AUTHOR=Ren Shengwei , Yang Kaili , Xu Liyan , Fan Qi , Gu Yuwei , Pang Chenjiu , Zhao Dongqing 

TITLE=Machine learning analysis with the comprehensive index of corneal tomographic and biomechanical parameters in detecting pediatric subclinical keratoconus

JOURNAL=Frontiers in Bioengineering and Biotechnology

VOLUME=Volume 11 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/bioengineering-and-biotechnology/articles/10.3389/fbioe.2023.1273500

DOI=10.3389/fbioe.2023.1273500

ISSN=2296-4185

ABSTRACT=Background: Keratoconus (KC) occurs at puberty while the diagnosis results were focused on adults that is lacking in pediatric patients. Early diagnosis of pediatric KC can avoid the progression and improve the quality of life of patients. The study aimed to evaluate the ability of corneal tomographic and biomechanical variables through machine learning analysis to detect subclinical keratoconus (SKC) in a pediatric population.
Methods: Fifty-two KC, 52 SKC, and 52 control pediatric eyes matched with age and gender were recruited in the case-control study. The corneal tomographic and biomechanical parameters were measured by professionals. Linear mixed effects test was used to compare the differences among the three groups and the Least Significant Difference analysis was further used to conduct pairwise comparisons. The receiver operating characteristic (ROC) curve and the Delong test were used to evaluate the diagnostic ability. Variables were performed for multivariate logistic regression in the machine learning analysis, using a stepwise variable selection to decrease overfitting, and comprehensive indexes for detecting pediatric SKC eyes were produced in each step.
Results: PE, BAD-D, and TBI had the highest values of the area under the AUC value in identifying pediatric KC eyes, and the corresponding cut-off values were 12 μm, 2.48, and 0.6 separately. For discriminating SKC eyes, the highest AUC (95%(CI) was found in SP A1 with a value was 0.84(0.765, 0.915), and BAD-D was the best parameter among the corneal tomographic parameters with the AUC(95%CI) value of 0.817(0.729, 0.886). Three models were generated in the machine learning analysis, and Model 3 (y= 0.400*PE +1.982* DA Ratio Max[2mm] – 0.072*SP A1–3.245) had the highest AUC(95%CI) value, with 90.4% sensitivity and 76.9% specificity and the cut-off value providing the best Youden index was 0.19. 
Conclusions: The criteria of parameters for diagnosing pediatric KC and SKC eyes were inconsistent with the adult population. Combined corneal tomography and biomechanical parameters could enhance the early diagnosis of young patients and improve the inadequate representation of pediatric KC research.