Longitudinal Analysis of Step Counts in Parkinson’s Disease Patients: Insights from a Web-Based Application

Abstract

Parkinson’s disease (PD) is a chronic neurological disorder that affects millions of people worldwide. One of the common motor symptoms associated with PD is gait impairment, leading to reduced step count and mobility. Monitoring and analyzing step count data can provide valuable insights into the progression of the disease and the effectiveness of various treatments. The generalized additive model (GAM) model presents the following variables: sex (Male vs. Female, p = 0.03), handedness (Right vs. Left/Both, p = 0.015), PD status of father (Yes vs. No, p = 0.056), COVID-19 status (Yes vs. No, p = 0.008), cohort (PD vs. healthy control, p < 0.0001), the cubic regression spline with three basis functions of age by cohorts (p<0.0001) and the random effect of the individual age trajectories (p = 0.0001) are statistically significant for daily step counts. A web application specifically tailored for step count analysis in PD patients was also developed and it provides a user-friendly interface for patients, caregivers, and healthcare professionals to track and analyze step count data, facilitating personalized treatment plans and enhancing the management of PD.

Exploring Longitudinal MRI-Based Deep Learning Analysis in Parkinson’s Patients - A Short Survey Focus on Handedness

“…DL constitutes a specialized domain within the broader field of AI and sets itself apart from traditional machine learning models. Convolutional neural networks (CNN) are frequently employed for classification tasks in image classification and segmentation challenges in computer vision [52][53][54][55][56][57][58][59][60][61]. CNNs demonstrate exceptional efficiency…”