Automatic Diagnosis of Diabetic Retinopathy Stage Focusing Exclusively on Retinal Hemorrhage
<i>Background and Objectives</i>: The present study evaluated the detection of diabetic retinopathy (DR) using an automated fundus camera focusing exclusively on retinal hemorrhage (RH) using a deep convolutional neural network, which is a machine-learning technology. <i>Materials...
| প্রধান লেখক: | , , , , , , , |
|---|---|
| বিন্যাস: | প্রবন্ধ |
| ভাষা: | English |
| প্রকাশিত: |
MDPI AG
2022-11-01
|
| মালা: | Medicina |
| বিষয়গুলি: | |
| অনলাইন ব্যবহার করুন: | https://www.mdpi.com/1648-9144/58/11/1681 |
| _version_ | 1827644148633567232 |
|---|---|
| author | Yoshihiro Tokuda Hitoshi Tabuchi Toshihiko Nagasawa Mao Tanabe Hodaka Deguchi Yuki Yoshizumi Zaigen Ohara Hiroshi Takahashi |
| author_facet | Yoshihiro Tokuda Hitoshi Tabuchi Toshihiko Nagasawa Mao Tanabe Hodaka Deguchi Yuki Yoshizumi Zaigen Ohara Hiroshi Takahashi |
| author_sort | Yoshihiro Tokuda |
| collection | DOAJ |
| description | <i>Background and Objectives</i>: The present study evaluated the detection of diabetic retinopathy (DR) using an automated fundus camera focusing exclusively on retinal hemorrhage (RH) using a deep convolutional neural network, which is a machine-learning technology. <i>Materials and Methods</i>: This investigation was conducted via a prospective and observational study. The study included 89 fundus ophthalmoscopy images. Seventy images passed an image quality review and were graded as showing no apparent DR (<i>n</i> = 51), mild nonproliferative DR (NPDR; <i>n</i> = 16), moderate NPDR (<i>n</i> = 1), severe NPDR (<i>n</i> = 1), and proliferative DR (<i>n</i> = 1) by three retinal experts according to the International Clinical Diabetic Retinopathy Severity scale. The RH numbers and areas were automatically detected and the results of two tests—the detection of mild-or-worse NPDR and the detection of moderate-or-worse NPDR—were examined. <i>Results</i>: The detection of mild-or-worse DR showed a sensitivity of 0.812 (95% confidence interval: 0.680–0.945), specificity of 0.888, and area under the curve (AUC) of 0.884, whereas the detection of moderate-or-worse DR showed a sensitivity of 1.0, specificity of 1.0, and AUC of 1.0. <i>Conclusions</i>: Automated diagnosis using artificial intelligence focusing exclusively on RH could be used to diagnose DR requiring ophthalmologist intervention. |
| first_indexed | 2024-03-09T18:09:54Z |
| format | Article |
| id | doaj.art-679aedae592e446788810858a1c7aecc |
| institution | Directory Open Access Journal |
| issn | 1010-660X 1648-9144 |
| language | English |
| last_indexed | 2024-03-09T18:09:54Z |
| publishDate | 2022-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Medicina |
| spelling | doaj.art-679aedae592e446788810858a1c7aecc2023-11-24T09:11:54ZengMDPI AGMedicina1010-660X1648-91442022-11-015811168110.3390/medicina58111681Automatic Diagnosis of Diabetic Retinopathy Stage Focusing Exclusively on Retinal HemorrhageYoshihiro Tokuda0Hitoshi Tabuchi1Toshihiko Nagasawa2Mao Tanabe3Hodaka Deguchi4Yuki Yoshizumi5Zaigen Ohara6Hiroshi Takahashi7Inouye Eye Hospital, 4-3, Kanda-Surugadai, Chiyoda-ku, Tokyo 101-0062, JapanDepartment of Ophthalmology, Saneikai Tsukazaki Hospital, Himeji 671-1227, JapanDepartment of Ophthalmology, Saneikai Tsukazaki Hospital, Himeji 671-1227, JapanDepartment of Ophthalmology, Saneikai Tsukazaki Hospital, Himeji 671-1227, JapanDepartment of Ophthalmology, Saneikai Tsukazaki Hospital, Himeji 671-1227, JapanDepartment of Ophthalmology, Saneikai Tsukazaki Hospital, Himeji 671-1227, JapanDepartment of Ophthalmology, Saneikai Tsukazaki Hospital, Himeji 