• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利:CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验

基于时间感知的深度学习的儿童及青少年近视预测。

Myopia prediction for children and adolescents via time-aware deep learning.

机构信息

Big Data Research Center, University of Electronic Science and Technology of China, Chengdu, 611731, People's Republic of China.

Department of Ophthalmology, West China Hospital, Sichuan University, Chengdu, 610041, People's Republic of China.

出版信息

Sci Rep. 2023 Apr 3;13(1):5430. doi: 10.1038/s41598-023-32367-0.

DOI:10.1038/s41598-023-32367-0
PMID:37012269
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC10070443/
Abstract

This is a retrospective analysis. Quantitative prediction of the children's and adolescents' spherical equivalent based on their variable-length historical vision records. From October 2019 to March 2022, we examined uncorrected visual acuity, sphere, astigmatism, axis, corneal curvature and axial length of 75,172 eyes from 37,586 children and adolescents aged 6-20 years in Chengdu, China. 80% samples consist of the training set, the 10% form the validation set and the remaining 10% form the testing set. Time-Aware Long Short-Term Memory was used to quantitatively predict the children's and adolescents' spherical equivalent within two and a half years. The mean absolute prediction error on the testing set was 0.103 ± 0.140 (D) for spherical equivalent, ranging from 0.040 ± 0.050 (D) to 0.187 ± 0.168 (D) if we consider different lengths of historical records and different prediction durations. Time-Aware Long Short-Term Memory was applied to captured the temporal features in irregularly sampled time series, which is more in line with the characteristics of real data and thus has higher applicability, and helps to identify the progression of myopia earlier. The overall error 0.103 (D) is much smaller than the criterion for clinically acceptable prediction, say 0.75 (D).

摘要

这是一项回顾性分析。基于儿童可变长度的历史视力记录,对其球镜等效值进行定量预测。2019 年 10 月至 2022 年 3 月,我们检查了来自中国成都 37586 名 6-20 岁儿童和青少年的 75172 只未矫正视力、球镜、散光、轴位、角膜曲率和眼轴。80%的样本由训练集组成,10%来自验证集,其余 10%来自测试集。使用时间感知长短期记忆模型定量预测儿童和青少年未来两年半的球镜等效值。在测试集上,球镜等效值的平均绝对预测误差为 0.103 ± 0.140(D),范围为 0.040 ± 0.050(D)至 0.187 ± 0.168(D),具体取决于历史记录的长度和预测持续时间的不同。时间感知长短期记忆模型用于捕获不规则采样时间序列中的时间特征,这更符合真实数据的特点,因此具有更高的适用性,并有助于更早识别近视的进展。整体误差 0.103(D)远小于临床可接受预测的标准,例如 0.75(D)。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d482/10070443/81dbc2059a70/41598_2023_32367_Fig4_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d482/10070443/715710239d43/41598_2023_32367_Fig1_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d482/10070443/0e23eba1ac8e/41598_2023_32367_Fig2_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d482/10070443/daa8d94bc1b5/41598_2023_32367_Fig3_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d482/10070443/81dbc2059a70/41598_2023_32367_Fig4_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d482/10070443/715710239d43/41598_2023_32367_Fig1_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d482/10070443/0e23eba1ac8e/41598_2023_32367_Fig2_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d482/10070443/daa8d94bc1b5/41598_2023_32367_Fig3_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d482/10070443/81dbc2059a70/41598_2023_32367_Fig4_HTML.jpg

相似文献

1
Myopia prediction for children and adolescents via time-aware deep learning.基于时间感知的深度学习的儿童及青少年近视预测。
Sci Rep. 2023 Apr 3;13(1):5430. doi: 10.1038/s41598-023-32367-0.
2
Phakic intraocular lenses for the treatment of refractive errors: an evidence-based analysis.用于治疗屈光不正的有晶状体眼人工晶状体:一项基于证据的分析。
Ont Health Technol Assess Ser. 2009;9(14):1-120. Epub 2009 Oct 1.
3
A Novel Time-Aware Deep Learning Model Predicting Myopia in Children and Adolescents.一种预测儿童和青少年近视的新型时间感知深度学习模型。
Ophthalmol Sci. 2024 Jun 13;4(6):100563. doi: 10.1016/j.xops.2024.100563. eCollection 2024 Nov-Dec.
4
Laser in situ keratomileusis for myopia and hyperopia using the Lasersight 200 laser in 300 consecutive eyes.使用Lasersight 200激光对300只连续的眼睛进行近视和远视的准分子原位角膜磨镶术。
J Refract Surg. 2000 Nov-Dec;16(6):716-23. doi: 10.3928/1081-597X-20001101-07.
5
Prediction of myopia development among Chinese school-aged children using refraction data from electronic medical records: A retrospective, multicentre machine learning study.基于电子病历中的屈光数据预测中国学龄儿童近视进展:一项回顾性、多中心机器学习研究。
PLoS Med. 2018 Nov 6;15(11):e1002674. doi: 10.1371/journal.pmed.1002674. eCollection 2018 Nov.
6
Laser in situ keratomileusis for myopia from -5.50 to -11.50 diopters with astigmatism.用于治疗-5.50至-11.50屈光度近视合并散光的准分子原位角膜磨镶术
J Refract Surg. 1998 Jan-Feb;14(1):19-25. doi: 10.3928/1081-597X-19980101-06.
7
Distribution, Progression, and Associated Factors of Refractive Status of Children in Lhasa, Tibet, after COVID-19 Quarantine.拉萨市新冠疫情隔离后儿童屈光状态的分布、进展及相关因素。
Ophthalmic Res. 2022;65(3):321-327. doi: 10.1159/000522548. Epub 2022 Feb 16.
8
Intraocular lens implantation and laser in situ keratomileusis (bioptics) to correct high myopia and hyperopia with astigmatism.人工晶状体植入术和准分子原位角膜磨镶术(联合手术)矫正高度近视和远视合并散光。
J Refract Surg. 2001 Mar-Apr;17(2 Suppl):S234-7. doi: 10.3928/1081-597X-20010302-19.
9
Excimer laser in situ keratomileusis to correct compound myopic astigmatism.准分子激光原位角膜磨镶术矫正复合性近视散光。
J Refract Surg. 1997 Sep-Oct;13(6):511-20. doi: 10.3928/1081-597X-19970901-07.
10
[Intraocular lens power calculation for high myopic eyes with cataract: comparison of three formulas].[白内障高度近视眼人工晶状体屈光度计算:三种公式的比较]
Zhonghua Yan Ke Za Zhi. 2017 Apr 11;53(4):260-265. doi: 10.3760/cma.j.issn.0412-4081.2017.04.007.

