In vivo measurement and objective classification of healthy, injured and pathological shoulder complex function

Stroud Larreal, Lindsay Ann 2011. In vivo measurement and objective classification of healthy, injured and pathological shoulder complex function. PhD Thesis, Cardiff University. Item availability restricted.

Preview	PDF - Accepted Post-Print Version Download (89MB) | Preview
	PDF - Supplemental Material Restricted to Repository staff only Download (80kB)

Abstract

ysis (MA) techniques have been previously developed at Cardiff University to assess shoulder function following International Society of Biomechanics (ISB) recommendations. However, errors in the system significantly affect shoulder kinematics measurements. Image registration techniques (IRT) were developed to accurately measure GH joint kinematics using dynamic single-plane fluoroscopy. 3D computer bone models of the humerus and scapula were generated from magnetic resonance imaging (MRI) scans using Simpleware Software (Simpleware Ltd). Accurate 3D to two dimensional ( 2D) image registration was performed using Joint- Track software (Banks, S.A.). Full kinematics descriptions of the GH joint and of the scapula were obtained. The pattern of rotation agrees with what other researchers have previously measured. Humeral head translation was measured towards the glenoid centre (3�0.9mm medially, 2.7�0.9mm inferiorly and then superiorly and 6.7�2mm posteriorly) during abduction and (2.8�0.9mm medially, 3.6�0.9mm superiorly and then inferiorly and 5.3�2.1mm anteriorly) during scaption. The centering of the humeral head is believed to provide joint congruency for optimal shoulder function. To investigate the errors commonly associated with MA, a comparison between the kinematics outputs from the MA measuring system and IRT was performed. Greater GH joint elevation was recorded with IRT (54.8� and 82.6� for abduction and scaption respectively) compared to MA (51.1� and 75.2� for abduction and scaption respectively). Furthermore, differences between IRT and MA recordings in GH joint plane of elevation (6.7� and 1.9� abduction and scaption respectively) and axial rotation (24.1� and 23.0� abduction and scaption respectively) were measured. Discrepancies in measured rotations between MA and IRT can be attributed to factors related to differences in the analytical approach as well as the errors commonly associated with the techniques. Additions and improvements to the original Cardiff MA protocol for measuring and analysing shoulder biomechanics were made and healthy and shoulder patient function was subsequently investigated. The glenohumeral (GH) joint centre of rotation (CoR) estimation by means of the instantaneous helical axis (IHA) method was included in the Cardiff model using International Shoulder Group (ISG) routines. With the original protocol, only regression equations (MRE) based on scapula geometry were used to estimate GH joint CoR. Differences between IHA and MRE were investigated by comparing the estimated CoR positions relative to the scapula anatomical coordinate system (ACS). The MRE significantly overestimated the GH joint CoR in the anterior position (by 4 cm) compared to the IHA method and to the work of other research groups. The iii MRE also estimated the GH joint CoR laterally to the scapula ACS although imaging studies identified GH joint CoR medially to the scapula ACS. Trunk contribution to overall arm elevation was assessed between unilateral (UE) and bilateral (BE) arm elevations. BE was found to significantly decrease trunk lateral and axial rotation with respect to UE; however, trunk flexion was significantly greater. This in turn resulted in significantly different scapula rotations between UE and BE with up to 3� difference in scapula retraction during abduction between UE and BE. Consequently in shoulder complex biomechanics studies, particular attention should be made to minimise trunk rotations. Shoulder function asymmetry was investigated between dominant and nondominant shoulders. Significantly greater GH elevation and scapula lateral rotation were measured in dominant arms compared to non-dominant arms, with a difference of up to 7.6� and 7.0� respectively between the two arms. Asymmetry between the two shoulders could be attributed to soft tissue imbalance from more frequent use of the dominant shoulder compared to the non-dominant. Physiological range of motion (during static and dynamic trials) and 15 activities of daily living (ADLs) were recorded with skin markers attached to bony landmarks as well as with the AMC (and the SL for physiological ROMs). Static and dynamic trials measured differences in thorax and scapula rotations which may have arisen from muscle stabilisation. Acromioclavicular (AC) and scapula lateral rotations were underestimated (by up to 8� and 20� respectively) using the skin fixed markers. Joint and segment rotations are comparable to published studies that follow ISB recommendations The kinematics of patients with four different shoulder conditions (clavicle fracture, multidirectional instability, irreparable rotator cuff tear and GH dislocation) was measured. The effect and the extent of the IoP was investigated during physiological ROMs elevation and ADLs recordings by comparing their function to healthy and contralateral shoulders. The results from this study were used to develop a novel application for the Cardiff Dempster Shafer (DS) objective classifier. The classification tool was used to characterise shoulder complex function of 40 participants. Non injured or pathological (NIoP) and IoP shoulder function was characterised with 72.5% accuracy. Eight patients were misclassified as having NIoP shoulder function while two healthy participants were misclassified as having IoP function. A weak correlation between scoring questionnaires with the NIoP and IoP classification indices was found (-0.16298 and 0.180187 respectively). This might be explained by the subjective nature of the scores. The studies described in this thesis contributed towards advancements in shoulder complex kinematics studies at Cardiff University as well as with the international shoulder researcher’s community. An appreciation was gained of the challenges faced when using MA and IRT to measure shoulder motion as well as a better understanding of joint function in healthy and IoP shoulders.

Item Type:	Thesis (PhD)
Status:	Unpublished
Schools:	Engineering
Subjects:	R Medicine > R Medicine (General) T Technology > TA Engineering (General). Civil engineering (General)
Uncontrolled Keywords:	Shoulder kinematics ; Motion analysis ; Image registration ; Objective classification
Date of First Compliant Deposit:	30 March 2016
Last Modified:	19 Mar 2016 22:18
URI:	https://orca.cardiff.ac.uk/id/eprint/11414

Actions (repository staff only)

Edit Item