Subject-specific finite element models could improve understanding of how spinal loading varies between people, based on differences in morphology and tissue properties. However, determining accurate subject-specific intervertebral disc (IVD) properties can be difficult due to the spine's complex behaviour, in six degrees of freedom. Previous studies optimising IVD properties have utilised axial compression alone or range of motion data in three axes. This study aimed to optimise IVD properties using 6-axis force-moment data, and compare the resultant model's accuracy against a model optimised using IVD pressure data. Additionally, model vertebral alignment was assessed to determine if differences between imaged specimen alignment and in vitro 6-axis test alignment affected the optimisation process. A finite element model of a porcine lumbar motion segment was developed, with generic IVD properties. The model loading and boundary conditions replicated in vitro 6-axis stiffness matrix testing of the same specimen. The model was then optimised twice, once using experimental IVD pressures and once using forces and moments. A second model with geometry based on the specimen's vertebral alignment from the 6-axis testing was also developed and optimised. The 6-axis force-moment optimised model had more accurate overall 6-axis load-displacement behaviour, but less accurate IVD pressures than the pressure optimised model. Neither optimised model fully captured spinal behaviours, due to model and optimisation process limitations. The 6-axis vertebral alignment model had lower error and different optimised IVD properties than the imaged vertebral alignment model. Thus, vertebral alignment affected segment stiffness, so should be considered when developing spine models.

AbstractObjectivesThe syndesmosis joint, located between the tibia and fibula, is critical to maintaining the stability and function of the ankle joint. Damage to the ligaments that support this joint can lead to ankle instability, chronic pain, and a range of other debilitating conditions. Understanding the kinematics of a healthy joint is critical to better quantify the effects of instability and pathology. However, measuring this movement is challenging due to the anatomical structure of the syndesmosis joint. Biplane Video Xray (BVX) combined with Magnetic Resonance Imaging (MRI) allows direct measurement of the bones but the accuracy of this technique is unknown. The primary objective is to quantify this accuracy for measuring tibia and fibula bone poses by comparing with a gold standard implanted bead method.MethodsWritten informed consent was given by one participant who had five tantalum beads implanted into their distal tibia and three into their distal fibula from a previous study. Three-dimensional (3D) models of the tibia and fibula were segmented (Simpleware Scan IP, Synopsis) from an MRI scan (Magnetom 3T Prisma, Siemens). The beads were segmented from a previous CT and co-registered with the MRI bone models to calculate their positions. BVX (125 FPS, 1.25ms pulse width) was recorded whilst the participant performed level gait across a raised platform. The beads were tracked, and the bone position of the tibia and fibula were calculated at each frame (DSX Suite, C-Motion Inc.). The beads were digitally removed from the X-rays (MATLAB, MathWorks) allowing for blinded image-registration of the MRI models to the radiographs. The mean difference and standard deviation (STD) between bead-generated and image-registered bone poses were calculated for all degrees of freedom (DOF) for both bones.ResultsThe absolute mean tibia and fibula bone position differences (Table 1) between the bead and BVX poses were found to be less than 0.5 mm for both bones. The bone rotation differences were found to be less than 1° for all axes except for the fibula Z axis rotation which was found to be 1.46°. One study1 has reported the kinematics of the syndesmosis joint and reported maximum ranges of motion of 9.3°and translations of 3.3mm for the fibula. The results show that the accuracy of the methodology is sufficient to quantify these small movements.ConclusionsBVX combined with MRI can be used to accurately measure the syndesmosis joint. Future work will look at quantifying the accuracy of the talus to provide further understanding of normal ankle kinematics and to quantify the kinematics across a healthy population to act as a comparator for future patient studies.Declaration of Interest(b) declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported:I declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research project.

