Effects of Simulated Marker Placement Deviations on Running Kinematics and Evaluation of a Morphometric-Based Placement Feedback Method.

Sean T Osis,Blayne A Hettinga,Shari Macdonald,Reed Ferber,Alejandro Lucia

doi:10.1371/journal.pone.0147111

Sean T Osis, Blayne A Hettinga + Show 3 more

Open Access

https://doi.org/10.1371/journal.pone.0147111

Copy DOI

Journal: PloS one	Publication Date: Jan 14, 2016
Citations: 26	License type: CC BY 4.0

Affiliation: University of Calgary, Running Injury Clinic

Abstract

In order to provide effective test-retest and pooling of information from clinical gait analyses, it is critical to ensure that the data produced are as reliable as possible. Furthermore, it has been shown that anatomical marker placement is the largest source of inter-examiner variance in gait analyses. However, the effects of specific, known deviations in marker placement on calculated kinematic variables are unclear, and there is currently no mechanism to provide location-based feedback regarding placement consistency. The current study addresses these disparities by: applying a simulation of marker placement deviations to a large (n = 411) database of runners; evaluating a recently published method of morphometric-based deviation detection; and pilot-testing a system of location-based feedback for marker placements. Anatomical markers from a standing neutral trial were moved virtually by up to 30 mm to simulate deviations. Kinematic variables during running were then calculated using the original, and altered static trials. Results indicate that transverse plane angles at the knee and ankle are most sensitive to deviations in marker placement (7.59 degrees of change for every 10 mm of marker error), followed by frontal plane knee angles (5.17 degrees for every 10 mm). Evaluation of the deviation detection method demonstrated accuracies of up to 82% in classifying placements as deviant. Finally, pilot testing of a new methodology for providing location-based feedback demonstrated reductions of up to 80% in the deviation of outcome kinematics.

Highlights

Anatomical marker placement has been identified as the single largest source of variance in repeated gait analyses using motion-capture techniques, resulting in inter-examiner differences of up to 34 degrees for certain joint angles [1]
While these results provide valuable insights, there is a lack of contextual information about how downstream kinematic data may be affected by marker deviations described by the IQRR
Two questions arise: 1) which anatomical markers have the greatest influence over outcome kinematics, and 2) what is an ‘acceptable’ value for the IQRR at each location to indicate that deviations in marker placement are within some acceptable limit

Summary

Introduction

Anatomical marker placement has been identified as the single largest source of variance in repeated gait analyses using motion-capture techniques, resulting in inter-examiner differences of up to 34 degrees for certain joint angles [1] This problem is relevant for large, multi-site biomechanical gait studies where inter-tester differences can preclude pooling of data [2, 3]. It has been shown that the IQRR is capable of indicating divergence in marker placements between a single Novice and Expert, and measurements over time can show how a single Novice alters their placements [4] While these results provide valuable insights, there is a lack of contextual information about how downstream kinematic data may be affected by marker deviations described by the IQRR. Two questions arise: 1) which anatomical markers have the greatest influence over outcome kinematics, and 2) what is an ‘acceptable’ value for the IQRR at each location to indicate that deviations in marker placement are within some acceptable limit

Methods

Results

Conclusion