圆锥角膜自动筛选的高效堆叠集成学习模型。

IF 4 1区 医学 Q1 OPHTHALMOLOGY
Zahra J Muhsin, Rami Qahwaji, Ibrahim Ghafir, Mo'ath AlShawabkeh, Muawyah Al Bdour, Saif Aldeen AlRyalat, Majid Al-Taee
{"title":"圆锥角膜自动筛选的高效堆叠集成学习模型。","authors":"Zahra J Muhsin, Rami Qahwaji, Ibrahim Ghafir, Mo'ath AlShawabkeh, Muawyah Al Bdour, Saif Aldeen AlRyalat, Majid Al-Taee","doi":"10.1186/s40662-025-00440-6","DOIUrl":null,"url":null,"abstract":"<p><strong>Background: </strong>Despite extensive research on keratoconus (KC) detection with traditional machine learning models, stacking ensemble learning approaches remain underexplored. This paper presents a stacking ensemble learning method to enhance automated KC screening.</p><p><strong>Methods: </strong>This study utilizes a clinical dataset containing detailed corneal data from 2491 cases classified as non-KC (NKC), subclinical KC (SCKC) and clinical KC (CKC). Each cornea is represented by 79 features extracted from Pentacam imaging. Following extensive pre-processing, key corneal features that are strongly correlated with the target diagnosis are identified. These features are the keratometry of the steepest anterior point, surface variance index, vertical asymmetry index, height decentration index, and height asymmetry index. A novel stacking ensemble model is developed using the selected features to improve corneal classification into NKC, SCKC, and CKC by integrating top tree-based classifiers (random forest, gradient boosting, decision trees) with a support vector machine meta-classifier.</p><p><strong>Results: </strong>The pre-processing and feature selection techniques reduced the model's parameters to just 6.33% of the original dataset, improving classification performance, and cutting over 85% of the training time. The performance of the developed model was validated and tested on unseen data. Experimental results showed that the model outperforms existing studies, achieving 99.72% accuracy, precision, sensitivity, F1, and F2 scores, with a Matthews correlation coefficient of 0.995. It accurately classified all NKC and CKC cases, with just one misclassification involving an SCKC case. The model also demonstrated consistent performance on 100 additional unseen test cases, underscoring its generalizability and robustness in KC screening.</p><p><strong>Conclusions: </strong>By combining the strengths of diverse base models and key Pentacam indices, the stacking ensemble approach ensures reliable, accurate KC screening, providing clinicians with an automated tool for early detection and better patient management.</p>","PeriodicalId":12194,"journal":{"name":"Eye and Vision","volume":"12 1","pages":"25"},"PeriodicalIF":4.0000,"publicationDate":"2025-06-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12186405/pdf/","citationCount":"0","resultStr":"{\"title\":\"Highly efficient stacking ensemble learning model for automated keratoconus screening.\",\"authors\":\"Zahra J Muhsin, Rami Qahwaji, Ibrahim Ghafir, Mo'ath AlShawabkeh, Muawyah Al Bdour, Saif Aldeen AlRyalat, Majid Al-Taee\",\"doi\":\"10.1186/s40662-025-00440-6\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Background: </strong>Despite extensive research on keratoconus (KC) detection with traditional machine learning models, stacking ensemble learning approaches remain underexplored. This paper presents a stacking ensemble learning method to enhance automated KC screening.</p><p><strong>Methods: </strong>This study utilizes a clinical dataset containing detailed corneal data from 2491 cases classified as non-KC (NKC), subclinical KC (SCKC) and clinical KC (CKC). Each cornea is represented by 79 features extracted from Pentacam imaging. Following extensive pre-processing, key corneal features that are strongly correlated with the target diagnosis are identified. These features are the keratometry of the steepest anterior point, surface variance index, vertical asymmetry index, height decentration index, and height asymmetry index. A novel stacking ensemble model is developed using the selected features to improve corneal classification into NKC, SCKC, and CKC by integrating top tree-based classifiers (random forest, gradient boosting, decision trees) with a support vector machine meta-classifier.</p><p><strong>Results: </strong>The pre-processing and feature selection techniques reduced the model's parameters to just 6.33% of the original dataset, improving classification performance, and cutting over 85% of the training time. The performance of the developed model was validated and tested on unseen data. Experimental results showed that the model outperforms existing studies, achieving 99.72% accuracy, precision, sensitivity, F1, and F2 scores, with a Matthews correlation coefficient of 0.995. It accurately classified all NKC and CKC cases, with just one misclassification involving an SCKC case. The model also demonstrated consistent performance on 100 additional unseen test cases, underscoring its generalizability and robustness in KC screening.</p><p><strong>Conclusions: </strong>By combining the strengths of diverse base models and key Pentacam indices, the stacking ensemble approach ensures reliable, accurate KC screening, providing clinicians with an automated tool for early detection and better patient management.</p>\",\"PeriodicalId\":12194,\"journal\":{\"name\":\"Eye and Vision\",\"volume\":\"12 1\",\"pages\":\"25\"},\"PeriodicalIF\":4.0000,\"publicationDate\":\"2025-06-24\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12186405/pdf/\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Eye and Vision\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1186/s40662-025-00440-6\",\"RegionNum\":1,\"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":"Eye and Vision","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1186/s40662-025-00440-6","RegionNum":1,"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