AbstractObjectivesInvestigate Magnetic Resonance Imaging (MRI) as an alternative to Computerised Tomography (CT) when calculating kinematics using Biplane Video X-ray (BVX) by quantifying the accuracy of a combined MRI-BVX methodology by comparing with results from a gold-standard bead-based method.MethodsWritten informed consent was given by one participant who had four tantalum beads implanted into their distal femur and proximal tibia from a previous study. Three-dimensional (3D) models of the femur and tibia were segmented (Simpleware Scan IP, Synopsis) from an MRI scan (Magnetom 3T Prisma, Siemens). Anatomical Coordinate Systems (ACS) were applied to the bone models using automated algorithms1. The beads were segmented from a previous CT and co-registered with the MRI bone models to calculate their positions. BVX (60 FPS, 1.25 ms pulse width) was recorded whilst the participant performed a lunge. The beads were tracked, and the ACS position of the femur and tibia were calculated at each frame (DSX Suite, C-Motion Inc.). The beads were digitally removed from the X-rays (MATLAB, MathWorks) allowing for blinded image-registration of the MRI models to the radiographs. The mean difference and standard deviation (STD) between bead-generated and image-registered bone poses were calculated for all degrees of freedom (DOF) for both bones. Using the principles defined by Grood and Suntay2, 6 DOF kinematics of the tibiofemoral joint were calculated (MATLAB, MathWorks). The mean difference and STD between these two sets of kinematics were calculated.ResultsThe absolute mean femur and tibia ACS position differences (Table 1) between the bead and image-registered poses were found to be within 0.75mm for XYZ, with all STD within ±0.5mm. Mean rotation differences for both bones were found to be within 0.2º for XYZ (Table 1). The absolute mean tibiofemoral joint translations (Table 1) were found to be within ±0.7mm for all DOF, with the smallest absolute mean in compression-distraction. The absolute mean tibiofemoral rotations were found to be within 0.25º for all DOF (Table 1), with the smallest mean was found in abduction-adduction. The largest mean and STD were found in internal-external rotation due to the angle of the X-rays relative to the joint movement, increasing the difficulty of manual image registration in that plane.ConclusionThe combined MRI-BVX method produced bone pose and tibiofemoral kinematics accuracy similar to previous CT results3. This allows for confidence in future results, especially in clinical applications where high accuracy is needed to understand the effects of disease and the efficacy of surgical interventions. Acknowledgements: This research was supported by the Engineering and Physical Sciences Research Council (EPSRC) doctoral training grant (EP/T517951/1).Declaration of Interest(b) declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported:I declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research project.

AbstractObjectivesSpinal disorders such as back pain incur a substantial societal and economic burden. Unfortunately, there is lack of understanding and treatment of these disorders are further impeded by the inability to assess spinal forces in vivo. The aim of this project is to address this challenge by developing and testing a novel image-driven approach that will assess the forces in an individual's spine in vivo by incorporating information acquired from multimodal imaging (magnetic resonance imaging (MRI) and biplane X-rays) in a subject-specific model.MethodsMagnetic resonance and biplane X-ray imaging are used to capture information about the anatomy, tissues, and motion of an individual's spine as they perform a range of everyday activities. This information is then utilised in a subject-specific computational model based on the finite element method to predict the forces in their spine. The project is also utilising novel machine learning algorithms and in vitro, six-axis mechanical testing on human, porcine and bovine samples to develop and test the modelling methods rigorously.Results & DiscussionMRI sequences have been identified that provide high-quality image data and information on different tissue types which will be used to predict subject-specific disc properties. In-vivo protocols to capture motion analysis, EMG muscle activity, and video X-rays of the spine have been designed with planned data collection of 15 healthy volunteers. Preliminary modelling work has evaluated potential machine learning approaches and quantified the sensitivity of the models developed to material properties.ConclusionThe development and testing of these image-driven subject-specific spine models will provide a new tool for determining forces in the spine. It will also provide new tools for measuring and modelling spine movement and quantifying the properties of the spinal tissues.AcknowledgmentsFunding from the EPSRC: EP/V036602/1 (Meakin, Holsgrove & Javadi) and EP/V032275/1 (Holt & Williams).Declaration of Interest(b) declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported:I declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research project.

