基于智能手机的深度学习模型，用于白内障和多种角膜疾病的广泛诊断和分流。

IF 3.7 2区医学 Q1 OPHTHALMOLOGY

British Journal of Ophthalmology Pub Date : 2024-09-20 DOI:10.1136/bjo-2023-324488

Yuta Ueno, Masahiro Oda, Takefumi Yamaguchi, Hideki Fukuoka, Ryohei Nejima, Yoshiyuki Kitaguchi, Masahiro Miyake, Masato Akiyama, Kazunori Miyata, Kenji Kashiwagi, Naoyuki Maeda, Jun Shimazaki, Hisashi Noma, Kensaku Mori, Tetsuro Oshika

{"title":"基于智能手机的深度学习模型，用于白内障和多种角膜疾病的广泛诊断和分流。","authors":"Yuta Ueno, Masahiro Oda, Takefumi Yamaguchi, Hideki Fukuoka, Ryohei Nejima, Yoshiyuki Kitaguchi, Masahiro Miyake, Masato Akiyama, Kazunori Miyata, Kenji Kashiwagi, Naoyuki Maeda, Jun Shimazaki, Hisashi Noma, Kensaku Mori, Tetsuro Oshika","doi":"10.1136/bjo-2023-324488","DOIUrl":null,"url":null,"abstract":"Aim: To develop an artificial intelligence (AI) algorithm that diagnoses cataracts/corneal diseases from multiple conditions using smartphone images.Methods: This study included 6442 images that were captured using a slit-lamp microscope (6106 images) and smartphone (336 images). An AI algorithm was developed based on slit-lamp images to differentiate 36 major diseases (cataracts and corneal diseases) into 9 categories. To validate the AI model, smartphone images were used for the testing dataset. We evaluated AI performance that included sensitivity, specificity and receiver operating characteristic (ROC) curve for the diagnosis and triage of the diseases.Results: The AI algorithm achieved an area under the ROC curve of 0.998 (95% CI, 0.992 to 0.999) for normal eyes, 0.986 (95% CI, 0.978 to 0.997) for infectious keratitis, 0.960 (95% CI, 0.925 to 0.994) for immunological keratitis, 0.987 (95% CI, 0.978 to 0.996) for cornea scars, 0.997 (95% CI, 0.992 to 1.000) for ocular surface tumours, 0.993 (95% CI, 0.984 to 1.000) for corneal deposits, 1.000 (95% CI, 1.000 to 1.000) for acute angle-closure glaucoma, 0.992 (95% CI, 0.985 to 0.999) for cataracts and 0.993 (95% CI, 0.985 to 1.000) for bullous keratopathy. The triage of referral suggestion using the smartphone images exhibited high performance, in which the sensitivity and specificity were 1.00 (95% CI, 0.478 to 1.00) and 1.00 (95% CI, 0.976 to 1.000) for 'urgent', 0.867 (95% CI, 0.683 to 0.962) and 1.00 (95% CI, 0.971 to 1.000) for 'semi-urgent', 0.853 (95% CI, 0.689 to 0.950) and 0.983 (95% CI, 0.942 to 0.998) for 'routine' and 1.00 (95% CI, 0.958 to 1.00) and 0.896 (95% CI, 0.797 to 0.957) for 'observation', respectively.Conclusions: The AI system achieved promising performance in the diagnosis of cataracts and corneal diseases.","PeriodicalId":9313,"journal":{"name":"British Journal of Ophthalmology","volume":null,"pages":null},"PeriodicalIF":3.7000,"publicationDate":"2024-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11503034/pdf/","citationCount":"0","resultStr":"{\"title\":\"Deep learning model for extensive smartphone-based diagnosis and triage of cataracts and multiple corneal diseases.