EUVIMEDEUVIMED entdecken
Multizentrische StudieKohortenstudie

Predicting Early Keratoconus Progression Using Biomechanics via Multi‐Machine Learning: A Multicenter 2‐Year Prospective Cohort Study

Yan Huo iD, Ruisi Xie, Zhengyuan Qu, Yutong Li, Shufan Wang, Jing Li, Jie Hou, Haohan Zou iD, Xiaona Li, Yan Wang iD

Clinical & Experimental Ophthalmology · 2026-05 · 54 (4) · 442-451

DOI: 10.1111/ceo.70076 ↗

Kurz gefasst

Multizentrische Studie in der Zeitschrift Clinical & Experimental Ophthalmology, online erschienen am 11.02.2026. Verfasst von Yan Huo und 9 weiteren Autor:innen. Beteiligte Einrichtung: School of Medicine, Nankai University, Tianjin, China. Hauptthemen laut MEDLINE: Keratoconus, Machine Learning, Cornea. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 34 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Tianjin Key Medical Discipline Construction, National Program on Key Research Project of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

This study aimed to develop and validate a biomechanical early progression index (BEPI) based on machine learning algorithms, intended to predict early keratoconus progression at the first visit.

Methods

This multicenter prospective cohort study recruited 247 eyes of 247 participants from 3 centers; 225 completed all visits and were divided into a training and validation cohort. Corvis ST and Pentacam were used to obtain biomechanical and tomographic data. Based on disease progression during the follow‐up period, predictive models were constructed using biomechanical data by applying Random Forest, XGBoost, Naïve Bayes classifier, K‐Nearest Neighbours and Support Vector Machine.

Results

In the training cohort ( n = 157), 66 eyes remained stable after a mean period of 25.18 ± 4.17 months, whereas 91 eyes demonstrated progression after 6.74 ± 2.53 months. The random forest provided the highest predictive performance and was designated as BEPI. With a cutoff value of 0.528, BEPI achieved 0.989 sensitivity and 0.985 specificity in distinguishing between stable and progressive cases. In the external validation cohort ( n = 68), 31 eyes remained stable after 23.16 ± 6.75 months and 37 eyes progressed after 7.03 ± 3.42 months of follow‐up. BEPI demonstrated the highest predictive performance, with an accuracy of 0.941 and an F1‐score of 0.947.

Conclusion

The machine learning‐based index BEPI derived from corneal biomechanics enabled the accurate prediction of early keratoconus progression at the first visit, with an accuracy of 0.941. BEPI provides clinicians with a novel index for the accurate evaluation of early keratoconus progression, facilitating personalised clinical decision‐making to preserve vision and improve prognosis.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

KeratoconusMachine LearningCorneaHumansProspective StudiesDisease ProgressionFemaleBiomechanical PhenomenaMaleAdultCorneal TopographyPrediction AlgorithmsYoung AdultFollow-Up StudiesPredictive Learning ModelsRandom ForestBoosting Machine Learning AlgorithmsCorneal Pachymetry

Stichwörter der Autor:innen

biomechanicsdisease progressionearly keratoconusmachine learningprognosis

Autor:innen und Institutionen

  • School of Medicine, Nankai University, Tianjin, China.
  • Clinical College of Ophthalmology, Tianjin Medical University, Tianjin, China.
  • Xi'an People's Hospital, Xi'an, Shaanxi, China.
  • Jinan Mingshui Eye Hospital, Ji'nan, Shandong, China.
  • Tianjin Eye Hospital, Tianjin Key Lab of Ophthalmology and Visual Science, Tianjin Eye Institute, Nankai University Affiliated Eye Hospital, Tianjin, China.
  • Nankai University Eye Institute, Nankai University, Tianjin, China.

ORCID-Profile: Yan Huo, Haohan Zou, Yan Wang

Förderung

  • National Natural Science Foundation of China
  • Tianjin Key Medical Discipline Construction
  • National Program on Key Research Project of China

Zitieren

APA

Huo, Y., Xie, R., Qu, Z., Li, Y., Wang, S., Li, J., Hou, J., Zou, H., Li, X., & Wang, Y. (2026). Predicting Early Keratoconus Progression Using Biomechanics via Multi‐Machine Learning: A Multicenter 2‐Year Prospective Cohort Study. Clinical & Experimental Ophthalmology, 54(4), 442-451. https://doi.org/10.1111/ceo.70076

Vancouver

Huo Y, Xie R, Qu Z, Li Y, Wang S, Li J, et al. Predicting Early Keratoconus Progression Using Biomechanics via Multi‐Machine Learning: A Multicenter 2‐Year Prospective Cohort Study. Clinical & Experimental Ophthalmology. 2026;54(4):442-451. doi:10.1111/ceo.70076

BibTeX
@article{Huo2026,
  author = {Huo, Yan and Xie, Ruisi and Qu, Zhengyuan and Li, Yutong and Wang, Shufan and Li, Jing and Hou, Jie and Zou, Haohan and Li, Xiaona and Wang, Yan},
  title = {Predicting Early Keratoconus Progression Using Biomechanics via Multi‐Machine Learning: A Multicenter 2‐Year Prospective Cohort Study},
  journal = {Clinical & Experimental Ophthalmology},
  year = {2026},
  volume = {54},
  number = {4},
  pages = {442--451},
  doi = {10.1111/ceo.70076}
}
Kostenlos starten →