Background: Both medial knee osteoarthritis and associated varus alignment have been proposed to alter knee joint loading and consequently overloading the medial compartment. Individuals with knee osteoarthritis and varus deformity are candidates for coronal plane corrective surgery, high tibial osteotomy. This study evaluated knee loading and contact location for a control group, a pre-surgery cohort and the same cohort 12 months postsurgery using a musculoskeletal modelling approach. Methods: Joint kinematics during gait were measured in 30 knee osteoarthritis patients, before and after high tibial osteotomy, and 28 healthy adults. Using a musculoskeletal model that incorporated patient-specific mechanical tibial femoral angle, the resulting muscle, ligament, and contact forces were calculated and the medial lateral condyle load distribution was analysed. Findings: Surgery changed medial compartment contact force throughout stance relative to pre-surgery. This reduction in medial compartment contact force pre- vs post-HTO is observed despite a significant increase in post-surgery walking speed compared to pre-HTO, where increased speed is typically associated with increased joint loading. Interpretation: This study has estimated the effects of high tibial osteotomy on knee loading using a generic model that incorporates a detailed knee model to better understand tibiofemoral contact loading. The findings support the aim of surgery to unload the medial knee compartment and lateralise joint contact forces.

To be able to assess the biomechanical and functional effects of ankle injury and disease it is necessary to characterise healthy ankle kinematics. Due to the anatomical complexity of the ankle, it is difficult to accurately measure the Tibiotalar and Subtalar joint angles using traditional marker-based motion capture techniques. Biplane Video X-ray (BVX) is an imaging technique that allows direct measurement of individual bones using high-speed, dynamic X-rays. The objective is to develop an in-vivo protocol for the hindfoot looking at the tibiotalar and subtalar joint during different activities of living. A bespoke raised walkway was manufactured to position the foot and ankle inside the field of view of the BVX system. Three healthy volunteers performed three gait and step-down trials while capturing Biplane Video X-Ray (125Hz, 1.25ms, 80kVp and 160 mA) and underwent MR imaging (Magnetom 3T Prisma, Siemens) which were manually segmented into 3D bone models (Simpleware Scan IP, Synopsis). Bone position and orientation for the Talus, Calcaneus and Tibia were calculated by manual matching of 3D Bone models to X-Rays (DSX Suite, C-Motion, Inc.). Kinematics were calculated using MATLAB (MathWorks, Inc. USA). Pilot results showed that for the subtalar joint there was greater range of motion (ROM) for Inversion and Dorsiflexion angles during stance phase of gait and reduced ROM for Internal Rotation compared with step down. For the tibiotalar joint, Gait had greater inversion and internal rotation ROM and reduced dorsiflexion ROM when compared with step down. The developed protocol successfully calculated the in-vivo kinematics of the tibiotalar and subtalar joints for different dynamic activities of daily living. These pilot results show the different kinematic profiles between two different activities of daily living. Future work will investigate translation kinematics of the two joints to fully characterise healthy kinematics.

Biplane video X-ray (BVX) – with models segmented from magnetic resonance imaging (MRI) – is used to directly track bones during dynamic activities. Investigating tibiofemoral kinematics helps to understand effects of disease, injury, and possible interventions. Develop a protocol and compare in-vivo kinematics during loaded dynamic activities using BVX and MRI. BVX (60 FPS) was captured whilst three healthy volunteers performed three repeats of lunge, stair ascent and gait. MRI scans were performed (Magnetom 3T Prisma, Siemens). 3D bone models of the tibia and femur were segmented (Simpleware Scan IP, Synopsis). Bone poses were obtained by manually matching bone models to X-rays (DSX Suite, C-Motion Inc.). Mean range of motion (ROM) of the contact points on the medial and lateral tibial plateau were calculated using custom MATLAB code (MathWorks). Results were filtered using an adaptive low pass Butterworth filter (Frequency range: 5-29Hz). Gait and Stair ascent activities from one participant's data showed increased ROM for medial-lateral (ML) translation in the medial compartment but decreased ROM in anterior-posterior (AP) translation when comparing against the same translations on the lateral compartment of the tibial plateau. Lunge activity showed increased ROM for both ML and AP translation in the medial compartment when compared with the lateral compartment. These results highlight the variability in condylar translations between different activities. Understanding healthy in-vivo kinematics across different activities allows the determination of suitable activities to best investigate the kinematic changes due to disease or injury and assess the efficacy of different interventions. Acknowledgements: This research was supported by the Engineering and Physical Sciences Research Council (EPSRC) doctoral training grant (EP/T517951/1).

