A Multiple Regression Approach to Normalization of Spatiotemporal Gait Features.

Normalization of gait data is performed to reduce the effects of intersubject variations due to physical characteristics. This study reports a multiple regression normalization approach for spatiotemporal gait data that takes into account intersubject variations in self-selected walking speed and physical properties including age, height, body mass, and sex. Spatiotemporal gait data including stride length, cadence, stance time, double support time, and stride time were obtained from healthy subjects including 782 children, 71 adults, 29 elderly subjects, and 28 elderly Parkinson's disease (PD) patients. Data were normalized using standard dimensionless equations, a detrending method, and a multiple regression approach. After normalization using dimensionless equations and the detrending method, weak to moderate correlations between walking speed, physical properties, and spatiotemporal gait features were observed (0.01 < |r| < 0.88), whereas normalization using the multiple regression method reduced these correlations to weak values (|r| <0.29). Data normalization using dimensionless equations and detrending resulted in significant differences in stride length and double support time of PD patients; however the multiple regression approach revealed significant differences in these features as well as in cadence, stance time, and stride time. The proposed multiple regression normalization may be useful in machine learning, gait classification, and clinical evaluation of pathological gait patterns.

A multiple regression normalization approach to evaluation of gait in total knee arthroplasty patients.

BACKGROUND: Gait features characteristic of a cohort may be difficult to evaluate due to differences in subjects' demographic factors and walking speed. The aim of this study was to employ a multiple regression normalization method that accounts for subject age, height, body mass, gender, and self-selected walking speed in the evaluation of gait in unilateral total knee arthroplasty patients. METHODS: Three-dimensional gait analysis was performed on 45 total knee arthroplasty patients and 31 aged-matched controls walking at their self-selected speed. Gait data peaks including joint angles, ground reaction forces, net joint moments, and net joint powers were normalized using subject body mass, standard dimensionless equations, and a multiple regression approach that modeled subject age, height, body mass, gender, and self-selected walking speed. FINDINGS: Normalizing gait data using subject body mass, dimensionless equations, and multiple regression approach resulted in a significantly lower knee adduction moment and knee extensor power in total knee arthroplasty patients compared to controls (p<0.05). In contrast to normalization using body mass and dimensionless equations, multiple regression normalization greatly reduced variance in gait data by minimizing correlations with subject demographic factors and walking speed, resulting in significantly higher peak hip extension angles and peak hip flexion powers in total knee arthroplasty patients (p<0.05). INTERPRETATION: Total knee arthroplasty patients generate greater hip extension angles and hip flexor power and have a lower knee adduction moment than healthy controls. This gait pattern may be a strategy to reduce muscle and joint loading at the knee.

The effects of an ankle foot orthosis on cerebral palsy gait: A multiple regression analysis.

The aim of this study was twofold. Firstly, to develop a multiple regression normalization (MR) strategy to decorrelate physical properties and walking speed from spatiotemporal gait data in healthy children; and secondly, to use this MR approach to identify the effect of a solid ankle foot orthosis (AFO) on gait in children with cerebral palsy (CP). Spatiotemporal gait data during self-selected walking were obtained from 51 children with diplegic CP and 34 aged-matched healthy controls. Data were normalized using standard dimensionless equations (DS) and a MR approach. Stride length, stance time, swing time, and double support time were significantly different between children with CP and healthy controls using DS (p<;0.05); however, only stride length and swing time were significantly different when children with CP walked with and without an AFO. Normalizing gait data using DS demonstrated significant differences in cadence and step time in children with CP when wearing an AFO compared to the controls (p<;0.05). In contrast, MR normalization revealed significant differences in all spatiotemporal parameters between children with CP with and without an AFO, except double support time. After MR normalization, spatiotemporal parameters in children wearing an AFO became closer to those of the controls, except for double support time. The MR approach presented will assist in evaluating the effectiveness of conservative interventions such as AFOs in children with CP, as well as in surgery, and may be useful in gait classification using machine learning.

Classification of Parkinson's Disease Gait Using Spatial-Temporal Gait Features.

Quantitative gait assessment is important in diagnosis and management of Parkinson's disease (PD); however, gait characteristics of a cohort are dispersed by patient physical properties including age, height, body mass, and gender, as well as walking speed, which may limit capacity to discern some pathological features. The aim of this study was twofold. First, to use a multiple regression normalization strategy that accounts for subject age, height, body mass, gender, and self-selected walking speed to identify differences in spatial-temporal gait features between PD patients and controls; and second, to evaluate the effectiveness of machine learning strategies in classifying PD gait after gait normalization. Spatial-temporal gait data during self-selected walking were obtained from 23 PD patients and 26 aged-matched controls. Data were normalized using standard dimensionless equations and multiple regression normalization. Machine learning strategies were then employed to classify PD gait using the raw gait data, data normalized using dimensionless equations, and data normalized using the multiple regression approach. After normalizing data using the dimensionless equations, only stride length, step length, and double support time were significantly different between PD patients and controls (p < 0.05); however, normalizing data using the multiple regression method revealed significant differences in stride length, cadence, stance time, and double support time. Random Forest resulted in a PD classification accuracy of 92.6% after normalizing gait data using the multiple regression approach, compared to 80.4% (support vector machine) and 86.2% (kernel Fisher discriminant) using raw data and data normalized using dimensionless equations, respectively. Our multiple regression normalization approach will assist in diagnosis and treatment of PD using spatial-temporal gait data.