TY - JOUR
T1 - Development of machine learning models to predict posterior capsule rupture based on the EUREQUO registry
AU - Triepels, Ron J. M. A.
AU - Segers, Maartje H. M.
AU - Rosen, Paul
AU - Nuijts, Rudy M. M. A.
AU - van den Biggelaar, Frank J. H. M.
AU - Henry, Ype P.
AU - Stenevi, Ulf
AU - Tassignon, Marie‐José
AU - Young, David
AU - Behndig, Anders
AU - Lundström, Mats
AU - Dickman, Mor M.
PY - 2023/2/15
Y1 - 2023/2/15
N2 - 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.
AB - 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.
KW - artificial intelligence
KW - Bayesian network
KW - cataract surgery
KW - logistic regression
KW - machine learning
KW - multi‐layer perceptron
KW - posterior capsule rupture
U2 - 10.1111/aos.15648
DO - 10.1111/aos.15648
M3 - Article
SN - 1755-375X
JO - Acta Ophthalmologica
JF - Acta Ophthalmologica
ER -