Review article: Myopia - Genetically inherited or environmental influences

Tjoa, Stanley Evander Emeltan; Putra, S E D

doi:10.1088/1755-1315/293/1/012037

Cited by 3 publications

(1 citation statement)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Usually light entering the eyes of myopic patients is not focused on the retina, but in front of the retina, which is also known as a refractive error [5][6].The harm of myopia is not as simple as blurred vision. It is an important factor in diseases such as retinal detachment, macular degeneration and glaucoma.…”

Section: Introductionmentioning

confidence: 99%

Application of machine learning method in clinical fitting of Ortho-K lens for myopia correction

Gong

et al. 2022

Preprint

View full text Add to dashboard Cite

PurposeAiming at the problem that the complex structure of Ortho-K lens makes the clinical fitting process cumbersome, time-consuming and labor-intensive, a machine learning method was proposed to predict the Alignment Curve (AC) and Target Power (TP) of orthokeratology in vision shaping treatment (VST) to assist in fitting.MethodsThe clinical data of 1235 patients were selected in this study. The AC parameters and TP parameters of the Ortho-K lens worn by the patients were used as target predictors, respectively. A feature engineering study was performed on the input variables to construct the most appropriate feature input dimension and type, and four machine learning algorithms (including Random Forest, Gaussian Process, Support Vector Machine, and Linear Regression) were used to predict AC and TP, respectively. Then optimize the hyper-parameters of the machine learning model to achieve the best predictive performance of the machine learning model.ResultsGaussian Process and Random Forest performed best in predicting AC and TP, with R2 of 0.84 and 0.88, respectively. During the testing process of the test set, the prediction results of the best machine learning model showed a high degree of consistency with the traditional clinical diagnosis results.ConclusionThe machine learning predictive diagnosis method proposed in this study is significantly more efficient than other auxiliary diagnosis methods. The predictive ability of the machine learning model reduces the number of times the patient needs to replace the test piece, which greatly improves the success rate and diagnostic efficiency of the test piece.

show abstract

Section: Introductionmentioning

confidence: 99%