(results will display both Free Papers & Poster)
New algorithm based on artificial intelligence to calculate the power of an intra-ocular lens (IOL) using the refractive error of the first operated eye
Session Details
Session Title: Cataract
Session Date/Time: Friday 21/02/2020 | 10:30-13:00
Paper Time: 12:36
Venue: Fes 1.
First Author: L.GAUTHIER FRANCE
Co Author(s): B. Victor
Abstract Details
Purpose:
To present a new approach to calculate the emmetropic power of an IOL based on multiparametric analyse of the the 4-8 week post-operative refractive error of the first operated eye
Setting:
Private practice – Ophthalmology department, Clinique Côte Basque Sud, Saint Jean de LUZ, France.
Methods:
A base of 534 bilateral operated cataract patients (1068 eyes) was established with known preoperative biometric data (iol master 500) and refractive result measured between 4 and 8 weeks after surgery.All previous operated eyes have been excluded from the study.
Based on the post-refractive results of two eyes of the same patient, we trained a machine learning algorithm to learn with several key variables (AL, KM, Age, Sexe, ACD...) to predict post-refractive result of the contralateral eye of the same patient in order to adjust the IOL power computed by the SRK-T formula.
Results:
The use of the algorithm increased significantly the accuracy of the IOL calculation.
The algorithm achieves 85% of eyes within +/- 0.50 D of the target refraction compared to 70% with SRKT, 96% of eyes within +/- 0.75 D of the target refraction compared to 95% with SRKT and 99% of eyes within +/- 1 D of the target refraction compared to 96% with SRKT.
The algorithm improves accuracy especially for eyes with extreme biometric a measurement.
According to the different parameters of the algorithm, the refractive result of the contralateral eye is the more correlated to refractive compensation achieved by the algorithm.
Conclusions:
The algorithm based on the refractive result of first operated eye increases dramatically the accuracy of the IOL calculation.
This algorithm is different from simple compensation of first eye refractive error proportion, as it involves multiple parameters to determine this compensation.
The time surgery between two eyes of same patients is currently around one week in routine surgery. In the cases of surgery designed to provide an early refractive result, the use of this algorithm offers an optimum guarantee of accuracy.
Financial Disclosure:
... gains financially from product or procedure presented, ... is employed by a for-profit company with an interest in the subject of the presentation, ... has significant investment interest in a company producing, developing or supplying product or procedure presented