Cardiff University | Prifysgol Caerdydd ORCA
Online Research @ Cardiff 
WelshClear Cookie - decide language by browser settings

Development of novel methodologies to quantify, analyse and classify in-vivo knee function affected by aging, osteoarthritis and total knee replacement

Watling, Daniel 2014. Development of novel methodologies to quantify, analyse and classify in-vivo knee function affected by aging, osteoarthritis and total knee replacement. PhD Thesis, Cardiff University.
Item availability restricted.

PDF - Accepted Post-Print Version
Download (9MB) | Preview
[img] PDF - Supplemental Material
Restricted to Repository staff only

Download (91kB)


Total knee replacement is effective at reducing pain resulting from end stage knee osteoarthritis but patient clinical outcome remains poor. This study develops methodologies for the assessment and classification of knee function to further the understanding of the mechanical effects of osteoarthritis, identify targets for treatment and objectively evaluate patient functional recovery. This study additionally develops methodologies to investigate how internal knee structures function during dynamic in-vivo activities using novel approaches to MRI. The uncertainties in the assessment of patients’ knee function were first investigated. Stair gait analysis was investigated to complement traditional level gait analysis. Measurements with the force plate interacting with step one displayed lowest inter-subject variability and were carried forwards to patient assessments. The level and stair gait of three age groups of healthy volunteers was investigated to select a healthy control group. A trend of peak sagittal moments at the hip, knee and ankle all decreasing with aging and frontal plane joint moments at the hip and ankle increase with aging was found. Age related changes were non-linear becoming more prominent towards old age. Young and middle aged healthy volunteers were combined to form a larger, homogenous cohort for patient comparisons. Inter-subject variability is neither helped nor hindered by the inclusion stair gait analysis when compared to level gait ensuring no adverse affects in distinguishing functional changes from naturally occurring individual variation. Patient completion of stair gait was unfortunately very poor however and the practicalities of stair gait in the patient population were found to be preventative. The accuracy of using principal component analysis over traditional parameterisation to classify osteoarthritic or healthy knee function was found to improve classification accuracy when using the Cardiff Dempster Shafer Theory Classifier. Knee measures were found to have poor classification accuracy with hip and ankle adaptations best discriminating healthy and patient gait. The adaptations to lower limb biomechanics observed in older healthy adults appear to become magnified with osteoarthritis. The classification methodology developed throughout the study resulted in an in and out of sample classification accuracy of 97.9% in determining osteoarthritic or healthy knee function. The hip, knee, ankle and ground reaction force biomechanical data of 12 patients before total knee replacement and one year post surgery were assessed using principal component analysis and the Cardiff Demspter Shafer Classifier. 42% of patients experience no functional benefit and 25% of patients recovered lower limb function characteristic of young and middle aged healthy volunteers. Classification of total knee replacement outcome was found to correlate with clinical outcome measures but implant type, BMI, weight loss and presence of comorbidities were poor indicators of patient outcome. Patient age and pre surgery function were found to correlate to the function observed post surgery, indicating that the earlier timing of intervention (and visualisation of this using the classification methodology) may improve functional and clinical outcome of patients with end stage knee arthritis. Novel, high resolution MR imaging and analysis techniques to quantify 3D, patient specific, invivo menisco-tibial kinematics and meniscus shape change were created. Repeatability was high with largest errors due to MRI image quality. Posterior translations of both menisci were found with increasing knee flexion angle and large meniscus translations, as much as 3.4mm in the anterior-posterior direction and 4mm in the medial-lateral direction were observed with changing transverse plane knee joint rotations. Load bearing introduced substantial additional posterior motion of both menisci, up to 4.6mm (medial meniscus) and 5.2mm (lateral meniscus) in the flexed knee. Large variability was observed between subjects suggesting patient specific response to load bearing is an important consideration in the treatment and rehabilitation of soft tissue injuries and prevention of early onset arthritis. Flexion of the knee was found to introduce compression in the menisci while loading contributed greatest posterior translation of both menisci. The posterior horns of both menisci were also found to rotate towards the centre of the tibial plateau during flexion and maintaining this during treatment of soft tissue injuries may aid the prevention of secondary pathologies.

Item Type: Thesis (PhD)
Status: Unpublished
Schools: Engineering
Subjects: T Technology > TJ Mechanical engineering and machinery
Uncontrolled Keywords: Motion analysis; Biomechanics; Classification; Medical imaging; Osteoarthritis; Gait.
Funders: ARUKBBC, Depuy International Ltd
Date of First Compliant Deposit: 30 March 2016
Last Modified: 19 Mar 2016 23:49

Actions (repository staff only)

Edit Item Edit Item


Downloads per month over past year

View more statistics