背景:尽管使用传统的机器学习模型对圆锥角膜(KC)检测进行了广泛的研究,但堆叠集成学习方法仍未得到充分的探索。提出了一种提高KC自动筛选的叠加集成学习方法。方法:本研究使用了包含2491例非KC (NKC)、亚临床KC (SCKC)和临床KC (CKC)患者详细角膜数据的临床数据集。每个角膜由Pentacam成像提取的79个特征表示。经过广泛的预处理,确定了与目标诊断密切相关的关键角膜特征。这些特征是前最陡点的角化测量、表面方差指数、垂直不对称指数、高度分散指数和高度不对称指数。通过将基于顶树的分类器(随机森林、梯度增强、决策树)与支持向量机元分类器相结合,建立了一种新的层叠集成模型,利用所选特征将角膜分类为NKC、SCKC和CKC。结果:预处理和特征选择技术将模型的参数降低到原始数据集的6.33%,提高了分类性能,减少了85%以上的训练时间。在未知数据上验证了所开发模型的性能。实验结果表明,该模型优于已有研究,准确率、精密度、灵敏度、F1和F2得分均达到99.72%,马修斯相关系数为0.995。它准确地分类了所有NKC和CKC病例,只有一个错误分类涉及SCKC病例。该模型还在100个额外的未见过的测试用例上展示了一致的性能,强调了它在KC筛选中的泛化性和健壮性。结论:通过结合不同基础模型和Pentacam关键指数的优势,堆叠集成方法确保可靠,准确的KC筛查,为临床医生提供早期发现和更好的患者管理的自动化工具。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Highly efficient stacking ensemble learning model for automated keratoconus screening.

Background: Despite extensive research on keratoconus (KC) detection with traditional machine learning models, stacking ensemble learning approaches remain underexplored. This paper presents a stacking ensemble learning method to enhance automated KC screening.

Methods: This study utilizes a clinical dataset containing detailed corneal data from 2491 cases classified as non-KC (NKC), subclinical KC (SCKC) and clinical KC (CKC). Each cornea is represented by 79 features extracted from Pentacam imaging. Following extensive pre-processing, key corneal features that are strongly correlated with the target diagnosis are identified. These features are the keratometry of the steepest anterior point, surface variance index, vertical asymmetry index, height decentration index, and height asymmetry index. A novel stacking ensemble model is developed using the selected features to improve corneal classification into NKC, SCKC, and CKC by integrating top tree-based classifiers (random forest, gradient boosting, decision trees) with a support vector machine meta-classifier.

Results: The pre-processing and feature selection techniques reduced the model's parameters to just 6.33% of the original dataset, improving classification performance, and cutting over 85% of the training time. The performance of the developed model was validated and tested on unseen data. Experimental results showed that the model outperforms existing studies, achieving 99.72% accuracy, precision, sensitivity, F1, and F2 scores, with a Matthews correlation coefficient of 0.995. It accurately classified all NKC and CKC cases, with just one misclassification involving an SCKC case. The model also demonstrated consistent performance on 100 additional unseen test cases, underscoring its generalizability and robustness in KC screening.

Conclusions: By combining the strengths of diverse base models and key Pentacam indices, the stacking ensemble approach ensures reliable, accurate KC screening, providing clinicians with an automated tool for early detection and better patient management.

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
Eye and Vision
Eye and Vision OPHTHALMOLOGY-
CiteScore
8.60
自引率
2.40%
发文量
89
审稿时长
15 weeks
期刊介绍: Eye and Vision is an open access, peer-reviewed journal for ophthalmologists and visual science specialists. It welcomes research articles, reviews, methodologies, commentaries, case reports, perspectives and short reports encompassing all aspects of eye and vision. Topics of interest include but are not limited to: current developments of theoretical, experimental and clinical investigations in ophthalmology, optometry and vision science which focus on novel and high-impact findings on central issues pertaining to biology, pathophysiology and etiology of eye diseases as well as advances in diagnostic techniques, surgical treatment, instrument updates, the latest drug findings, results of clinical trials and research findings. It aims to provide ophthalmologists and visual science specialists with the latest developments in theoretical, experimental and clinical investigations in eye and vision.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信