Aretha Zhu, Priya Tailor, Rashika Verma, Isis Zhang, Brian Schott, Catherine Ye, Bernard Szirth, Miriam Habiel, Albert S Khouri
{"title":"在 COVID-19 期间,为服务不足的社区实施深度学习人工智能视力威胁疾病筛查。","authors":"Aretha Zhu, Priya Tailor, Rashika Verma, Isis Zhang, Brian Schott, Catherine Ye, Bernard Szirth, Miriam Habiel, Albert S Khouri","doi":"10.1177/1357633X231158832","DOIUrl":null,"url":null,"abstract":"<p><strong>Introduction: </strong>Age-related macular degeneration, diabetic retinopathy, and glaucoma are vision-threatening diseases that are leading causes of vision loss. Many studies have validated deep learning artificial intelligence for image-based diagnosis of vision-threatening diseases. Our study prospectively investigated deep learning artificial intelligence applications in student-run non-mydriatic screenings for an underserved, primarily Hispanic community during COVID-19.</p><p><strong>Methods: </strong>Five supervised student-run community screenings were held in West New York, New Jersey. Participants underwent non-mydriatic 45-degree retinal imaging by medical students. Images were uploaded to a cloud-based deep learning artificial intelligence for vision-threatening disease referral. An on-site tele-ophthalmology grader and remote clinical ophthalmologist graded images, with adjudication by a senior ophthalmologist to establish the gold standard diagnosis, which was used to assess the performance of deep learning artificial intelligence.</p><p><strong>Results: </strong>A total of 385 eyes from 195 screening participants were included (mean age 52.43 ± 14.5 years, 40.0% female). A total of 48 participants were referred for at least one vision-threatening disease. Deep learning artificial intelligence marked 150/385 (38.9%) eyes as ungradable, compared to 10/385 (2.6%) ungradable as per the human gold standard (<i>p</i> < 0.001). Deep learning artificial intelligence had 63.2% sensitivity, 94.5% specificity, 32.0% positive predictive value, and 98.4% negative predictive value in vision-threatening disease referrals. Deep learning artificial intelligence successfully referred all 4 eyes with multiple vision-threatening diseases. Deep learning artificial intelligence graded images (35.6 ± 13.3 s) faster than the tele-ophthalmology grader (129 ± 41.0) and clinical ophthalmologist (68 ± 21.9, <i>p</i> < 0.001).</p><p><strong>Discussion: </strong>Deep learning artificial intelligence can increase the efficiency and accessibility of vision-threatening disease screenings, particularly in underserved communities. Deep learning artificial intelligence should be adaptable to different environments. Consideration should be given to how deep learning artificial intelligence can best be utilized in a real-world application, whether in computer-aided or autonomous diagnosis.</p>","PeriodicalId":50024,"journal":{"name":"Journal of Telemedicine and Telecare","volume":" ","pages":"1590-1597"},"PeriodicalIF":3.5000,"publicationDate":"2024-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10014445/pdf/10.1177_1357633X231158832.pdf","citationCount":"0","resultStr":"{\"title\":\"Implementation of deep learning artificial intelligence in vision-threatening disease screenings for an underserved community during COVID-19.\",\"authors\":\"Aretha Zhu, Priya Tailor, Rashika Verma, Isis Zhang, Brian Schott, Catherine Ye, Bernard Szirth, Miriam Habiel, Albert S Khouri\",\"doi\":\"10.1177/1357633X231158832\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Introduction: </strong>Age-related macular degeneration, diabetic retinopathy, and glaucoma are vision-threatening diseases that are leading causes of vision loss. Many studies have validated deep learning artificial intelligence for image-based diagnosis of vision-threatening diseases. Our study prospectively investigated deep learning artificial intelligence applications in student-run non-mydriatic screenings for an underserved, primarily Hispanic community during COVID-19.</p><p><strong>Methods: </strong>Five supervised student-run community screenings were held in West New York, New Jersey. Participants underwent non-mydriatic 45-degree retinal imaging by medical students. Images were uploaded to a cloud-based deep learning artificial intelligence for vision-threatening disease referral. An on-site tele-ophthalmology grader and remote clinical ophthalmologist graded images, with adjudication by a senior ophthalmologist to establish the gold standard diagnosis, which was used to assess the performance of deep learning artificial intelligence.</p><p><strong>Results: </strong>A total of 385 eyes from 195 screening participants were included (mean age 52.43 ± 14.5 years, 40.0% female). A total of 48 participants were referred for at least one vision-threatening disease. Deep learning artificial intelligence marked 150/385 (38.9%) eyes as ungradable, compared to 10/385 (2.6%) ungradable as per the human gold standard (<i>p</i> < 0.001). Deep learning artificial intelligence had 63.2% sensitivity, 94.5% specificity, 32.0% positive predictive value, and 98.4% negative predictive value in vision-threatening disease referrals. Deep learning artificial intelligence successfully referred all 4 eyes with multiple vision-threatening diseases. Deep learning artificial intelligence graded images (35.6 ± 13.3 s) faster than the tele-ophthalmology grader (129 ± 41.0) and clinical ophthalmologist (68 ± 21.9, <i>p</i> < 0.001).