Measuring Everyday Life Motor Activities in Children and Adolescents with Neuromotor Impairments
Open access
Author
Date
2022Type
- Doctoral Thesis
ETH Bibliography
yes
Altmetrics
Abstract
In pediatric neurorehabilitation, children and adolescents with congenital and acquired illnesses and injuries of the developing brain are treated and cared for. These patients often present neuromotor impairments that result in difficulties in executing everyday life motor activities, such as walking or grasping an object. They undergo intensive therapy programs with an emphasis on reducing these limitations and fostering their functional independence in everyday life. To assess the patients’ progress during rehabilitation, usually, motor capacity (i.e., what a child can do in a standardized environment) is measured at the clinic. However, after discharge, motor performance (i.e., what a child does do in its habitual environment) becomes more important and it remains unclear whether children and adolescents can translate their improvements during rehabilitation into everyday life at home or school.
Wearable inertial sensors provide a promising solution to overcome this limitation. Technological progress has made these sensors small-sized, lightweight, energy-efficient, and thus applicable for unobtrusive long-term measurements of motor activities in the patients’ habitual environment. However, to derive meaningful outcome measures of everyday life motor activities, the unlabeled raw data generated by these sensors needs to be analyzed by appropriate data processing algorithms. Over the last decade, many algorithms have been developed that were predominantly designed for adult patient populations and rarely for pediatric populations. Hence, existing algorithms needed to be adapted to the needs of pediatric rehabilitation and validated in children and adolescents with neuromotor impairments.
Therefore, the primary aim of this thesis was to develop and validate a data processing algorithm that derives clinically meaningful motor performance measures about the daily motor activities of children and adolescents with neuromotor impairments based on data from a wearable inertial sensor system. The secondary aim was to apply this new technology to pediatric patients to measure their motor performance after rehabilitation and investigate if children and adolescents can translate their motor capacity achieved during rehabilitation into daily life.
The thesis began with a systematic review to get an overview of the technological possibilities of wearable inertial sensors to quantify everyday life motor activities in people with mobility impairments. This part was followed by summarizing the mobility and self-care goals of children and adolescents undergoing rehabilitation and conducting an international surveywith pediatric health professionals to identify the needs of pediatric rehabilitation. The findings of these studies led to the development of an algorithm that determines functional hand use with wrist sensors; the duration of lying, sitting, and standing positions with a trunk and a thigh sensor; the distance and speed of self-propelled wheeling periods with a wrist and a wheel sensor; and the duration, distance, and speed of walking periods with an ankle sensor.
As a next step, the algorithm was validated in the target population by three studies. The first study analyzed the accuracy of posture and mobility-related measures. It showed that the algorithm’s measurement error in estimating the duration of lying, sitting, standing, self-propelled wheeling, and walking was less than 10%. The second study investigated the validity of the functional hand use measures. The correlation coefficient between the sensor-based and video-based measures was 0.7. The third study determined the accuracy of sensor-based gait speed estimations. It revealed that the algorithm’s measurement error is small enough to detect clinically important changes of 0.1 m/s. These validity studies showed that the newly developed algorithm derives valid estimates of the children’s and adolescents’ motor performance.
In the final study of this thesis, children and adolescents with neuromotor impairments wore the sensors in their habitual environment after rehabilitation. This study revealed that children and adolescents with neuromotor impairments are less active on weekends than on school days and that their daily performed motor activities varied substantially between days. Consequently, we recommend measuring everyday life motor activities on seven consecutive days in future applications of the sensor system. This measurement protocol covers school days and weekend days, and the number of measurement days is sufficient to obtain reliable estimates of the children’s and adolescents’ motor performance. The study also showed that children and adolescents were willing to wear the sensors for a week in daily life.
Additionally, the same study investigated if children and adolescents with neuromotor impairments can translate their motor capacity achieved during rehabilitation into daily life by comparing their motor performance in daily life with their motor capacity assessed at the clinic. The capacity of children and adolescents explained only 13% to 58% of their motor performance in daily life. These weak correlations showed that motor assessments conducted at the clinic only partially reflect the patient’s performance at home and school. This confirms that capacity and performance are two different constructs and underpins the need to complement clinical assessments with performance measures conducted in the patients’ habitual environment. Moreover, it confirms that wearable sensors and the algorithm developed in this thesis capture essential information about the patients’ functioning in daily life, which adds value to clinical practice and rehabilitation research. Show more
Permanent link
https://doi.org/10.3929/ethz-b-000580069Publication status
publishedExternal links
Search print copy at ETH Library
Contributors
Examiner: Taylor, William R.
Examiner: Labruyere, Rob
Examiner: Newman, Christopher J.
Examiner: Lambercy, Olivier
Publisher
ETH ZurichOrganisational unit
03827 - Gassert, Roger / Gassert, Roger
More
Show all metadata
ETH Bibliography
yes
Altmetrics