Journal article 897 views
7th International Society for Physical Activity and Health Congress
Journal of Physical Activity and Health, Volume: 15, Issue: 10 Suppl 1, Pages: S46 - S47
Swansea University Author:
Gareth Stratton
Full text not available from this repository: check for access using links below.
DOI (Published version): 10.1123/jpah.2018-0535
Abstract
Introduction: Accelerometers are widely used to study physical activity and have been shown to be informative of motion mechanics. Whilst Process-oriented assessment is an important tool in the development of children’s fundamental movement skills, current methods of assessment are cumbersome and su...
Published in: | Journal of Physical Activity and Health |
---|---|
ISSN: | 1543-3080 1543-5474 |
Published: |
2018
|
Online Access: |
Check full text
|
URI: | https://cronfa.swan.ac.uk/Record/cronfa46092 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Abstract: |
Introduction: Accelerometers are widely used to study physical activity and have been shown to be informative of motion mechanics. Whilst Process-oriented assessment is an important tool in the development of children’s fundamental movement skills, current methods of assessment are cumbersome and subjective. We present a novel analysis framework for activity assessment that uses accelerometry to create sophisticated motion maps and demonstrate their utility in profiling and categorising movement mechanics, objectively, in a diverse range of fundamental movements. Methods: Acceleration data were collected from ankle and wrist mounted sensors. Children aged 9 - 12 years were assessed in a multi-stage fitness test and a fundamental movement skill (FMS) challenge. Acceleration and magnetometer data were used to construct spectrograms, phase maps of motion and a performance sphere. Specific activity components were analysed through pattern recognition using machine learning, and dynamic time-warping. Results: Novel analyses of FMS displayed patterns clearly linked to specific activities such as throwing, jumping and body roll. These were sufficient to classify performance into activity categories using pattern recognition and a training set from expert observer scores. Discussion: Novel analyses of children’s mechanical motion patterns can be achieved for FMS using lightweight, low cost wearable sensors. These motion maps can be predictive of performance based on limited sampling allowing population profiling of FMS. Further, the use of computerised pattern recognition and classification gives an objective scoring of complex motion, normally requiring subjective assessment by expert human observers. |
---|---|
College: |
Faculty of Science and Engineering |
Issue: |
10 Suppl 1 |
Start Page: |
S46 |
End Page: |
S47 |