Shady T Awwad, Bassel Hammoud, Jad F Assaf, Lara Asroui, James Bradley Randleman, Cynthia J Roberts, Douglas D Koch, Jawad Kaisania, Carl-Joe Mehanna, Shadi Elbassuoni
{"title":"Thickness Speed Progression Index: Machine Learning Approach for Keratoconus Detection.","authors":"Shady T Awwad, Bassel Hammoud, Jad F Assaf, Lara Asroui, James Bradley Randleman, Cynthia J Roberts, Douglas D Koch, Jawad Kaisania, Carl-Joe Mehanna, Shadi Elbassuoni","doi":"10.1016/j.ajo.2024.11.011","DOIUrl":null,"url":null,"abstract":"<p><strong>Purpose: </strong>To develop and validate a pachymetry-based machine learning (ML) index for differentiating keratoconus, keratoconus suspect, and normal corneas.</p><p><strong>Design: </strong>Development and validation of an ML diagnostic algorithm.</p><p><strong>Methods: </strong>This retrospective study included 349 eyes of 349 patients with normal, frank keratoconus (KC), and KC suspect (KCS) corneas. KCS corneas included topographically/tomographically normal (TNF) and borderline fellow eyes (TBF) of patients with asymmetric KC. Six parameters were derived from the corneal thickness progression map on the Galilei Dual Scheimpflug-Placido system and fed into a machine-learning algorithm to create the Thickness Speed Progression Index. The model was trained with 5-fold cross-validation using a random search over 7 different ML algorithms, and the best model and hyperparameters were selected.</p><p><strong>Results: </strong>A total of 133 normal eyes, 141 KC eyes, and 75 KCS eyes, subdivided into 34 TNF and 41 TBF eyes, were included. In experiment 1 (normal and KC), the best model (Random Forest) achieved an accuracy of 100% and area under the receiver operating characteristic (AUROC) of 1.00 for both normal and KC groups. In experiment 2 (normal, KCS, and KC), the model achieved an overall accuracy of 91%, and AUROC curves of 0.93, 0.83, and 0.99 in detecting normal, KCS, and KC corneas respectively. In experiment 3 (normal, TNF, TBF, and KC), the model achieved an accuracy of 87% with AUROC curves of 0.91, 0.60, 0.77, and 0.94 for normal, TNF, TBF, and KC corneas, respectively.</p><p><strong>Conclusions: </strong>Using data solely based on pachymetry, ML algorithms such as the Thickness Speed Progression Index are able to discriminate normal corneas from KC and KCSs corneas with reasonable accuracy.</p>","PeriodicalId":7568,"journal":{"name":"American Journal of Ophthalmology","volume":" ","pages":"188-201"},"PeriodicalIF":4.1000,"publicationDate":"2024-11-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"American Journal of Ophthalmology","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1016/j.ajo.2024.11.011","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
Abstract
Purpose: To develop and validate a pachymetry-based machine learning (ML) index for differentiating keratoconus, keratoconus suspect, and normal corneas.
Design: Development and validation of an ML diagnostic algorithm.
Methods: This retrospective study included 349 eyes of 349 patients with normal, frank keratoconus (KC), and KC suspect (KCS) corneas. KCS corneas included topographically/tomographically normal (TNF) and borderline fellow eyes (TBF) of patients with asymmetric KC. Six parameters were derived from the corneal thickness progression map on the Galilei Dual Scheimpflug-Placido system and fed into a machine-learning algorithm to create the Thickness Speed Progression Index. The model was trained with 5-fold cross-validation using a random search over 7 different ML algorithms, and the best model and hyperparameters were selected.
Results: A total of 133 normal eyes, 141 KC eyes, and 75 KCS eyes, subdivided into 34 TNF and 41 TBF eyes, were included. In experiment 1 (normal and KC), the best model (Random Forest) achieved an accuracy of 100% and area under the receiver operating characteristic (AUROC) of 1.00 for both normal and KC groups. In experiment 2 (normal, KCS, and KC), the model achieved an overall accuracy of 91%, and AUROC curves of 0.93, 0.83, and 0.99 in detecting normal, KCS, and KC corneas respectively. In experiment 3 (normal, TNF, TBF, and KC), the model achieved an accuracy of 87% with AUROC curves of 0.91, 0.60, 0.77, and 0.94 for normal, TNF, TBF, and KC corneas, respectively.
Conclusions: Using data solely based on pachymetry, ML algorithms such as the Thickness Speed Progression Index are able to discriminate normal corneas from KC and KCSs corneas with reasonable accuracy.
期刊介绍:
The American Journal of Ophthalmology is a peer-reviewed, scientific publication that welcomes the submission of original, previously unpublished manuscripts directed to ophthalmologists and visual science specialists describing clinical investigations, clinical observations, and clinically relevant laboratory investigations. Published monthly since 1884, the full text of the American Journal of Ophthalmology and supplementary material are also presented online at www.AJO.com and on ScienceDirect.
The American Journal of Ophthalmology publishes Full-Length Articles, Perspectives, Editorials, Correspondences, Books Reports and Announcements. Brief Reports and Case Reports are no longer published. We recommend submitting Brief Reports and Case Reports to our companion publication, the American Journal of Ophthalmology Case Reports.
Manuscripts are accepted with the understanding that they have not been and will not be published elsewhere substantially in any format, and that there are no ethical problems with the content or data collection. Authors may be requested to produce the data upon which the manuscript is based and to answer expeditiously any questions about the manuscript or its authors.