A new handheld fundus camera combined with visual artificial intelligence facilitates diabetic retinopathy screening
AIM: To explore the performance in diabetic retinopathy (DR) screening of artificial intelligence (AI) system by evaluating the image quality of a handheld Optomed Aurora fundus camera in comparison to traditional tabletop fundus cameras and the diagnostic accuracy of DR of the two modalities. METHO...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Press of International Journal of Ophthalmology (IJO PRESS)
2022-04-01
|
| Series: | International Journal of Ophthalmology |
| Subjects: | |
| Online Access: | http://ies.ijo.cn/en_publish/2022/4/20220416.pdf |
| _version_ | 1818315555137912832 |
|---|---|
| author | Shang Ruan Yang Liu Wei-Ting Hu Hui-Xun Jia Shan-Shan Wang Min-Lu Song Meng-Xi Shen Da-Wei Luo Tao Ye Feng-Hua Wang |
| author_facet | Shang Ruan Yang Liu Wei-Ting Hu Hui-Xun Jia Shan-Shan Wang Min-Lu Song Meng-Xi Shen Da-Wei Luo Tao Ye Feng-Hua Wang |
| author_sort | Shang Ruan |
| collection | DOAJ |
| description | AIM: To explore the performance in diabetic retinopathy (DR) screening of artificial intelligence (AI) system by evaluating the image quality of a handheld Optomed Aurora fundus camera in comparison to traditional tabletop fundus cameras and the diagnostic accuracy of DR of the two modalities. METHODS: Overall, 630 eyes were included from three centers and screened by a handheld camera (Aurora, Optomed, Oulu, Finland) and a table-top camera. Image quality was graded by three masked and experienced ophthalmologists. The diagnostic accuracy of the handheld camera and AI system was evaluated in assessing DR lesions and referable DR. RESULTS: Under nonmydriasis status, the handheld fundus camera had better image quality in centration, clarity, and visible range (1.47, 1.48, and 1.40) than conventional tabletop cameras (1.30, 1.28, and 1.18; P<0.001). Detection of retinal hemorrhage, hard exudation, and macular edema were comparable between the two modalities, in principle, with the area under the curve of the handheld fundus camera slightly lower. The sensitivity and specificity for the detection of referable DR with the handheld camera were 82.1% (95%CI: 72.1%-92.2%) and 97.4% (95%CI: 95.4%-99.5%), respectively. The performance of AI detection of DR using the Phoebus Algorithm was satisfactory; however, Phoebus showed a high sensitivity (88.2%, 95%CI: 79.4%-97.1%) and low specificity (40.7%, 95%CI: 34.1%-47.2%) when detecting referable DR. CONCLUSION: The handheld Aurora fundus camera combined with autonomous AI system is well-suited in DR screening without mydriasis because of its high sensitivity of DR detection as well as its image quality, but its specificity needs to be improved with better modeling of the data. Use of this new system is safe and effective in the detection of referable DR in real world practice. |
| first_indexed | 2024-12-13T09:07:24Z |
| format | Article |
| id | doaj.art-afd88062cbc64c0ab5955071ee6049c2 |
| institution | Directory Open Access Journal |
| issn | 2222-3959 2227-4898 |
| language | English |
| last_indexed | 2024-12-13T09:07:24Z |
| publishDate | 2022-04-01 |
| publisher | Press of International Journal of Ophthalmology (IJO PRESS) |
| record_format | Article |
| series | International Journal of Ophthalmology |
| spelling | doaj.art-afd88062cbc64c0ab5955071ee6049c22022-12-21T23:53:03ZengPress of International Journal of Ophthalmology (IJO PRESS)International Journal of Ophthalmology2222-39592227-48982022-04-0115462062710.18240/ijo.2022.04.1620220416A new handheld fundus camera combined with visual artificial intelligence facilitates diabetic retinopathy screeningShang Ruan0Yang Liu1Wei-Ting Hu2Hui-Xun Jia3Shan-Shan Wang4Min-Lu Song5Meng-Xi Shen6Da-Wei Luo7Tao Ye8Feng-Hua Wang9Feng-Hua Wang. Department of Ophthalmology, Shanghai General Hospital, Shanghai Jiao Tong University, School of Medicine, Haining Road No.100, Shanghai 200000, China. shretina@sjtu.edu.cnDepartment of Ophthalmology, Shanghai General Hospital, Shanghai Jiao Tong University, School of Medicine, Shanghai 200080, ChinaShanghai Phoebus Medical Co. Ltd., Shanghai 200000, ChinaDepartment of Ophthalmology, Shanghai General Hospital, Shanghai Jiao Tong University, School of Medicine, Shanghai 200080, ChinaDepartment of Ophthalmology, Shanghai General Hospital, Shanghai Jiao Tong University, School of Medicine, Shanghai 200080, ChinaDepartment of Ophthalmology, Shanghai General Hospital, Shanghai Jiao Tong University, School of Medicine, Shanghai 200080, ChinaDepartment of Ophthalmology, Shanghai General Hospital, Shanghai Jiao Tong University, School of Medicine, Shanghai 200080, ChinaDepartment of Ophthalmology, Shanghai General Hospital, Shanghai Jiao Tong University, School of Medicine, Shanghai 200080, ChinaWest Nanjing Road Community Health Center, Shanghai 200000, ChinaDepartment of