671-1227, JapanDepartment of Ophthalmology, Nippon Medical School, Bunkyo-ku, Tokyo 113-8603, Japan<i>Background and Objectives</i>: The present study evaluated the detection of diabetic retinopathy (DR) using an automated fundus camera focusing exclusively on retinal hemorrhage (RH) using a deep convolutional neural network, which is a machine-learning technology. <i>Materials and Methods</i>: This investigation was conducted via a prospective and observational study. The study included 89 fundus ophthalmoscopy images. Seventy images passed an image quality review and were graded as showing no apparent DR (<i>n</i> = 51), mild nonproliferative DR (NPDR; <i>n</i> = 16), moderate NPDR (<i>n</i> = 1), severe NPDR (<i>n</i> = 1), and proliferative DR (<i>n</i> = 1) by three retinal experts according to the International Clinical Diabetic Retinopathy Severity scale. The RH numbers and areas were automatically detected and the results of two tests—the detection of mild-or-worse NPDR and the detection of moderate-or-worse NPDR—were examined. <i>Results</i>: The detection of mild-or-worse DR showed a sensitivity of 0.812 (95% confidence interval: 0.680–0.945), specificity of 0.888, and area under the curve (AUC) of 0.884, whereas the detection of moderate-or-worse DR showed a sensitivity of 1.0, specificity of 1.0, and AUC of 1.0. <i>Conclusions</i>: Automated diagnosis using artificial intelligence focusing exclusively on RH could be used to diagnose DR requiring ophthalmologist intervention.https://www.mdpi.com/1648-9144/58/11/1681fundus ophthalmoscopydiabetic retinopathyretinal hemorrhagedeep learningdeep convolutional neural network |
| spellingShingle | Yoshihiro Tokuda Hitoshi Tabuchi Toshihiko Nagasawa Mao Tanabe Hodaka Deguchi Yuki Yoshizumi Zaigen Ohara Hiroshi Takahashi Automatic Diagnosis of Diabetic Retinopathy Stage Focusing Exclusively on Retinal Hemorrhage Medicina fundus ophthalmoscopy diabetic retinopathy retinal hemorrhage deep learning deep convolutional neural network |
| title | Automatic Diagnosis of Diabetic Retinopathy Stage Focusing Exclusively on Retinal Hemorrhage |
| title_full | Automatic Diagnosis of Diabetic Retinopathy Stage Focusing Exclusively on Retinal Hemorrhage |
| title_fullStr | Automatic Diagnosis of Diabetic Retinopathy Stage Focusing Exclusively on Retinal Hemorrhage |
| title_full_unstemmed | Automatic Diagnosis of Diabetic Retinopathy Stage Focusing Exclusively on Retinal Hemorrhage |
| title_short | Automatic Diagnosis of Diabetic Retinopathy Stage Focusing Exclusively on Retinal Hemorrhage |
| title_sort | automatic diagnosis of diabetic retinopathy stage focusing exclusively on retinal hemorrhage |
| topic | fundus ophthalmoscopy diabetic retinopathy retinal hemorrhage deep learning deep convolutional neural network |
| url | https://www.mdpi.com/1648-9144/58/11/1681 |
| work_keys_str_mv | AT yoshihirotokuda automaticdiagnosisofdiabeticretinopathystagefocusingexclusivelyonretinalhemorrhage AT hitoshitabuchi automaticdiagnosisofdiabeticretinopathystagefocusingexclusivelyonretinalhemorrhage AT toshihikonagasawa automaticdiagnosisofdiabeticretinopathystagefocusingexclusivelyonretinalhemorrhage AT maotanabe automaticdiagnosisofdiabeticretinopathystagefocusingexclusivelyonretinalhemorrhage AT hodakadeguchi automaticdiagnosisofdiabeticretinopathystagefocusingexclusivelyonretinalhemorrhage AT yukiyoshizumi automaticdiagnosisofdiabeticretinopathystagefocusingexclusivelyonretinalhemorrhage AT zaigenohara automaticdiagnosisofdiabeticretinopathystagefocusingexclusivelyonretinalhemorrhage AT hiroshitakahashi automaticdiagnosisofdiabeticretinopathystagefocusingexclusivelyonretinalhemorrhage |