Predicting executive functioning from walking features in Parkinson's disease using machine learning

Abstract

Parkinson's disease is characterized by motor and cognitive deficits. While previous work suggests a relationship between both, direct empirical evidence is scarce or inconclusive. Therefore, we examined the relationship between walking features and executive functioning in patients with Parkinson's disease using state-of-the-art machine learning approaches. A dataset of 103 geriatric Parkinson inpatients, who performed four walking conditions with varying difficulty levels depending on single task walking and additional motor and cognitive demands, was analyzed. Walking features were quantified using an inertial measurement unit (IMU) system positioned at the patient's lower back. The analyses included five imputation methods and four regression approaches to predict executive functioning, as measured using the Trail-Making Test (TMT). Multiple imputation by chained equations (MICE) in combination with support vector regression (SVR) reduce the mean absolute error by about 4.95% compared to baseline. Importantly, predictions solely based on walking features obtained with support vector regression mildly but significantly correlated with Δ-TMT values. Specifically, this effect was primarily driven by step time variability, double limb support time variability, and gait speed in the dual task condition with cognitive demands. Taken together, our data provide direct evidence for a link between executive functioning and specific walking features in Parkinson's disease.

Keywords: Executive functioning; Machine learning; Parkinson’s disease; Walking features.

Fig. 1

Boxplot for each walking condition: top left ST normal, top right ST fast, bottom left DT walking-motor, and bottom right DT walking-cognitive. The boxplots show when a particular feature was dropped using RFE. Both meta features (M, orange) and walking features (W, blue) were used. All RFE-values vary between 1 (least important feature, therefore dropped first) and 21 (most important feature) and are provided as the average of 5 independent RFE runs. Abbreviations: ^a UPDRS-III total score, ^b DIA-S total score, ^c Walking aid usage in percent, ^d Number of steps, ^e Time in seconds, ^f Disease duration in years, ^g Gait speed in meters per second.

Fig. 2

Boxplot for each walking condition: top left ST normal, top right ST fast, bottom left DT walking-motor, and bottom right DT walking-cognitive. The boxplots show when a particular feature was dropped using RFE, when using walking features only. All RFE-values vary between 1 (least important feature, therefore dropped first) and 11 (most important feature) and are provided as the average of 5 independent RFE runs. Abbreviations: ^a Number of steps, ^b Time in seconds, ^c Gait speed in meters per second.