Development of machine learning models to predict posterior capsule rupture based on the EUREQUO registry

Triepels, Ron J. M. A. (author): Department of Data Analytics and Digitalisation, Maastricht University, Maastricht, Netherlands

Segers, Maartje H. M. (author): University Eye Clinic, Maastricht University Medical Center+, Maastricht, Netherlands,Yukioka Hospital

Rosen, Paul (author): Department of Ophthalmology, Oxford Eye Hospital, Oxford, United Kingdom

Nuijts, Rudy M. M. A. (author): University Eye Clinic, Maastricht University Medical Center+, Maastricht, Netherlands,Yukioka Hospital

van den Biggelaar, Frank J. H. M. (author): University Eye Clinic, Maastricht University Medical Center+, Maastricht, Netherlands,Yukioka Hospital

Henry, Ype P. (author): Department of Ophthalmology, Amsterdam UMC, Amsterdam, Netherlands,Amsterdam UMC - Vrije Universiteit Amsterdam

Stenevi, Ulf (author): Department of Ophthalmology, Sahlgrenska University Hospital, Göteborg, Sweden

Tassignon, Marie-José (author): Department of Ophthalmology, Antwerp University Hospital, Edegem, Belgium

Young, David (author): Department of Mathematics and Statistics, University of Strathclyde, Glasgow, United Kingdom

Lundström, Mats (author): Lund University,Lunds universitet,Oftalmologi, Lund,Sektion IV,Institutionen för kliniska vetenskaper, Lund,Medicinska fakulteten,Ophthalmology, Lund,Section IV,Department of Clinical Sciences, Lund,Faculty of Medicine

Dickman, Mor M. (author): University Eye Clinic, Maastricht University Medical Center+, Maastricht, Netherlands,Yukioka Hospital

(creator_code:org_t)

2023-02-15
2023
English.
In: Acta Ophthalmologica. - : John Wiley & Sons. - 1755-375X .- 1755-3768. ; 101:6, s. 644-650

Related links:: https://doi.org/10.1...; show more...; https://umu.diva-por... (primary) (Raw object); http://dx.doi.org/10... (free); https://urn.kb.se/re...; https://doi.org/10.1...; https://lup.lub.lu.s...; show less...

Abstract Subject headings

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.

MEDICIN OCH HÄLSOVETENSKAP -- Klinisk medicin -- Oftalmologi (hsv//swe)
MEDICAL AND HEALTH SCIENCES -- Clinical Medicine -- Ophthalmology (hsv//eng)

By the author/editor: Triepels, Ron J. ...; Segers, Maartje ...; Rosen, Paul; Nuijts, Rudy M. ...; van den Biggelaa ...; Henry, Ype P.; show more...; Stenevi, Ulf; Tassignon, Marie ...; Young, David; Behndig, Anders; Lundström, Mats; Dickman, Mor M.; show less...

About the subject

MEDICAL AND HEALTH SCIENCES: MEDICAL AND HEAL ...; and Clinical Medicin ...; and Ophthalmology

Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.