Optical motion capture (OMC) is the current gold standard for motion analysis, however measuring patellofemoral kinematics is not possible using the technique. One approach to measuring in-vivo kinematics is to use biplane video X-ray (BVX) and 3D models generated from MRI to track the movement of the patellar. Understanding how the patellar is moving during different loaded dynamic activities can help with understanding the effects of different interventions when treating disease or injury. Objective To develop a protocol and compare patellofemoral kinematics for different activities using biplane video X-ray (BVX) Methods Two healthy volunteers performed level walk, lunge, and stair ascent activities while simultaneous capturing BVX and synchronised OMC. Participants undertook MR imaging (Magnetom 3T Prisma, Siemens) which was manually segmented into 3D bone models (Simpleware Scan IP, Synopsis). Bone position and orientation for the patellar and femur were calculated by manual matching of 3D Bone models to X-Rays (DSX Suite, C-Motion, Inc.). Patellofemoral kinematics were calculated using Visual 3D (C-Motion, Inc.). Results Initial results show that patellar flexion(+) (PF) was greatest during lunge (52.1o) compared with stair ascent (49.4o) and stance phase of gait (5.4o), however stair ascent had the largest PF range of motion (ROM) of 48.8o. The lunge activity had the greatest ROM for patellar lateral rotation (12.8o) compared with stair ascent (8.7o) and gait (3.7o). Patellar lateral (+) tilt was found to be greatest during gait (8.4o) compared with stair ascent (6.7o) and lunge (6.8o). Conclusions These results highlight the variability of patellofemoral kinematics between different loaded dynamic activities. When considering the influence and efficacy of patellofemoral interventions it is important to investigate different activities to fully understand their effects. Future work will look at more dynamic activities and to investigate further the effect of different activities on patellofemoral tracking. Declaration of Interest (b) declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported:I declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research project.

