A Machine Learning Pipeline for Gait Analysis in a Semi Free-Living Environment.

This paper presents a novel approach to creating a graphical summary of a subject's activity during a protocol in a Semi Free-Living Environment. Thanks to this new visualization, human behavior, in particular locomotion, can now be condensed into an easy-to-read and user-friendly output. As time series collected while monitoring patients in Semi Free-Living Environments are often long and complex, our contribution relies on an innovative pipeline of signal processing methods and machine learning algorithms. Once learned, the graphical representation is able to sum up all activities present in the data and can quickly be applied to newly acquired time series. In a nutshell, raw data from inertial measurement units are first segmented into homogeneous regimes with an adaptive change-point detection procedure, then each segment is automatically labeled. Then, features are extracted from each regime, and lastly, a score is computed using these features. The final visual summary is constructed from the scores of the activities and their comparisons to healthy models. This graphical output is a detailed, adaptive, and structured visualization that helps better understand the salient events in a complex gait protocol.

The Use of Inertial Measurement Units for the Study of Free Living Environment Activity Assessment: A Literature Review.

This article presents an overview of fifty-eight articles dedicated to the evaluation of physical activity in free-living conditions using wearable motion sensors. This review provides a comprehensive summary of the technical aspects linked to sensors (types, number, body positions, and technical characteristics) as well as a deep discussion on the protocols implemented in free-living conditions (environment, duration, instructions, activities, and annotation). Finally, it presents a description and a comparison of the main algorithms and processing tools used for assessing physical activity from raw signals.

Adaptive Change-Point Detection for Studying Human Locomotion.

This paper presents an innovative method to analyze inertial signals recorded in a semi-controlled environment. It uses an adaptive and supervised change point detection procedure to decompose the signals into homogeneous segments, allowing a refined analysis of the successive phases within a gait protocol. Thanks to a training procedure, the algorithm can be applied to a wide range of protocols and handles different levels of granularity. The method is tested on a cohort of 15 healthy subjects performing a complex protocol composed of different activities and shows promising results for the automated and adaptive study of human gait and activity.Clinical relevance- A new approach to study human activity and locomotion in Free-Living Environments FLEs through an adaptive change-point detection which isolates homogeneous phases.

A Movement Monitor Based on Magneto-Inertial Sensors for Non-Ambulant Patients with Duchenne Muscular Dystrophy: A Pilot Study in Controlled Environment.

UNLABELLED: Measurement of muscle strength and activity of upper limbs of non-ambulant patients with neuromuscular diseases is a major challenge. ActiMyo® is an innovative device that uses magneto-inertial sensors to record angular velocities and linear accelerations that can be used over long periods of time in the home environment. The device was designed to insure long-term stability and good signal to noise ratio, even for very weak movements. In order to determine relevant and pertinent clinical variables with potential for use as outcome measures in clinical trials or to guide therapy decisions, we performed a pilot study in non-ambulant neuromuscular patients. We report here data from seven Duchenne Muscular Dystrophy (DMD) patients (mean age 18.5 ± 5.5 years) collected in a clinical setting. Patients were assessed while wearing the device during performance of validated tasks (MoviPlate, Box and Block test and Minnesota test) and tasks mimicking daily living. The ActiMyo® sensors were placed on the wrists during all the tests. Software designed for use with the device computed several variables to qualify and quantify muscular activity in the non-ambulant subjects. Four variables representative of upper limb activity were studied: the rotation rate, the ratio of the vertical component in the overall acceleration, the hand elevation rate, and an estimate of the power of the upper limb. The correlations between clinical data and physical activity and the ActiMyo® movement parameters were analyzed. The mean of the rotation rate and mean of the elevation rate appeared promising since these variables had the best reliability scores and correlations with task scores. Parameters could be computed even in a patient with a Brooke functional score of 6. The variables chosen are good candidates as potential outcome measures in non-ambulant patients with Duchenne Muscular Dystrophy and use of the ActiMyo® is currently being explored in home environment. TRIAL REGISTRATION: ClinicalTrials.gov NCT01611597.