Capturing Movement Patterns in Pre-School Children using Markerless Motion Capture Technology

Research output: ThesisPh.D. thesis

4 Downloads (Pure)

Abstract

Many school children and adolescents experience musculoskeletal pain in the lower extremities. For some individuals, the pain is persistent and can affect both health and career choices later in life. As musculoskeletal pain is common already at school age, and since the consequences can be severe, it would be natural to investigate if any identifiable risk factors precede musculoskeletal pain at this age. A potential musculoskeletal risk factor is dynamic knee valgus (DKV) that is defined as a medial position of the knee during dynamic weight-bearing movement. DKV has been investigated in adults and adolescent populations and is, generally, considered a risk factor for the development of both traumatic and overuse injuries in the lower extremities. DKV is also known to be more prevalent among females, and some evidence suggests that this sex-difference accelerates at puberty. Although DKV is well studied in older ages, it is unknown to what extent DKV is present among children, if there are any sex-differences, and if there are any associations between early-in-life DKV and musculoskeletal health later in life. The quantification of DKV requires some form of motion capture. Traditional methods of motion capture involve the precise attachment of markers onto the skin of the investigated subjects and, therefore, tend to require long participant preparation times. Furthermore, the attached markers can interfere with the natural movement patterns of the investigated subject. Therefore, traditional methods of marker-based motion capture are not feasible for collecting large amounts of data in young children. Marker-less motion capture is a relatively novel technology that allows motion capture without the use of markers or anthropometric measurements. This leads to much shorter participant preparation times and allows the subjects to move naturally without impedance from attached equipment. This technology, therefore, holds promising feasibility, but the validity is not fully established, especially not in preschool populations. The overall aim of this thesis was to identify applicable motion capture measures in preschool children, and use these to create a dataset for future investigations of potential associations between early-in-life motor patterns and later musculoskeletal health. The specific foci of this thesis are first, to establish and evaluate the concurrent reliability and agreement between a marker-based and a novel markerless motion-capture system when measuring kinematics and jump performance in a sample of 14 preschool children (Study I). Secondly, to establish and evaluate the week-to-week test-retest reliability of selected kinematics measured in 63 preschool children using the marker-less motion capture system (Study II). Thirdly, to report the results from a cross-sectional study examining DKV, measured using the marker-less motion capture system, in a sample of 458 4 to 6-year-old preschool children. Specifically, to report on the prevalence of DKV, as well as potential differences of DKV concerning sex, age, and motor performance (Study III). In study I, we found that the measurements of five kinematic variables as well as jump length and height could be measured with acceptable reliability and agreement by the marker-less motion capture system. In study II, we found that the week-to-week test-retest reliability of single estimates of the same five kinematic variables was poor. Our results indicated that high levels of biological variation caused the poor week-to-week reliability rather than instrumental variability. In study III, we found that roughly 75% of 4 to 6-year-old children land with some degree of DKV, but also that girls are twice as common as boys among those children who land with the most pronounced levels of DKV. Overall, this research project was successful in creating a dataset containing complete motion capture data from 604 preschool children. This dataset is valuable for future prospective research projects. The markerless motion capture technology used in this thesis is now at a level of maturity where it is feasible to use in large scale projects, and where selected kinematic and jump performance measures have demonstrated acceptable validity in preschool children. In general, preschool children show high levels of week-to-week variability in their kinematic jump-landing patterns, although their week-to-week jump-length performance is relatively stable. Roughly 75% of preschool children land with some level of DKV, and girls are twice as common as boys among those children with the most extreme values of DKV. More research is needed to clarify the potential relationship between motor performance and kinematic landing strategies, and whether DKV early in life may interfere with later musculoskeletal pain or injury.
Original languageEnglish
Supervisors/Advisors
  • Lauridsen, Henrik, Supervisor
  • Hestbæk, Lise, Supervisor
  • Holsgaard-Larsen, Anders, Supervisor
Publication statusPublished - 7. Jul 2020

Fingerprint

Dive into the research topics of 'Capturing Movement Patterns in Pre-School Children using Markerless Motion Capture Technology'. Together they form a unique fingerprint.

Cite this