Title of Work

Error Associated with the Use of an Electromagnetic Motion Capture System During Full-Arc Shoulder Internal and External Rotation.

Document Type

Poster Session

Publication/Presentation Date

Spring 4-13-2019

City of Publication or Presentation

St. Louis Park

Conference Name

Minnesota American Physical Therapy Association Meeting


Purpose and Hypothesis: Electromagnetic (EM) tracking devices have been used in many biomechanical applications, including those for three-dimensional (3-D) shoulder kinematics. These systems have been found to be both reliable and accurate, however limited data from recent studies has suggested that large errors may occur when measuring shoulder rotational motion. The purpose of this study is to investigate the magnitude of measurement error between an inclinometer and a 3-D EM tracking system throughout maximal internal rotation (IR) and external rotation (ER) of the dominant arm. Our hypothesis is that a significant difference will be found in shoulder internal rotation (IR) and external rotation (ER) at end ranges between these two measurement systems, and that there will be a progressive increase in magnitude of error across the arc of motion.

Subjects: This study included 52 subjects (31 females, 21 males). This study utilized a convenience sample of DPT/PTA students at St. Catherine University. Inclusion criteria included being 18 years of age or older and currently free of shoulder pain. Exclusion criteria included: 1) metal implants; 2) history of/or current fracture, dislocation, surgery; 3) current pain that limited shoulder ROM; 4) not currently part of supervised shoulder rehabilitation program and 5) previous history of sensitivity to tape/adhesive that required medical attention.

Materials and Methods: One examiner placed the dominant arm of each subject, who were in a seated position, at 90° of both shoulder abduction and elbow flexion. The shoulder was then passively placed, with confirmation via an inclinometer, into each of the following positions: neutral rotation, 20° increments of IR, 45° & 90°of ER, and end-range IR and ER. A 3-D EM tracking system captured all scapular and humerothoracic kinematics at these each of these static positions. The 3-D kinematic data was collected with three Model 800 Trakstar sensors and a mid-range transmitter. Kinematic data was gathered at 100 Hz and processed with Motion Monitor software.

Results: Paired T-tests with Bonferroni corrections revealed significant differences in humerothoracic rotational values at all positions of IR & ER, except for neutral IR/ER and 45° of ER, with EM values being smaller than those gathered with the inclinometer. Absolute error between the two measurement devices increased with increasing IR motion, ranging from 0.5° at neutral to 22.2° at end-range IR. ER error was similar increasing from 1.0° at 45° to 25° at end-range. Root mean square error ranged from 10.7°-14.6° for IR and 7.4°-11.6° for ER.

Conclusions: 3-D EM tracking systems appear to underestimate both maximal IR and ER humerothoracic motion compared to standard clinical measures. Results from this study suggest that the amount of IR error using an EM system is similar to the error noted in this and a previous study for ER. Further research is needed to establish an offset for IR and ER motion gathered with an EM system in order to extrapolate to expected clinical motion values.

Clinical Relevance: Based on the results of this study, clinicians should use caution when comparing shoulder rotational values gathered with an EM system vs. clinical measures, such as an inclinometer.

This document is currently not available here.