\",\"authors\":\"Yuta Ueno, Masahiro Oda, Takefumi Yamaguchi, Hideki Fukuoka, Ryohei Nejima, Yoshiyuki Kitaguchi, Masahiro Miyake, Masato Akiyama, Kazunori Miyata, Kenji Kashiwagi, Naoyuki Maeda, Jun Shimazaki, Hisashi Noma, Kensaku Mori, Tetsuro Oshika\",\"doi\":\"10.1136/bjo-2023-324488\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Aim: To develop an artificial intelligence (AI) algorithm that diagnoses cataracts/corneal diseases from multiple conditions using smartphone images.Methods: This study included 6442 images that were captured using a slit-lamp microscope (6106 images) and smartphone (336 images). An AI algorithm was developed based on slit-lamp images to differentiate 36 major diseases (cataracts and corneal diseases) into 9 categories. To validate the AI model, smartphone images were used for the testing dataset. We evaluated AI performance that included sensitivity, specificity and receiver operating characteristic (ROC) curve for the diagnosis and triage of the diseases.Results: The AI algorithm achieved an area under the ROC curve of 0.998 (95% CI, 0.992 to 0.999) for normal eyes, 0.986 (95% CI, 0.978 to 0.997) for infectious keratitis, 0.960 (95% CI, 0.925 to 0.994) for immunological keratitis, 0.987 (95% CI, 0.978 to 0.996) for cornea scars, 0.997 (95% CI, 0.992 to 1.000) for ocular surface tumours, 0.993 (95% CI, 0.984 to 1.000) for corneal deposits, 1.000 (95% CI, 1.000 to 1.000) for acute angle-closure glaucoma, 0.992 (95% CI, 0.985 to 0.999) for cataracts and 0.993 (95% CI, 0.985 to 1.000) for bullous keratopathy. The triage of referral suggestion using the smartphone images exhibited high performance, in which the sensitivity and specificity were 1.00 (95% CI, 0.478 to 1.00) and 1.00 (95% CI, 0.976 to 1.000) for 'urgent', 0.867 (95% CI, 0.683 to 0.962) and 1.00 (95% CI, 0.971 to 1.000) for 'semi-urgent', 0.853 (95% CI, 0.689 to 0.950) and 0.983 (95% CI, 0.942 to 0.998) for 'routine' and 1.00 (95% CI, 0.958 to 1.00) and 0.896 (95% CI, 0.797 to 0.957) for 'observation', respectively.Conclusions: The AI system achieved promising performance in the diagnosis of cataracts and corneal diseases.\",\"PeriodicalId\":9313,\"journal\":{\"name\":\"British Journal of Ophthalmology\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":3.7000,\"publicationDate\":\"2024-09-20\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11503034/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"British Journal of Ophthalmology\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1136/bjo-2023-324488\",\"RegionNum\":2,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"OPHTHALMOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"British Journal of Ophthalmology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1136/bjo-2023-324488","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"OPHTHALMOLOGY","Score":null,"Total":0}

