Search research articles

关于 JoVE

概览领导团队博客 JoVE 帮助中心

作者

出版流程编辑委员会范围与政策同行评审常见问题投稿

图书馆员

用户评价订阅访问资源图书馆顾问委员会常见问题

研究

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments 存档

教育

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual 教师资源中心教师网站

使用条款与条件

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序

Same author

Development of a deep learning model to predict cycloplegic myopia and pre- myopia: based on a large-scale screening of school-aged children.

Eye (London, England)·2026

Same author

Repeatability of cone photoreceptor metrics with a commercial adaptive optics scanning laser ophthalmoscopy (Mona II) in healthy adults.

The British journal of ophthalmology·2026

Same author

Long-term observation of the endothelium morphology and corneal thickness in myopic adolescents treated with orthokeratology: a 10-year retrospective study.

International journal of ophthalmology·2026

Same author

Prevalence of Same-Sex Sexual Behavior in Termites: Persistence Under Mate-Seeking Stress Absence.

Insects·2026

Same author

Comparing myopia control efficacy and corneal power changes in children wearing corneal refractive therapy (CRT) and vision shaping treatment (VST) orthokeratology lenses with smaller back optic zone diameters.

Photodiagnosis and photodynamic therapy·2026

Same author

Effect of 0.01% Atropine on Choroidal Thickness and Ocular Growth in Pre-myopic Children.

Ophthalmic & physiological optics : the journal of the British College of Ophthalmic Opticians (Optometrists)·2026

Search research articles

相关实验视频

Updated: Jan 8, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

使用基于 fundus imageomics 的机器学习模型预测近视风险.

Xiaoling Zhang^1,2, Zixun Wang², Jingtao Yu²

¹Handan Eye Hospital (The Third Hospital of Handan), Handan, Hebei, China.

Scientific reports

|December 12, 2025

概括

使用视网膜图像的机器学习模型可以预测儿童近视风险. 最好的模型确定了早期近视风险分层的视网膜关键特征.

关键词:

轴长/角膜曲率比 (AL/CR) 是指轴长与角膜曲率之间的比率.彩色底部摄影 (CFP) 是一种摄影技术.深度学习的图像组学.机器学习 (ML) 是指机器学习.近视风险预测和预测

更多相关视频

In vivo Structural Assessments of Ocular Disease in Rodent Models using Optical Coherence Tomography

In vivo Structural Assessments of Ocular Disease in Rodent Models using Optical Coherence Tomography

Published on: July 24, 2020

Inducement and Evaluation of a Murine Model of Experimental Myopia

Inducement and Evaluation of a Murine Model of Experimental Myopia

Published on: January 22, 2019

相关实验视频

Last Updated: Jan 8, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

In vivo Structural Assessments of Ocular Disease in Rodent Models using Optical Coherence Tomography

In vivo Structural Assessments of Ocular Disease in Rodent Models using Optical Coherence Tomography

Published on: July 24, 2020

Inducement and Evaluation of a Murine Model of Experimental Myopia

Inducement and Evaluation of a Murine Model of Experimental Myopia

Published on: January 22, 2019

科学领域:

眼科和计算机成像学
医疗保健中的人工智能

背景情况:

近视是一个日益严重的全球健康问题,特别是在学龄儿童中.
早期预测近视风险对于及时干预和管理至关重要.
从彩色底部摄影 (CFP) 中对视网膜特征的定量分析为风险评估提供了潜力.

研究的目的:

开发一种机器学习 (ML) 模型,使用CFP数据来预测儿童近视风险.
为了确定与近视进展相关的关键视网膜影像学特征.
评估各种ML算法对近视风险分层的预测性能.

主要方法:

一项涉及2184名6至10岁儿童的横截面研究,采用CFP数据.
使用EVisionAI平台提取146个视网膜图像学特征,以及年龄和性别.
通过LASSO回归和专家审查进行特征选择,然后使用RF,XGBoost和LightGBM进行模型构建.

主要成果:

随机森林 (RF) 模型实现了最高的预测性能 (AUC = 0.798),超过了LightGBM和XGBoost.
鉴定到的关键预测因素包括年龄,鼻盘-鼻距离,缩区域和血管参数.
射频模型表现出高特异性 (0.80) 和中度灵敏性 (0.59),并进行了强大的校准.

结论:

量化CFP衍生的图像学结合ML可以有效地预测学龄儿童近视风险.
开发的RF模型利用年龄,视网膜距离和血管特征,显示出显著的临床价值.
这种方法为早期近视风险分层和个性化管理策略提供了有前途的工具.