Ophthalmology, Shanghai General Hospital, Shanghai Jiao Tong University, School of Medicine, Shanghai 200080, China; Shanghai Key Laboratory of Ocular Fundus DiseasesAIM: To explore the performance in diabetic retinopathy (DR) screening of artificial intelligence (AI) system by evaluating the image quality of a handheld Optomed Aurora fundus camera in comparison to traditional tabletop fundus cameras and the diagnostic accuracy of DR of the two modalities. METHODS: Overall, 630 eyes were included from three centers and screened by a handheld camera (Aurora, Optomed, Oulu, Finland) and a table-top camera. Image quality was graded by three masked and experienced ophthalmologists. The diagnostic accuracy of the handheld camera and AI system was evaluated in assessing DR lesions and referable DR. RESULTS: Under nonmydriasis status, the handheld fundus camera had better image quality in centration, clarity, and visible range (1.47, 1.48, and 1.40) than conventional tabletop cameras (1.30, 1.28, and 1.18; P<0.001). Detection of retinal hemorrhage, hard exudation, and macular edema were comparable between the two modalities, in principle, with the area under the curve of the handheld fundus camera slightly lower. The sensitivity and specificity for the detection of referable DR with the handheld camera were 82.1% (95%CI: 72.1%-92.2%) and 97.4% (95%CI: 95.4%-99.5%), respectively. The performance of AI detection of DR using the Phoebus Algorithm was satisfactory; however, Phoebus showed a high sensitivity (88.2%, 95%CI: 79.4%-97.1%) and low specificity (40.7%, 95%CI: 34.1%-47.2%) when detecting referable DR. CONCLUSION: The handheld Aurora fundus camera combined with autonomous AI system is well-suited in DR screening without mydriasis because of its high sensitivity of DR detection as well as its image quality, but its specificity needs to be improved with better modeling of the data. Use of this new system is safe and effective in the detection of referable DR in real world practice.http://ies.ijo.cn/en_publish/2022/4/20220416.pdfdiabetic retinopathyimage qualityhandheld cameraartificial intelligence |
| spellingShingle | Shang Ruan Yang Liu Wei-Ting Hu Hui-Xun Jia Shan-Shan Wang Min-Lu Song Meng-Xi Shen Da-Wei Luo Tao Ye Feng-Hua Wang A new handheld fundus camera combined with visual artificial intelligence facilitates diabetic retinopathy screening International Journal of Ophthalmology diabetic retinopathy image quality handheld camera artificial intelligence |
| title | A new handheld fundus camera combined with visual artificial intelligence facilitates diabetic retinopathy screening |
| title_full | A new handheld fundus camera combined with visual artificial intelligence facilitates diabetic retinopathy screening |
| title_fullStr | A new handheld fundus camera combined with visual artificial intelligence facilitates diabetic retinopathy screening |
| title_full_unstemmed | A new handheld fundus camera combined with visual artificial intelligence facilitates diabetic retinopathy screening |
| title_short | A new handheld fundus camera combined with visual artificial intelligence facilitates diabetic retinopathy screening |
| title_sort | new handheld fundus camera combined with visual artificial intelligence facilitates diabetic retinopathy screening |
| topic | diabetic retinopathy image quality handheld camera artificial intelligence |
| url | http://ies.ijo.cn/en_publish/2022/4/20220416.pdf |
| work_keys_str_mv | AT shangruan anewhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT yangliu anewhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT weitinghu anewhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT huixunjia anewhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT shanshanwang anewhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT minlusong anewhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT mengxishen anewhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT daweiluo anewhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT taoye anewhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT fenghuawang anewhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT shangruan newhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT yangliu newhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT weitinghu newhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT huixunjia newhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT shanshanwang newhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT minlusong newhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT mengxishen newhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT daweiluo newhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT taoye newhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening AT fenghuawang newhandheldfunduscameracombinedwithvisualartificialintelligencefacilitatesdiabeticretinopathyscreening |