</p><p><strong>Discussion: </strong>Deep learning artificial intelligence can increase the efficiency and accessibility of vision-threatening disease screenings, particularly in underserved communities. Deep learning artificial intelligence should be adaptable to different environments. Consideration should be given to how deep learning artificial intelligence can best be utilized in a real-world application, whether in computer-aided or autonomous diagnosis.</p>\",\"PeriodicalId\":50024,\"journal\":{\"name\":\"Journal of Telemedicine and Telecare\",\"volume\":\" \",\"pages\":\"1590-1597\"},\"PeriodicalIF\":3.5000,\"publicationDate\":\"2024-12-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10014445/pdf/10.1177_1357633X231158832.pdf\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Telemedicine and Telecare\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1177/1357633X231158832\",\"RegionNum\":3,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2023/3/13 0:00:00\",\"PubModel\":\"Epub\",\"JCR\":\"Q1\",\"JCRName\":\"HEALTH CARE SCIENCES & SERVICES\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Telemedicine and Telecare","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1177/1357633X231158832","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2023/3/13 0:00:00","PubModel":"Epub","JCR":"Q1","JCRName":"HEALTH CARE SCIENCES & SERVICES","Score":null,"Total":0}
引用次数: 0
摘要
导言:老年黄斑变性、糖尿病视网膜病变和青光眼是威胁视力的疾病,也是导致视力丧失的主要原因。许多研究已经验证了深度学习人工智能在基于图像的视力威胁性疾病诊断中的应用。我们的研究前瞻性地调查了深度学习人工智能在 COVID-19 期间由学生负责的非眼底筛查中的应用,该筛查主要针对服务不足的西班牙裔社区:方法:在新泽西州西纽约市举行了五次由学生监督的社区筛查。参加者在医科学生的指导下进行了非眼球45度视网膜成像。图像被上传到基于云的深度学习人工智能,用于威胁视力的疾病转诊。一名现场远程眼科分级师和远程临床眼科医生对图像进行分级,并由一名资深眼科医生进行裁定,以建立金标准诊断,用于评估深度学习人工智能的性能:共纳入195名筛查参与者的385只眼睛(平均年龄(52.43±14.5)岁,女性占40.0%)。共有 48 名参与者因至少一种威胁视力的疾病而被转诊。深度学习人工智能将150/385(38.9%)只眼睛标记为不可分级,而根据人类金标准将10/385(2.6%)只眼睛标记为不可分级(p p 讨论:深度学习人工智能可以提高威胁视力疾病筛查的效率和可及性,尤其是在服务不足的社区。深度学习人工智能应能适应不同的环境。无论是在计算机辅助诊断还是自主诊断中,都应考虑如何在实际应用中更好地利用深度学习人工智能。
Implementation of deep learning artificial intelligence in vision-threatening disease screenings for an underserved community during COVID-19.
Introduction: Age-related macular degeneration, diabetic retinopathy, and glaucoma are vision-threatening diseases that are leading causes of vision loss. Many studies have validated deep learning artificial intelligence for image-based diagnosis of vision-threatening diseases. Our study prospectively investigated deep learning artificial intelligence applications in student-run non-mydriatic screenings for an underserved, primarily Hispanic community during COVID-19.
Methods: Five supervised student-run community screenings were held in West New York, New Jersey. Participants underwent non-mydriatic 45-degree retinal imaging by medical students. Images were uploaded to a cloud-based deep learning artificial intelligence for vision-threatening disease referral. An on-site tele-ophthalmology grader and remote clinical ophthalmologist graded images, with adjudication by a senior ophthalmologist to establish the gold standard diagnosis, which was used to assess the performance of deep learning artificial intelligence.
Results: A total of 385 eyes from 195 screening participants were included (mean age 52.43 ± 14.5 years, 40.0% female). A total of 48 participants were referred for at least one vision-threatening disease. Deep learning artificial intelligence marked 150/385 (38.9%) eyes as ungradable, compared to 10/385 (2.6%) ungradable as per the human gold standard (p < 0.001). Deep learning artificial intelligence had 63.2% sensitivity, 94.5% specificity, 32.0% positive predictive value, and 98.4% negative predictive value in vision-threatening disease referrals. Deep learning artificial intelligence successfully referred all 4 eyes with multiple vision-threatening diseases. Deep learning artificial intelligence graded images (35.6 ± 13.3 s) faster than the tele-ophthalmology grader (129 ± 41.0) and clinical ophthalmologist (68 ± 21.9, p < 0.001).
Discussion: Deep learning artificial intelligence can increase the efficiency and accessibility of vision-threatening disease screenings, particularly in underserved communities. Deep learning artificial intelligence should be adaptable to different environments. Consideration should be given to how deep learning artificial intelligence can best be utilized in a real-world application, whether in computer-aided or autonomous diagnosis.
期刊介绍:
Journal of Telemedicine and Telecare provides excellent peer reviewed coverage of developments in telemedicine and e-health and is now widely recognised as the leading journal in its field. Contributions from around the world provide a unique perspective on how different countries and health systems are using new technology in health care. Sections within the journal include technology updates, editorials, original articles, research tutorials, educational material, review articles and reports from various telemedicine organisations. A subscription to this journal will help you to stay up-to-date in this fast moving and growing area of medicine.