Purpose: Total knee replacement (TKR) surgery is performed to reduce pain and improve the function of the lower limb. Most of the current literature focuses on knee kinematics pre and post-TKR, compared to healthy controls. However, there is limited research exploring hip and ankle kinematics before and after TKR and comparing them to non-pathological subjects (NP). This study aims to explore the differences in hip, knee and ankle kinematics in the sagittal and frontal planes pre to post-TKR and to investigate whether the lower limb joints’ kinematics post-TKR normalise and compare to NP. This knowledge would help to understand to which extent TKR helps to improve the whole lower limb function. Methods: In this longitudinal study, biomechanical data was collected using a CAST marker set on 21 patients before and approximately 12 months post-TKR (22 knees, primary TKR) and 20 NP. All participants gave written consent prior to data collection. Participants walked barefoot at self-selected speed in a motion capture lab (Qualisys camera system, Sweden) over a 10-meter walkway instrumented with 6 force plates (Bertec Inc., Ohio). Additionally, patients’ active-assisted knee range of movement (ROM) was recorded from a seated position for both visits. The Oxford Knee Score (OKS) was recorded pre and post-TKR and for NP. Visual 3D (C-Motion, Inc., MD) was used to calculate hip, knee and ankle kinematics in the sagittal and frontal planes. The median differences between pre and post-TKR data were calculated using the Wilcoxon signed-rank test (p-value ≤ 0.05). The median differences between post-TKR and NP data were determined with the Mann-Whitney U test (p-value ≤ 0.05). Results: Compared to NP (45% women), patients (25% women) pre-TKR were significantly older (median difference 19 years, p<0.001), heavier (median difference 22.2 Kg, p<0.001) and had a higher body mass index (median difference 7.32 Kg/m2, p<0.001). The OKS improved significantly pre to post-TKR (median difference 15.5, p=0.001) but post-TKR, the OKS was not comparable to the NP (median difference 12, p<0.001). The active-assisted knee ROM, gait variables, joint angles in the sagittal and frontal planes are displayed in Table 1 for all groups. The significant differences between patients pre to post-TKR and between NP and patients post-TKR are reported in Table 1 and highlighted in dark grey. When compared to pre-TKR, patients post-TKR showed a small but significant increase in sagittal hip ROM (median difference 1.7°, p=0.008) and their value was comparable to NP. However, when looking at post-TKR and NP hip peak angles, patients had a significantly larger hip flexion (median difference 4.3°, p=0.004) and, more prominently, a reduced hip extension (median difference 8.2°, p<0.001) compared to NP. Pre to post-TKR, patients displayed a significantly improved knee extension both in the active-assisted assessment (median difference 3.1°, p=0.004) and especially while walking (median difference 5.9°, p=0.001). This change resulted in a significantly increased sagittal knee ROM during gait (median difference 8.5°, p<0.001). Despite this, after the surgery, patients’ peak knee extension and ROM were still significantly reduced when compared to NP (median differences 6.9°, p<0.001 and 8.1°, p=0.001, respectively). Considering the ankle, patients had a significantly larger sagittal ROM pre to post-TKR (median difference 8.5°, p<0.001), mainly due to a significant increase in peak ankle plantarflexion (median difference 3.1°, p=0.005). When comparing post-TKR patients to NP, the sagittal ankle ROM values were similar but the patient’s peak dorsiflexion was significantly greater (median difference 4.7°, p=0.001) and their peak plantar flexion was significantly smaller (median difference 4.8°, p=0.019) than NP. In the frontal plane, the hip peak angles and ROM did not change significantly pre to post-TKR. When compared to NP, patients showed a reduced peak hip abduction (median difference 3.7°, p=0.002) and frontal hip ROM (median difference 3.6°, p<0.001) post-TKR. Pre to post-TKR there was a significant increase in peak knee abduction (median difference 2.4°, p=0.019), ankle inversion (median difference 1.4°, p=0.031) and a significant decrease in ankle eversion (median difference 2.4°, p=0.051); the post-TKR values for these variables were comparable to those of NP. The frontal ankle ROM did not change significantly pre to post-TKR but it was significantly reduced post-TKR compared to NP (median difference 4.6°, p=0.017). The walking speed increased by 14% post-TKR (median difference 0.13 m/s, p<0.001) but patients’ gait speed was 25% slower than NP one year after the surgery (median difference 0.31 m/s, p<0.001). The walking cycle duration was 6.9% longer in patients post-TKR compared to NP (median difference 0.08 s, p<0.001), despite the 5.7% decrease after the surgery (median difference 0.07 s, p,0.001). The stance time did not change pre to post-TKR and it was significantly longer in the patients’ group post-surgery when compared to NP (median difference 2.4%, p<0.001). Conclusions: Patients who underwent TKR showed an increased active-assisted knee extension and sagittal ROM one year after the surgery; this reflected positively in the knee extension and sagittal ROM during gait post-TKR. Pre to post-TKR, patients exhibited increased sagittal hip, ankle and mostly knee ROMs during gait, with augmented knee extension and ankle plantarflexion. In the frontal plane, only the knee peak valgus angle and ankle inversion increased and ankle eversion decreased pre to post-TKR. Patients did not display a return to healthy joints’ kinematics one year after TKR and this may be because the NP group in this study was much younger than the patients’ one. The sagittal hip and ankle ROM were comparable to NP but they were shifted towards the flexion range with markedly reduced extension at the hip, and towards the dorsiflexion range with reduced plantarflexion at the ankle. Patients showed a stiffer knee during gait as their knee sagittal ROM and peak extension were markedly limited compared to NP. Additionally, patients post-TKR had limited hip and ankle mobility in the frontal plane, unlike NP. The current investigation showed that patients had an improvement in hip, knee and ankle kinematics following TKR, mainly in the sagittal plane and at the knee joint. However, the lower limb kinematics was not comparable to that of NP in patients post-TKR, whose largest limitations were observed in hip and knee extension and knee ROM in the sagittal plane, and hip and ankle ROM in the frontal plane. This study suggests that it is important to consider hip and ankle joints alongside the knee when exploring patients’ kinematics to evaluate TKR outcomes and during the rehabilitation process. More research is needed to determine which are the factors that could affect hip and ankle mobility following TKR surgery.

