Publications

Automatic methods for detecting FOG using the freeze index (FI) have been widely proposed to systematically monitor FOG in real life and guide therapy optimizations. However, methods to estimate the FI have relied on a broad range of measurement technologies and computational methodologies, often lacking mathematical rigor. This lack of standardization has severely hindered the acceptance of FI by regulatory agencies as a reproducible, robust, effective and safe measure on which to base further developments. In this study, we formalize the definition of the FI and propose a rigorous, explicit estimation algorithm, which may serve as a standard for future applications. This standardization provides a consistent and reliable benchmark.

The deterioration of gait can be used as a biomarker for ageing and neurological diseases. Continuous gait monitoring and analysis are essential for early deficit detection and personalized reha‐ bilitation. The use of mobile and wearable inertial sensor systems for gait monitoring and analysis have been well explored with promising results in the literature. In this work, we developed an inertial sensor based in‐shoe gait analysis system for real‐time gait monitoring and investigated the optimal sampling frequency to capture all the information on walking patterns. An exploratory valida‐ tion study was performed using an optical motion capture system on four healthy adult subjects.

Motion capture systems are widely accepted as ground-truth for gait analysis and are used for the validation of other gait analysis systems. To date, their reliability and limitations in manual labeling of gait events have not been studied. Objectives: Evaluate manual labeling uncertainty and introduce a hybrid stride detection and gait-event estimation model for autonomous, long-term, and remote monitoring. Methods: Estimate inter-labeler inconsistencies by computing the limits-of-agreement. Develop a hybrid model based on dynamic time warping and convolutional neural network to identify valid strides and eliminate non-stride data in inertial (walking) data collected by a wearable device. Finally, detect gait events within a valid stride region.