A novel artificial intelligence-based classification of highly myopic eyes based on visual function and fundus features

IF 2.8 3区医学 Q1 OPHTHALMOLOGY Acta Ophthalmologica Pub Date : 2025-01-19 DOI:10.1111/aos.17026

Jiaqi Meng, Yunxiao Song, Wenwen He, Zhong-Lin Lu, Yuxi Chen, Ling Wei, Keke Zhang, Jiao Qi, Yu Du, Yi Lu, Xiangjia Zhu

{"title":"A novel artificial intelligence-based classification of highly myopic eyes based on visual function and fundus features","authors":"Jiaqi Meng, Yunxiao Song, Wenwen He, Zhong-Lin Lu, Yuxi Chen, Ling Wei, Keke Zhang, Jiao Qi, Yu Du, Yi Lu, Xiangjia Zhu","doi":"10.1111/aos.17026","DOIUrl":null,"url":null,"abstract":"<div>\n \n <section>\n \n <p><b>Aims/Purpose:</b> To develop a novel classification of highly myopic eyes using artificial intelligence (AI) and investigate its relationship with contrast sensitivity function (CSF) and fundus features.</p>\n \n <p><b>Methods:</b> 616 highly myopic eyes of 616 patients were enrolled. CSF was measured using the quantitative CSF method. Myopic macular degeneration (MMD) was graded according to the International META-PM Classification. Thickness of the macula and peripapillary retinal nerve fiber layer (p-RNFL) were assessed by fundus photography and optical coherence tomography, respectively. Classification was performed by combining CSF and fundus features with principal component analysis and k-means clustering.</p>\n \n <p><b>Results:</b> With 83.35% total variance explained, highly myopic eyes were classified into 4 categories. The percentages of categories 1 to 4 were 14.9%, 37.5%, 36.2%, and 11.4%, respectively. CSF of the eyes in category 1 were the highest, followed by those in category 2 and then category 3, while the lowest was seen in category 4. Compared to those in category 1, eyes in category 2 presented higher percentage of MMD2 and thinner temporal p-RNFL. Eyes in categories 3 and 4 presented significantly higher percentage of MMD≥3, thinner nasal macular thickness and p-RNFL (<i>p</i> < 0.05). Multivariate regression showed category 4 had higher MMD grades, thinner macular and p-RNFL thickness compared to category 3.</p>\n \n <p><b>Conclusions:</b> We proposed an AI-based classification of highly myopic eyes by integrating features from both visual function and fundus. It might be an important tool to comprehensively evaluate highly myopic eyes.</p>\n </section>\n </div>","PeriodicalId":6915,"journal":{"name":"Acta Ophthalmologica","volume":"103 S284","pages":""},"PeriodicalIF":2.8000,"publicationDate":"2025-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://onlinelibrary.wiley.com/doi/10.1111/aos.17026","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Acta Ophthalmologica","FirstCategoryId":"3","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1111/aos.17026","RegionNum":3,"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

Abstract

Aims/Purpose: To develop a novel classification of highly myopic eyes using artificial intelligence (AI) and investigate its relationship with contrast sensitivity function (CSF) and fundus features.

Methods: 616 highly myopic eyes of 616 patients were enrolled. CSF was measured using the quantitative CSF method. Myopic macular degeneration (MMD) was graded according to the International META-PM Classification. Thickness of the macula and peripapillary retinal nerve fiber layer (p-RNFL) were assessed by fundus photography and optical coherence tomography, respectively. Classification was performed by combining CSF and fundus features with principal component analysis and k-means clustering.

Results: With 83.35% total variance explained, highly myopic eyes were classified into 4 categories. The percentages of categories 1 to 4 were 14.9%, 37.5%, 36.2%, and 11.4%, respectively. CSF of the eyes in category 1 were the highest, followed by those in category 2 and then category 3, while the lowest was seen in category 4. Compared to those in category 1, eyes in category 2 presented higher percentage of MMD2 and thinner temporal p-RNFL. Eyes in categories 3 and 4 presented significantly higher percentage of MMD≥3, thinner nasal macular thickness and p-RNFL (p < 0.05). Multivariate regression showed category 4 had higher MMD grades, thinner macular and p-RNFL thickness compared to category 3.

Conclusions: We proposed an AI-based classification of highly myopic eyes by integrating features from both visual function and fundus. It might be an important tool to comprehensively evaluate highly myopic eyes.

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

基于视觉功能和眼底特征的高度近视眼人工智能分类

目的：利用人工智能（AI）建立高度近视的新分类方法，并探讨其与对比敏感度函数（CSF）和眼底特征的关系。方法：选取616例高度近视患者616只眼。采用定量脑脊液法测定脑脊液。近视黄斑变性（MMD）按照国际META-PM分级进行分级。分别用眼底摄影和光学相干断层扫描评估黄斑和乳头周围视网膜神经纤维层（p-RNFL）的厚度。将脑脊液和眼底特征结合主成分分析和k-means聚类进行分类。结果：高度近视分为4类，总方差解释率为83.35%。第1 ~ 4类所占比例分别为14.9%、37.5%、36.2%和11.4%。第1类眼睛CSF最高，第2、3类次之，第4类最低。与第1类相比，第2类眼睛的MMD2百分比更高，颞部p-RNFL更薄。第3、4类眼MMD≥3百分比、鼻黄斑厚度较薄、p- rnfl显著高于第3、4类眼（p < 0.05）。多因素回归分析显示，与第3类患者相比，第4类患者MMD分级更高，黄斑和p-RNFL厚度更薄。结论：我们提出了一种基于人工智能的高度近视眼分类方法，综合了视觉功能和眼底的特征。它可能是高度近视的一种重要的综合评价工具。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

Acta Ophthalmologica 医学-眼科学

CiteScore

7.60

自引率

5.90%

发文量

433

审稿时长

6 months

期刊介绍： Acta Ophthalmologica is published on behalf of the Acta Ophthalmologica Scandinavica Foundation and is the official scientific publication of the following societies: The Danish Ophthalmological Society, The Finnish Ophthalmological Society, The Icelandic Ophthalmological Society, The Norwegian Ophthalmological Society and The Swedish Ophthalmological Society, and also the European Association for Vision and Eye Research (EVER). Acta Ophthalmologica publishes clinical and experimental original articles, reviews, editorials, educational photo essays (Diagnosis and Therapy in Ophthalmology), case reports and case series, letters to the editor and doctoral theses.