Skeletal kinematics are traditionally measured by motion analysis methods such as optical motion capture (OMC). While easy to carry out and clinically relevant for certain applications, it...

The purpose of this study was to quantify the effect of total knee replacement (TKR) alignment on in-vivo knee function and loading in a unique patient cohort who have been identified as having a high rate of component mal-alignment. Post-TKR (82.4 ± 6.7 months), gait analysis was performed on 25 patients (27 knees), to calculate knee kinematics and kinetics. For a step activity, video fluoroscopic analysis quantified in-vivo implant kinematics. Frontal plane lower-limb alignment was defined by the Hip-Knee-Ankle angle (HKA) measured on long leg static X-rays. Transverse plane component rotation was calculated from computed tomography scans. Sagittal plane alignment was defined by measuring the flexion angle of the femoral component and the posterior tibial slope angle (PTSA). For gait analysis, a more varus HKA correlated with increased peak and dynamic joint kinetics, predicting 47.6% of Knee Adduction Angular Impulse variance. For the step activity, during step-up and single leg loaded, higher PTSA correlated with a posterior shift in medial compartment Anterior-Posterior (AP) translation. During step-down, higher PTSA correlated with reduced lateral compartment AP translation with a posterior shift in AP translation in both compartments. A more varus HKA correlated with a more posterior medial AP translation and inter-component rotation was related to transverse plan range of motion. This in-vivo study found that frontal plane lower-limb alignment had a significant effect on joint forces during gait but had minimal influence on in-vivo implant kinematics for step activity. PTSA was found to influence in-vivo TKR translations and is therefore an important surgical factor.

Purpose: Varus knee deformity can significantly increase medial knee joint loading that has been demonstrated to accelerate knee degeneration in presence of osteoarthritis (OA). High tibial osteotomy (HTO) surgery is performed in early stages of knee OA with the aim to unload the medial compartment of the tibiofemoral joint and slow down knee degeneration. Therefore, it is clinically important to measure joint loading to better understand whether the clinical aim of the surgery is achieved. External knee adduction moments (EKAM) are routinely used to infer medial tibiofemoral joint load. However, EKAM may not reflect internal joint loading. Musculoskeletal simulations have been shown to be of added value to better understand tibiofemoral joint contact forces compared to EKAM. Patient-specific gait analysis in combination with musculoskeletal modeling allows the analysis of knee joint loading in terms of compartmental contact forces and might be more sensitive to investigate changes in the knee loading pre and post HTO. It is of clinical importance to determine objectively the degree of success of HTO surgery by measuring whether normal knee joint contact forces are restored post operation. The purpose of this exploratory study was to implement a subject-specific simulation pipeline to measure tibiofemoral contact forces in patients undergoing HTO and to report preliminary findings on an individual basis. Methods: Three-dimensional gait analysis was performed on 6 patients before and approximately 12 months post HTO surgery using a modified Cleveland marker-set. Subjects walked at a self-selected speed on a 10 m walkway instrumented with force platforms for a minimum of 6 trials. Where data issues or outliers were identified, a minimum of 3 trials were used in the analysis. All data was analysed with an identical musculoskeletal modelling workflow. Tibiofemoral contact forces were calculated in OpenSim using a scaled musculoskeletal model. Patient-specific mechanical tibiofemoral angle was implemented in the pipeline. The model integrated an extended knee model allowing for 6 degrees of freedom of the tibiofemoral joint into a generic full-body model. The generic model was scaled to the subjects' anthropometry. Joint angles were calculated using inverse kinematics and then muscle forces and secondary knee kinematics were estimated using the concurrent optimization of muscle activations and kinematics algorithm. The magnitude of the total, medial and lateral tibiofemoral contact forces were determined during the first and second half of stance. For the same trials, external knee adduction moments (EKAM) were calculated in Visual 3D. For this exploratory study, EKAM and joint contact forces are presented for each participant pre and post HTO. Approval for this study was granted by the Wales Research Ethics Committee 3 (10/MRE09/28) and Cardiff and Vale University Health Board. Written informed consent was obtained from each participant prior to data collection. Results: Mechanical tibiofemoral angle ranged from 5° to 15.4° varus prior to surgery and was corrected to 2.4° varus to 2° valgus post operatively, as shown in Table 1. Table 2 presents individual changes to EKAM and tibiofemoral contact forces pre and post HTO. Participants 3,5 and 6 who underwent the biggest correction in surgery also reduced their total tibiofemoral peak contact force for the first half of stance, as well as reducing their medial tibiofemoral peak contact force. Interestingly, participant 4 showed increases in both EKAM peaks, total tibiofemoral peak forces and medial tibiofemoral peak forces. Conclusions: This study used a musculoskeletal model for the first time on a cohort of individuals who underwent HTO surgery. This exploratory study has shown preliminary data indicating that musculoskeletal modelling can be used as an indication of success of HTO surgery. Future research will be undertaken with a greater sample size, sufficiently powered in order to statistically quantify the significance of HTO surgery on tibiofemoral contact forces.View Large Image Figure ViewerDownload Hi-res image Download (PPT)

