Multimodal Fuzzy Logic-based Gait Evaluation System for Assessing Children with Cerebral Palsy

Gait analysis is crucial for identifying functional deviations from the normal gait cycle and is essential for the individualized treatment of motor disorders such as cerebral palsy (CP). The primary contribution of this study is the introduction of a multimodal fuzzy logic system-based gait index (FLS-GIS), designed to provide numerical scores for gait patterns in both healthy children and those with CP, before and after surgery. This study examines and evaluates the surgical outcomes in children with CP who have undergone Achilles tendon lengthening. The FLS-GIS utilizes hierarchical feature fusion and fuzzy logic models to systematically evaluate and score gait patterns, focusing on spatial and temporal features across the hip, knee, and ankle joints. The two FLS types-1 (FLS-GIS-T1) and type-2 (FLS-GIS-T2) indices, respectively, were implemented to comprehensively study gait profiles. Starting with the gait parameters of all subjects, the changes in gait parameters in post-surgery children reflect significant improvements in gait dynamics, bringing walking patterns in CP children closer to those of their typically healthy peers. Both FLS-GIS-T1 and FLS-GIS-T2 demonstrated significant improvements in post-surgery evaluations compared to pre-surgery assessments, with p values < 0.05 and < 0.001, respectively, when compared to traditional indices. The proposed FLS-based index offers clinicians a robust and standardized gait evaluation tool, characterized by a fixed range of values, enabling consistent assessment across various gait conditions.