Vijendran Sruthi, Alok Yash, Kuzhuppilly Neetha I R, Bhat Jayasheela R, Kamath Yogish S
Department of Ophthalmology, Kasturba Medical College, Manipal, Manipal Academy of Higher Education, Manipal, Karnataka, 576104, India.
Department of Community Medicine, Kasturba Medical College, Manipal, Manipal Academy of Higher Education, Manipal, Karnataka, 576104, India.
BMC Ophthalmol. 2025 May 30;25(1):323. doi: 10.1186/s12886-025-04160-2.
Artificial intelligence has become part of healthcare with a multitude of applications being customized to roles required in clinical practice. There has been an expanding growth and development of computer technology with increasing appearance in the ophthalmological universe with roles in detection of most ophthalmic diseases. This article attempts to study the efficacy of smartphones and their applications in detection of paediatric eye diseases.
On 24 January 2024, a comprehensive search was performed across five databases-PubMed, Scopus, Web of Science, Cumulative Index to Nursing and Allied Health Literature, and ProQuest-focusing on studies assessing smartphone-based disease detection and diagnostic accuracy compared to validated methods. Keywords and MeSH terms, including "smartphone," "eye diseases," and "children," were combined using Boolean operators and eligible studies were obtained. The inclusion criteria covered studies from 2000 to 2023, involving children under 18 years, and reporting diagnostic outcomes. Exclusions included studies not exclusive to eye disease, purely adult population studies, reviews, studies with non-availability of full text, and studies exploring other uses of smartphone and designs lacking diagnostic efficacy analysis. Article quality was assessed using the Joanna Briggs Institute Critical Appraisal Checklist.
A total of 2054 articles were retrieved. After removing 1112 duplicates, 507 records were excluded through title screening, followed by 333 through abstract screening. A full-text review of 83 articles led to the inclusion of 33 studies, involving 16,015 participants. Most of the studies (28, 84.84%) were of high quality, with five (15.15%) of moderate quality. Twelve smartphone applications assessed refractive errors using visual acuity tests or photorefraction, five detected amblyogenic risk factors, six identified strabismus, and three targeted leukocoria. Additional applications evaluated stereoacuity (two), eyelid position (one), chalazion (one), corneal diameter (one), and retinopathy of prematurity (two). Overall, these applications demonstrated the potential of smartphones in paediatric eye disease detection.
Smartphone applications are effective tools for detecting important causes of childhood eye disorders such as strabismus, retinopathy of prematurity, chalazion, and refractive errors. These technologies offer promising opportunities for teleophthalmology and integration into routine clinical practice.
人工智能已成为医疗保健的一部分，众多应用程序针对临床实践所需的角色进行了定制。随着计算机技术在眼科领域的应用日益增多，其在大多数眼科疾病的检测中发挥着作用，计算机技术也在不断发展。本文旨在研究智能手机及其应用程序在小儿眼病检测中的功效。
2024年1月24日，在五个数据库——PubMed、Scopus、科学网、护理学与健康相关文献累积索引以及ProQuest上进行了全面检索，重点关注与经过验证的方法相比，评估基于智能手机的疾病检测和诊断准确性的研究。使用布尔运算符组合关键词和医学主题词，包括“智能手机”“眼病”和“儿童”，并获取符合条件的研究。纳入标准涵盖2000年至2023年的研究，涉及18岁以下儿童，并报告诊断结果。排除标准包括不限于眼病的研究、纯粹的成人人群研究、综述、无法获取全文的研究以及探索智能手机其他用途且缺乏诊断效能分析的设计。使用乔安娜·布里格斯研究所批判性评价清单评估文章质量。
共检索到2054篇文章。去除1112篇重复文章后，通过标题筛选排除507条记录，随后通过摘要筛选排除333条记录。对83篇文章进行全文审查后，纳入33项研究，涉及16015名参与者。大多数研究（28项，84.84%）质量较高，5项（15.15%）质量中等。12款智能手机应用程序使用视力测试或验光法评估屈光不正，5款检测致弱视危险因素，6款识别斜视，3款针对白瞳症。其他应用程序评估了立体视（2款）、眼睑位置（1款）、睑板腺囊肿（1款）、角膜直径（1款）和早产儿视网膜病变（2款）。总体而言，这些应用程序展示了智能手机在小儿眼病检测中的潜力。
智能手机应用程序是检测小儿眼病重要病因（如斜视、早产儿视网膜病变、睑板腺囊肿和屈光不正）的有效工具。这些技术为远程眼科以及融入常规临床实践提供了广阔的机会。