引用本文的文献

1
Influence of dynamic changes of ocular biometric parameters on new-onset myopia in Chinese children: a 4-year cohort study.中国儿童眼部生物测量参数动态变化对新发近视的影响:一项4年队列研究
Sci Rep. 2025 Aug 4;15(1):28474. doi: 10.1038/s41598-025-14453-7.
2
Application of artificial intelligence in myopia prevention and control.人工智能在近视防控中的应用。
Pediatr Investig. 2025 Mar 18;9(2):114-124. doi: 10.1002/ped4.70001. eCollection 2025 Jun.
3
Interpretable machine learning models for predicting childhood myopia from school-based screening data.

本文引用的文献

1
Machine learning predicting myopic regression after corneal refractive surgery using preoperative data and fundus photography.使用术前数据和眼底摄影通过机器学习预测角膜屈光手术后的近视回退
Graefes Arch Clin Exp Ophthalmol. 2022 Nov;260(11):3701-3710. doi: 10.1007/s00417-022-05738-y. Epub 2022 Jun 24.
2
Myopia prediction: a systematic review.近视预测:系统综述。
Eye (Lond). 2022 May;36(5):921-929. doi: 10.1038/s41433-021-01805-6. Epub 2021 Oct 13.
3
IMI Impact of Myopia.近视的IMI影响
基于学校筛查数据预测儿童近视的可解释机器学习模型。
Sci Rep. 2025 Jun 5;15(1):19811. doi: 10.1038/s41598-025-05021-0.
4
Artificial intelligence in ophthalmology: opportunities, challenges, and ethical considerations.眼科领域的人工智能:机遇、挑战与伦理考量。
Med Hypothesis Discov Innov Ophthalmol. 2025 May 10;14(1):255-272. doi: 10.51329/mehdiophthal1517. eCollection 2025 Spring.
5
Artificial intelligence technology in ophthalmology public health: current applications and future directions.眼科公共卫生中的人工智能技术:当前应用与未来方向。
Front Cell Dev Biol. 2025 Apr 17;13:1576465. doi: 10.3389/fcell.2025.1576465. eCollection 2025.
6
Artificial Intelligence in Optometry: Current and Future Perspectives.验光领域的人工智能:现状与未来展望
Clin Optom (Auckl). 2025 Mar 12;17:83-114. doi: 10.2147/OPTO.S494911. eCollection 2025.
7
A Novel Time-Aware Deep Learning Model Predicting Myopia in Children and Adolescents.一种预测儿童和青少年近视的新型时间感知深度学习模型。
Ophthalmol Sci. 2024 Jun 13;4(6):100563. doi: 10.1016/j.xops.2024.100563. eCollection 2024 Nov-Dec.
Invest Ophthalmol Vis Sci. 2021 Apr 28;62(5):2. doi: 10.1167/iovs.62.5.2.
4
Pay Attention to Evolution: Time Series Forecasting with Deep Graph-Evolution Learning.关注演化:基于深度图演化学习的时间序列预测
IEEE Trans Pattern Anal Mach Intell. 2021 Apr 27;PP. doi: 10.1109/TPAMI.2021.3076155.
5
Predicting COVID-19 disease progression and patient outcomes based on temporal deep learning.基于时间深度学习的 COVID-19 疾病进展和患者预后预测。
BMC Med Inform Decis Mak. 2021 Feb 8;21(1):45. doi: 10.1186/s12911-020-01359-9.
6
Annual Myopia Progression and Subsequent 2-Year Myopia Progression in Singaporean Children.新加坡儿童的年度近视进展和随后的 2 年近视进展。
Transl Vis Sci Technol. 2020 Dec 7;9(13):12. doi: 10.1167/tvst.9.13.12. eCollection 2020 Dec.
7
Association of Age at Myopia Onset With Risk of High Myopia in Adulthood in a 12-Year Follow-up of a Chinese Cohort.近视发病年龄与中国队列人群 12 年随访中成年高度近视风险的相关性研究。
JAMA Ophthalmol. 2020 Nov 1;138(11):1129-1134. doi: 10.1001/jamaophthalmol.2020.3451.
8
Machine learning approaches to drug response prediction: challenges and recent progress.用于药物反应预测的机器学习方法:挑战与最新进展。
NPJ Precis Oncol. 2020 Jun 15;4:19. doi: 10.1038/s41698-020-0122-1. eCollection 2020.
9
Deep Learning for Improved Risk Prediction in Surgical Outcomes.深度学习在手术结局风险预测中的应用。
Sci Rep. 2020 Jun 9;10(1):9289. doi: 10.1038/s41598-020-62971-3.
10
Prediction of Myopia in Adolescents through Machine Learning Methods.通过机器学习方法预测青少年近视
Int J Environ Res Public Health. 2020 Jan 10;17(2):463. doi: 10.3390/ijerph17020463.