引用次数: 0

摘要

目的：开发一种人工智能（AI）算法，利用智能手机图像诊断多种情况下的白内障/角膜疾病：这项研究包括使用裂隙灯显微镜（6106 幅图像）和智能手机（336 幅图像）拍摄的 6442 幅图像。根据裂隙灯图像开发了一种人工智能算法，可将 36 种主要疾病（白内障和角膜疾病）分为 9 类。为了验证人工智能模型，测试数据集使用了智能手机图像。我们评估了人工智能的性能，包括诊断和分流疾病的灵敏度、特异性和接收器操作特征曲线（ROC）：对于正常眼睛，人工智能算法的 ROC 曲线下面积为 0.998（95% CI，0.992 至 0.999）；对于感染性角膜炎，人工智能算法的 ROC 曲线下面积为 0.986（95% CI，0.978 至 0.997）；对于免疫性角膜炎，人工智能算法的 ROC 曲线下面积为 0.960（95% CI，0.925 至 0.994）；对于角膜疤痕，人工智能算法的 ROC 曲线下面积为 0.987（95% CI，0.978 至 0.996）；对于角膜炎，人工智能算法的 ROC 曲线下面积为 0.997（95% CI，0.925 至 0.994）。997（95% CI，0.992 至 1.000），角膜沉积 0.993（95% CI，0.984 至 1.000），急性闭角型青光眼 1.000（95% CI，1.000 至 1.000），白内障 0.992（95% CI，0.985 至 0.999），大疱性角膜病 0.993（95% CI，0.985 至 1.000）。使用智能手机图像对转诊建议进行分流表现出很高的性能，其中 "紧急 "的灵敏度和特异性分别为 1.00（95% CI，0.478 至 1.00）和 1.00（95% CI，0.976 至 1.000），"紧急 "的灵敏度和特异性分别为 0.867（95% CI，0.683 至 0.962）和 1.00（95% CI，0.976 至 1.000）。962）和 1.00（95% CI，0.971 至 1.000），"常规 "和 "观察 "分别为 0.853（95% CI，0.689 至 0.950）和 0.983（95% CI，0.942 至 0.998）和 1.00（95% CI，0.958 至 1.00）和 0.896（95% CI，0.797 至 0.957）：人工智能系统在诊断白内障和角膜疾病方面表现良好。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Deep learning model for extensive smartphone-based diagnosis and triage of cataracts and multiple corneal diseases.

Aim: To develop an artificial intelligence (AI) algorithm that diagnoses cataracts/corneal diseases from multiple conditions using smartphone images.

Methods: This study included 6442 images that were captured using a slit-lamp microscope (6106 images) and smartphone (336 images). An AI algorithm was developed based on slit-lamp images to differentiate 36 major diseases (cataracts and corneal diseases) into 9 categories. To validate the AI model, smartphone images were used for the testing dataset. We evaluated AI performance that included sensitivity, specificity and receiver operating characteristic (ROC) curve for the diagnosis and triage of the diseases.

Results: The AI algorithm achieved an area under the ROC curve of 0.998 (95% CI, 0.992 to 0.999) for normal eyes, 0.986 (95% CI, 0.978 to 0.997) for infectious keratitis, 0.960 (95% CI, 0.925 to 0.994) for immunological keratitis, 0.987 (95% CI, 0.978 to 0.996) for cornea scars, 0.997 (95% CI, 0.992 to 1.000) for ocular surface tumours, 0.993 (95% CI, 0.984 to 1.000) for corneal deposits, 1.000 (95% CI, 1.000 to 1.000) for acute angle-closure glaucoma, 0.992 (95% CI, 0.985 to 0.999) for cataracts and 0.993 (95% CI, 0.985 to 1.000) for bullous keratopathy. The triage of referral suggestion using the smartphone images exhibited high performance, in which the sensitivity and specificity were 1.00 (95% CI, 0.478 to 1.00) and 1.00 (95% CI, 0.976 to 1.000) for 'urgent', 0.867 (95% CI, 0.683 to 0.962) and 1.00 (95% CI, 0.971 to 1.000) for 'semi-urgent', 0.853 (95% CI, 0.689 to 0.950) and 0.983 (95% CI, 0.942 to 0.998) for 'routine' and 1.00 (95% CI, 0.958 to 1.00) and 0.896 (95% CI, 0.797 to 0.957) for 'observation', respectively.

Conclusions: The AI system achieved promising performance in the diagnosis of cataracts and corneal diseases.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

British Journal of Ophthalmology 医学-眼科学

CiteScore

10.30

自引率

2.40%

发文量

213

审稿时长

3-6 weeks

期刊介绍： The British Journal of Ophthalmology (BJO) is an international peer-reviewed journal for ophthalmologists and visual science specialists. BJO publishes clinical investigations, clinical observations, and clinically relevant laboratory investigations related to ophthalmology. It also provides major reviews and also publishes manuscripts covering regional issues in a global context.