmoscopeimagesbasedontvl1opticalflowregistrationandikimeansclustering
AT yihe automatedconecellidentificationonadaptiveopticsscanninglaserophthalmoscopeimagesbasedontvl1opticalflowregistrationandikimeansclustering
AT jingwang automatedconecellidentificationonadaptiveopticsscanninglaserophthalmoscopeimagesbasedontvl1opticalflowregistrationandikimeansclustering
AT wanyueli automatedconecellidentificationonadaptiveopticsscanninglaserophthalmoscopeimagesbasedontvl1opticalflowregistrationandikimeansclustering
AT linaxing automatedconecellidentificationonadaptiveopticsscanninglaserophthalmoscopeimagesbasedontvl1opticalflowregistrationandikimeansclustering
AT xinzhang automatedconecellidentificationonadaptiveopticsscanninglaserophthalmoscopeimagesbasedontvl1opticalflowregistrationandikimeansclustering
AT guohuashi automatedconecellidentificationonadaptiveopticsscanninglaserophthalmoscopeimagesbasedontvl1opticalflowregistrationandikimeansclustering
_version_ 1724231460422942720

Similar Items