Digital Pedoscopic Assessment and Data-Driven Classification of Pediatric Rearfoot Alignment

Accurate assessment of pediatric foot biomechanics is challenging due to growth-related variability and limited quantitative tools. The supination and pronation angles of the ankle are critical for understanding lower limb alignment and pathological gait patterns.

This study introduces a novel digital pedoscopic system designed to enhance the quantitative evaluation of foot alignment and to demonstrate its clinical utility through clustering analysis of pediatric ankle angles.

Thirty-five pediatric patients (mean age = 6.17 ± 4.54 years) with neurological or developmental disorders were evaluated using a semi-automated digital pedoscopic system to obtain quantitative measurements of ankle alignment. Key anatomical landmarks, including the heel, calf, and knee centers, were manually identified from posterior images, and the system automatically calculated ankle pronation and supination angles. K-means clustering analysis was applied to classify participants based on their biomechanical profiles.

A total of thirty-five pediatric patients were assessed, and the revised abstract now explicitly reports this sample size to improve clarity. Data-driven k-means clustering of bilateral rearfoot angles identified three clearly defined alignment subgroups-neutral, pronated, and supinated-each exhibiting characteristic distribution patterns and degrees of inter-individual variability. These findings highlight the system's ability to quantitatively distinguish biomechanical phenotypes within a heterogeneous pediatric population. Visualization through scatter, box, and violin plots demonstrated distinct cluster-specific distributions and inter-individual variability in rearfoot alignment, demonstrating the feasibility of objective biomechanical stratification in pediatric populations.

The digital pedoscopic imaging system provides a reliable and reproducible approach for quantitative assessment of foot alignment in children. Clustering analysis enables stratification of biomechanical subtypes, supporting individualized rehabilitation strategies and longitudinal monitoring in pediatric clinical practice.