Development of machine learning models to predict posterior capsule rupture based on the EUREQUO registry
Acta Ophthalmologica · 2023-09 · 101 (6) · 644-650
DOI: 10.1111/aos.15648 ↗Kurz gefasst
Fachartikel in der Zeitschrift Acta Ophthalmologica, online erschienen am 15.02.2023. Verfasst von Ron J M A Triepels und 11 weiteren Autor:innen. Beteiligte Einrichtung: Department of Data Analytics and Digitalisation, Maastricht University, Maastricht, the Netherlands. Hauptthemen laut MEDLINE: Cataract. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
Purpose
To evaluate the performance of different probabilistic classifiers to predict posterior capsule rupture (PCR) prior to cataract surgery.
Methods
Three probabilistic classifiers were constructed to estimate the probability of PCR: a Bayesian network (BN), logistic regression (LR) model, and multi‐layer perceptron (MLP) network. The classifiers were trained on a sample of 2 853 376 surgeries reported to the European Registry of Quality Outcomes for Cataract and Refractive Surgery (EUREQUO) between 2008 and 2018. The performance of the classifiers was evaluated based on the area under the precision‐recall curve (AUPRC) and compared to existing scoring models in the literature. Furthermore, direct risk factors for PCR were identified by analysing the independence structure of the BN.
Results
The MLP network predicted PCR overall the best (AUPRC 13.1 ± 0.41%), followed by the BN (AUPRC 8.05 ± 0.39%) and the LR model (AUPRC 7.31 ± 0.15%). Direct risk factors for PCR include preoperative best‐corrected visual acuity (BCVA), year of surgery, operation type, anaesthesia, target refraction, other ocular comorbidities, white cataract, and corneal opacities.
Conclusions
Our results suggest that the MLP network performs better than existing scoring models in the literature, despite a relatively low precision at high recall. Consequently, implementing the MLP network in clinical practice can potentially decrease the PCR rate.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Data Analytics and Digitalisation, Maastricht University, Maastricht, the Netherlands.
- University Eye Clinic, Maastricht University Medical Center+, Maastricht, the Netherlands.
- Department of Ophthalmology, Oxford Eye Hospital, Oxford, UK.
- Department of Ophthalmology, Amsterdam UMC, Amsterdam, the Netherlands.
- Department of Ophthalmology, Sahlgrenska University Hospital, Göteborg, Sweden.
- Department of Ophthalmology, Antwerp University Hospital, Edegem, Belgium.
ORCID-Profile: Ron J M A Triepels, Maartje H M Segers, Anders Behndig, Mats Lundström, Mor M Dickman
Zitieren
APA
Triepels, R. J. M. A., Segers, M. H. M., Rosen, P., Nuijts, R. M. M. A., van den Biggelaar, F. J. H. M., Henry, Y. P., Stenevi, U., Tassignon, M. J., Young, D., Behndig, A., Lundström, M., & Dickman, M. M. (2023). Development of machine learning models to predict posterior capsule rupture based on the EUREQUO registry. Acta Ophthalmologica, 101(6), 644-650. https://doi.org/10.1111/aos.15648
Vancouver
Triepels RJMA, Segers MHM, Rosen P, Nuijts RMMA, van den Biggelaar FJHM, Henry YP, et al. Development of machine learning models to predict posterior capsule rupture based on the EUREQUO registry. Acta Ophthalmologica. 2023;101(6):644-650. doi:10.1111/aos.15648
BibTeX
@article{Triepels2023,
author = {Triepels, Ron J M A and Segers, Maartje H M and Rosen, Paul and Nuijts, Rudy M M A and van den Biggelaar, Frank J H M and Henry, Ype P and Stenevi, Ulf and Tassignon, Marie-José and Young, David and Behndig, Anders and Lundström, Mats and Dickman, Mor M},
title = {Development of machine learning models to predict posterior capsule rupture based on the EUREQUO registry},
journal = {Acta Ophthalmologica},
year = {2023},
volume = {101},
number = {6},
pages = {644--650},
doi = {10.1111/aos.15648}
}
EUVIMED