MedPath

Validation of a Predictive Algorithm to Determine the Effectiveness of Orthokeratology for Myopia Control

Conditions
Myopia
Interventions
Device: Orthokeratology lenses
Registration Number
NCT04275635
Lead Sponsor
Sun Yat-sen University
Brief Summary

This is a prospective study to validate a predictive algorithm for identifying fast progressing myopes.

Detailed Description

Orthokeratology (ortho-K) has been demonstrated to slow myopic progression and reduce axial elongation in young patients, but this treatment is limited by the need for contact lens wear, which is the common cause for keratitis in children, and therefore cautious use is recommended. There is a need to identify the patients that could benefit most from this treatment. In order to do so, we conduct a retrospective study and create a large database (n = 10,000) of de-identified data to train an algorithm for identifying fast progressing myopes. In addition, we will perform a prospective study to validate this predictive algorithm and determine the effectiveness of Orthokeratology among different individual patients in China.

Recruitment & Eligibility

Status
UNKNOWN
Sex
All
Target Recruitment
3000
Inclusion Criteria
  • -6.0D≤SER≤-0.5D
  • Astigmatism≤2.0D
Exclusion Criteria
  • Contraindications of wearing Ortho-K.
  • Diagnosis of strabismus, amblyopia and other refractive development of the eye or systemic diseases.
  • Currently involved in other clinical studies.

Study & Design

Study Type
OBSERVATIONAL
Study Design
Not specified
Arm && Interventions
GroupInterventionDescription
Children with myopiaOrthokeratology lensesA total of 3,000 children from Zhongshan Ophthalmic Center is required to undergo ophthalmic examinations and complete questionnaires at baseline and 1yr after wearing Ortho-K.
Primary Outcome Measures
NameTimeMethod
AUROC of the prediction algorithm for identifying fast progressing myopes1 year

Age-specific axial length (AL) changes previously described by Wang et al.(IOVS, 52 (11), 7949-53, 2011) are used as cut-off values to determine whether a child is a fast progressor or not. A child whose AL change falls on or above the cut-off value is considered to be a fast progressor.

Secondary Outcome Measures
NameTimeMethod
Performance of an algorithm for predicting AL1 year

The investigators will use mean absolute error (MAE), R square to evaluate the performance.

Sensitivity and specificity of the prediction algorithm for identifying fast progressing myopes1 year

The investigators will estimate sensitivity and specificity of the predictive algorithm for identifying fast progressing myopes.

Performance of an algorithm for predicting spherical equivalent refractive error1 year

The investigators will use mean absolute error (MAE), R square to evaluate the performance.

Trial Locations

Locations (1)

Zhognshan Ophthalmic Center (Zhujiang New Town Branch), Sun Yat-sen University

🇨🇳

Guangzhou, Guangdong, China

© Copyright 2025. All Rights Reserved by MedPath