One of the main surgical goals when performing a total knee replacement (TKR) is to ensure the implants are properly aligned and correctly sized; however, understanding the effect of alignment and ...

Purpose: Exercise prescription plays a fundamental role in the treatment of knee pathologies such as knee osteoarthritis, where over 90,000 total knee replacement patients receive regular post-surgery physiotherapy each year in the UK. Physiotherapists rely on home-based exercise prescription yet have limited knowledge of patient engagement at home and find it difficult to objectively monitor patient progress, attribute functional improvement (or lack of) to adherence/non-adherence and prescribe personalised interventions. The research vision is to facilitate unobtrusive sensor driven home monitoring/feedback of knee rehabilitation exercises. As a first step, this study sought to fine tune a machine learning algorithm to classify between different knee exercises and understand the impact of different feature selection parameters on classification performance. Methods: 8 volunteers (4 healthy, 4 with self-reported history of knee pain/pathology but not receiving treatment) performed 15 repetitions of 4 knee rehabilitation exercises (sit to stand (STS), knee flexion (KFL), knee extension (KEX) and weight shifting (WSH)) whilst wearing lower limb Xsens inertial sensors sampling at 60 Hz. A total of 85 features were extracted per exercise repetition from tri-axial accelerometer data provided by sensors placed on the foot, shank, thigh and pelvis. These were defined as the 90th percentile spectral edge frequency ((SEF) in X,Y,Z axis) and signal mean, max min, variance, skewness, and kurtosis (all in the X,Y,Z axis) per sensor in addition to repetition length. Participants were split into training and testing datasets using a Leave-One-Group-Out cross validation (cv) where a participant with multiple repetitions represented a group. Within each of the 8 cv-folds, features were scaled, a univariate feature selection method was implemented to reduce the feature set to top ranking features and a linear support vector machine (SVM) classifier was performed to determine how well 4 different knee rehabilitation exercises could be distinguished from each other. This process was repeated using 3 different score function parameters (Mutual Information Classification (MIC), F-value Classification (F-Class) and Chi2) with the number of features being selected ranging from 1 - 10 to determine the optimal score function and number of features whilst optimising classification performance. The F1 score (weighted harmonic mean of precision and recall with a best score of 1) was computed within each cv-fold as an average F1 score across each of the exercises classes and then combined across cv-folds as a median and 1st - 3rd interquartile range. Results: Feature selection methods using the MIC and the F-Class score functions consistently outperformed that using the Chi2 (Figure 1). The MIC approach was chosen as the optimal feature selection method as the F-Class algorithm was marginally outperformed using fewer features with the optimal number of features (ie. fewest whilst retaining classifier performance) being 3. All 8 cv-folds using this method selected Z-axis acceleration 90th percentile SEF of the foot and shank sensors whilst the third feature varied slightly between cv-folds (mean Z-axis acceleration of the foot sensor (n=3/8 cv-folds); Y-axis mean acceleration of the thigh sensor (n=2/8 cv-folds), Z-axis acceleration variance of the foot (n=1/8 cv-folds) or shank (n=1/8 cv-folds) sensors and Z-axis mean acceleration of the shank sensor (n=1/8 cv-folds). Inspection of the individual F1-scores for each exercise using the optimal feature selection method revealed a reduced ability to classify KEX and KFL in comparison to STS and WSH exercises (Figure 2). Conclusions: Changing parameter settings within a univariate feature selection algorithm that identifies top ranking features was found to alter the performance of a linear SVM seeking to classify between 4 knee rehabilitation exercises. This finding confirms the impact that incorrect feature selection parameter settings may have on classification performance. From the parameters and feature selection methods considered, the combination of an MIC score function and the selection of the top 3 features appear optimal for this dataset. Whilst the 3 features selected appeared to consistently classify STS and WSH exercises well, this was not the case for KEX and KFL exercises with both varying considerably between the cv-folds and warrants further investigation. Future work will consider additional features that may improve the ability to discriminate between KEX and KFL exercises. Additional feature selection methods such as recursive feature elimination and removing features with low variance will be investigated to see whether SVM classification performance can be further enhanced.

