Influence of Different Types of Retinal Cameras on the Performance of Deep Learning Algorithms in Diabetic Retinopathy Screening
<b>Background:</b> The aim of this study was to assess the performance of regional graders and artificial intelligence algorithms across retinal cameras with different specifications in classifying an image as gradable and ungradable. <b>Methods:</b> Study subjects were inclu...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2022-10-01
|
Series: | Life |
Subjects: | |
Online Access: | https://www.mdpi.com/2075-1729/12/10/1610 |
_version_ | 1797472038202376192 |
---|---|
author | Ramyaa Srinivasan Janani Surya Paisan Ruamviboonsuk Peranut Chotcomwongse Rajiv Raman |
author_facet | Ramyaa Srinivasan Janani Surya Paisan Ruamviboonsuk Peranut Chotcomwongse Rajiv Raman |
author_sort | Ramyaa Srinivasan |
collection | DOAJ |
description | <b>Background:</b> The aim of this study was to assess the performance of regional graders and artificial intelligence algorithms across retinal cameras with different specifications in classifying an image as gradable and ungradable. <b>Methods:</b> Study subjects were included from a community-based nationwide diabetic retinopathy screening program in Thailand. Various non-mydriatic fundus cameras were used for image acquisition, including Kowa Nonmyd, Kowa Nonmyd α-DⅢ, Kowa Nonmyd 7, Kowa Nonmyd WX, Kowa VX 10 α, Kowa VX 20 and Nidek AFC 210. All retinal photographs were graded by deep learning algorithms and human graders and compared with a standard reference. <b>Results:</b> Images were divided into two categories as gradable and ungradable images. Four thousand eight hundred fifty-two participants with 19,408 fundus images were included, of which 15,351 (79.09%) were gradable images and the remaining 4057 (20.90%) were ungradable images. <b>Conclusions:</b> The deep learning (DL) algorithm demonstrated better sensitivity, specificity and kappa than the human graders for all eight types of non-mydriatic fundus cameras. The deep learning system showed, more consistent diagnostic performance than the human graders across images of varying quality and camera types. |
first_indexed | 2024-03-09T19:56:20Z |
format | Article |
id | doaj.art-f740e8cccf084e1f9bd01f3b840d6218 |
institution | Directory Open Access Journal |
issn | 2075-1729 |
language | English |
last_indexed | 2024-03-09T19:56:20Z |
publishDate | 2022-10-01 |
publisher | MDPI AG |
record_format | Article |
series | Life |
spelling | doaj.art-f740e8cccf084e1f9bd01f3b840d62182023-11-24T00:57:15ZengMDPI AGLife2075-17292022-10-011210161010.3390/life12101610Influence of Different Types of Retinal Cameras on the Performance of Deep Learning Algorithms in Diabetic Retinopathy ScreeningRamyaa Srinivasan0Janani Surya1Paisan Ruamviboonsuk2Peranut Chotcomwongse3Rajiv Raman4Shri Bhagwan Mahavir Vitreoretinal Services, Sankara Nethralaya (Main Campus), No 41 (Old 18), College Road, Chennai 600006, Tamil Nadu, IndiaShri Bhagwan Mahavir Vitreoretinal Services, Sankara Nethralaya (Main Campus), No 41 (Old 18), College Road, Chennai 600006, Tamil Nadu, IndiaDepartment of Ophthalmology, Rajavithi Hospital, College of Medicine, Rangsit University, Bangkok 12000, ThailandDepartment of Ophthalmology, Rajavithi Hospital, College of Medicine, Rangsit University, Bangkok 12000, ThailandShri Bhagwan Mahavir Vitreoretinal Services, Sankara Nethralaya (Main Campus), No 41 (Old 18), College Road, Chennai 600006, Tamil Nadu, India<b>Background:</b> The aim of this study was to assess the performance of regional graders and artificial intelligence algorithms across retinal cameras with different specifications in classifying an image as gradable and ungradable. <b>Methods:</b> Study subjects were included from a community-based nationwide diabetic retinopathy screening program in Thailand. Various non-mydriatic fundus cameras were used for image acquisition, including Kowa Nonmyd, Kowa Nonmyd α-DⅢ, Kowa Nonmyd 7, Kowa Nonmyd WX, Kowa VX 10 α, Kowa VX 20 and Nidek AFC 210. All retinal photographs were graded by deep learning algorithms and human graders and compared with a standard reference. <b>Results:</b> Images were divided into two categories as gradable and ungradable images. Four thousand eight hundred fifty-two participants with 19,408 fundus images were included, of which 15,351 (79.09%) were gradable images and the remaining 4057 (20.90%) were ungradable images. <b>Conclusions:</b> The deep learning (DL) algorithm demonstrated better sensitivity, specificity and kappa than the human graders for all eight types of non-mydriatic fundus cameras. The deep learning system showed, more consistent diagnostic performance than the human graders across images of varying quality and camera types.https://www.mdpi.com/2075-1729/12/10/1610diabetic retinopathyartificial intelligenceretinal cameraretinal images |
spellingShingle | Ramyaa Srinivasan Janani Surya Paisan Ruamviboonsuk Peranut Chotcomwongse Rajiv Raman Influence of Different Types of Retinal Cameras on the Performance of Deep Learning Algorithms in Diabetic Retinopathy Screening Life diabetic retinopathy artificial intelligence retinal camera retinal images |
title | Influence of Different Types of Retinal Cameras on the Performance of Deep Learning Algorithms in Diabetic Retinopathy Screening |
title_full | Influence of Different Types of Retinal Cameras on the Performance of Deep Learning Algorithms in Diabetic Retinopathy Screening |
title_fullStr | Influence of Different Types of Retinal Cameras on the Performance of Deep Learning Algorithms in Diabetic Retinopathy Screening |
title_full_unstemmed | Influence of Different Types of Retinal Cameras on the Performance of Deep Learning Algorithms in Diabetic Retinopathy Screening |
title_short | Influence of Different Types of Retinal Cameras on the Performance of Deep Learning Algorithms in Diabetic Retinopathy Screening |
title_sort | influence of different types of retinal cameras on the performance of deep learning algorithms in diabetic retinopathy screening |
topic | diabetic retinopathy artificial intelligence retinal camera retinal images |
url | https://www.mdpi.com/2075-1729/12/10/1610 |
work_keys_str_mv | AT ramyaasrinivasan influenceofdifferenttypesofretinalcamerasontheperformanceofdeeplearningalgorithmsindiabeticretinopathyscreening AT jananisurya influenceofdifferenttypesofretinalcamerasontheperformanceofdeeplearningalgorithmsindiabeticretinopathyscreening AT paisanruamviboonsuk influenceofdifferenttypesofretinalcamerasontheperformanceofdeeplearningalgorithmsindiabeticretinopathyscreening AT peranutchotcomwongse influenceofdifferenttypesofretinalcamerasontheperformanceofdeeplearningalgorithmsindiabeticretinopathyscreening AT rajivraman influenceofdifferenttypesofretinalcamerasontheperformanceofdeeplearningalgorithmsindiabeticretinopathyscreening |