Gait Analysis Through Foot Pattern Recognition for Children With Cerebral Palsy

Abstract

In this study, we simplified the analysis of kinetic gait data using pattern recognition. Gait patterns were studied in 42 spastic children with cerebral palsy (age range: 3 to 17 years old), and 24 ageand sex-matched children. Gait analysis was performed using the DynoGraphy (CDG) system (Infortronic, Holland). The foot enrollment and the role of the heel or forefoot were assessed to form the gaitline. The bipedal phase was examined using a cyclogram, which is a cyclic characteristic formed by the changing position of the application point of the resultant normal force on a vertical supporting horizontal plane during motion. Based on the pattern recognition, the gait patterns of the subjects could be classified into 4 different patterns in both the gaitline and the cyclogram. The classification of the gait was parallel to the clinical evaluation of cerebral palsy obtained based on Minear's classification of daily activity (p < 0.05). The correlation between the gaitline and cyclogram was also highly significant (p < 0.05). The results of this study suggest that an automated pattern recognition program might provide an additional method for comprehensive gait evaluation in children with cerebral palsy.

Gait classification in children with cerebral palsy: A systematic review

“…Adequate sample definitions and inclusion criteria were provided in only 4 of the 18 publications included in the evaluation [13,15,17,18]. None of the studies provided information about the functional severity of participants.…”

“…Many studies did not report descriptive statistics, such as means, ranges or variances for each of the gait variables used to define the classification groups. Some studies [3,13,15,17,18,54] provided an illustrative representation of the distinguishing characteristics of each gait pattern, which may assist in the user-friendliness of the classification system.…”

“…In one study using qualitative construction methods, Wong et al [17] examined the correlation of their classification systems with Minear's classification of daily activity limitation, which were all measured on a 4-point ordinal scale. Pearson correlations were selected to examine this concurrent validity, however justification for using parametric techniques for ordinal data was not provided.…”

“…The diversity of gait deviations observed in children with CP has led to repeated efforts to develop gait classification systems to assist in diagnosis, clinical decision-making and communication [2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19]. Gait classifications may enable clinicians to differentiate gait patterns into clinically significant categories that assist with clinical decision-making.…”