Axial and choroidal changes as key predictors of myopia control in red-light therapy: Evidence from machine learning models
Hu X, Wang Z, Li R, Sun B, Wang D, Zong H, Zhang X, Wei R, Jie H
Photodiagnosis and photodynamic therapy · 1 cita
Etiquetas de revisión
Financiación no declaradaDatos neutrales que nuestra revisión registró sobre cómo se hizo este estudio. Describen el método, nunca si nos gusta el resultado.
Cómo se estudió
- Diseño
- Estudio de cohortes (indexado por PubMed)
- Estudiado en
- Personas
- Resultado principal
- Marcadores de salud y funcionamiento
Quién lo pagó
- Financiación
- Financiación no declarada
Publicación
- Publicado
- 2026-01-08 · Photodiagnosis Photodyn Ther · vol. 57 · pág. 105342
- Editorial
- Elsevier BV
- Citado
- 3 citas · más que el 99% de estudios similares · 16.4× el promedio de su campo
- Impacto
- Entre el 10 % más citado de su campo
- Referencias
- 46 trabajos
- Acceso
- Acceso abierto (revista) · CC-BY-NC-ND
- Áreas de investigación
- Ophthalmology and Visual Impairment Studies · Retinal Diseases and Treatments · Corneal surgery and disorders
- Palabras clave
- Key (lock), Control (management), Foundation (evidence), Deep learning
- MeSH
- choroid, humans, myopia, sensitivity and specificity, cohort studies, adolescent, child, female, male, axial length, eye, machine learning, red light, boosting machine learning algorithms, predictive learning models
9 autores
- Xiaoxue Hu🇨🇳Wuhan Children's Hospital; Huazhong University of Science and Technology
- Zixun Wang🇨🇳Tianjin Medical University Eye Hospital; Tianjin Medical University
- Rui Li🇨🇳The Central Hospital of Xiao gan
- Boxuan Sun🇨🇳Tianjin Medical University Eye Hospital; Tianjin Medical University
- D. X. Wang🇨🇳Wuhan Children's Hospital; Huazhong University of Science and Technology
- Hui Zong🇨🇳Wuhan Children's Hospital; Huazhong University of Science and Technology
- Xiaoling Zhang🇨🇳 · autor de correspondenciaHan Dan City Eye Hospital - The Third Hospital of Handan
- Ruihua Wei🇨🇳 · autor de correspondenciaTianjin Medical University Eye Hospital; Tianjin Medical University
- Hong Jie🇨🇳 · autor de correspondenciaWuhan Children's Hospital
Resumen
El texto original del estudio, en el idioma en que se publicó.
Background
Repeated low-level red-light (RLRL) therapy has emerged as a promising non-invasive intervention for myopia control. However, the predictive factors underlying its efficacy remain insufficiently explored.
Methods
This multicenter cohort study included 538 pediatric patients who underwent RLRL treatment with a minimum follow-up of one year. Baseline ocular parameters and dynamic changes in axial length (AL) and choroidal thickness (ChT) over three months were collected. Multiple feature selection approaches-LASSO regression, Boruta, recursive elimination, and multivariate regression-were applied. Seven machine learning algorithms were trained, and their performance was evaluated using the area under the curve (AUC), sensitivity, specificity, and F1 score. SHAP and LIME analyses were utilized for interpretability.
Results
Logistic regression and gradient boosting models demonstrated the highest discriminative ability. XGBoost achieved optimal performance (AUC: 0.90-0.92; accuracy: 87.7-88.2 %; F1 score: 0.72-0.86). Across one- and two-year prediction models, six stable predictors were identified: age, baseline AL, anterior chamber depth (ACD), RNFL thickness at the temporal inferior quadrant (TI), ChangeAL, and ChangeChT. SHAP analysis revealed that ChangeAL was the dominant short-term predictor, whereas ChangeChT was most influential for long-term outcomes. Older age and greater choroidal thickening were consistently associated with a protective effect against myopia progression.
Conclusions
We developed and validated an interpretable machine learning model that accurately predicts short- and long-term outcomes of RLRL therapy in children. ChangeAL and ChangeChT serve as key dynamic biomarkers for treatment monitoring. These findings provide a foundation for personalized clinical decision-making in myopia management.
Resumen vía Europe PMC. Los derechos de autor siguen siendo de los autores o la editorial (CC BY-NC-ND).
Confianza de la comunidad
Cargando…
¿Cuánto confías en los hallazgos de este estudio?
Comentarios
¿Algo está mal aquí?
Señala este estudio si su información, etiquetas o financiación parecen incorrectas. Un editor revisa cada señalamiento.
Información educativa sobre investigación publicada. No es consejo médico ni una recomendación para empezar o dejar nada.