Accurate measurement of in-vivo joint kinematics is important for understanding normal and pathological knee function and evaluating outcome of surgical procedures. Fluoroscopy and model based image registration (MBIR) provides an accurate and minimally-invasive technique for calculating in-vivo kinematics. This study builds upon existing MBIR protocols and looks at quantifying the errors present in the protocols, with the aim of developing a biplane fluoroscopy system to investigate in-vivo kinematics of the knee. A retrospective single plane fluoroscopy study was performed on a unique TKR patient group with mal-aligned knee replacements to understand the influence of surgical frontal plane alignment and function. Significant interactions between frontal plane alignment and knee joint kinetics and kinematics were detected using marker-based motion capture. While these interactions were not replicated within in-vivo knee kinematics measured during a step-up activity using fluoroscopy and MBIR, results highlighted interactions with other surgical measures of alignment such as posterior tibial slope angle and Hip-Knee-Ankle angle. A study was undertaken to examine in-vivo kinematics using three dimensional (3D) models generated from magnetic resonance imaging (MRI) combined with fluoroscopy and synchronised motion analysis. These studies, in which the fluoroscopy was performed at Llandough Hospital X-ray Department, highlighted key technical limitations associated with the currently adopted protocol, and two primary sources of error in determining in-vivo kinematics; generation of three dimensional (3D) bone models and MBIR processing to calculate in-vivo kinematics. A validation protocol was developed to determine the accuracy of magnetic resonance imaging (MRI) derived 3D bone models. This was performed by imaging five ovine hind limbs using MRI and computed tomography (CT) followed by complete dissection and structured light scanning of the femora and tibiae to calculate the true geometry. The results showed that MRI derived 3D bone models had a RMS error of 0.8 mm when compared with the other modalities. This error was deemed acceptable as it was not larger than the 3D voxel dimension. A validation study was performed to investigate the accuracy of a biplane C-arm system in calculating skeletal kinematics using MBIR. It examined the static and dynamic accuracy associated with using both Sawbones and an ovine hind limb during a simulated step up activity. Three different dynamic velocities were investigated. Errors were shown to increase with higher velocities highlighting the importance of calculating errors during representative dynamic tasks. The results also highlighted important hardware limitations with the C-arm system. An in-house combined motion analysis and biplane fluoroscopy system was established at Cardiff University. An updated and validated MBIR protocol was performed on 5 healthy volunteers during a step up and down task. 3D models of bone and cartilage were used in combination with biplane fluoroscopy images to calculate in-vivo kinematics and estimate contact point positions. The validation and MBIR protocols in this thesis have contributed to the development and understanding of the limitations associated with a new unique bespoke biplane X-ray system being designed and manufactured currently at Cardiff University.

Purpose: Total knee replacements (TKR) are considered to be the gold standard for treating pain and improving function in end stage OA. Despite this there is still up to 20% of patients who are dissatisfied with their outcome and have chronic pain and poor function. During surgery, factors including implant position and alignment can greatly influence outcome for the patient. There is a lot of debate on what is the optimum frontal plane alignment, with the current consensus that neutral alignment should be aimed for. However there is currently a lack of understanding on how the frontal plane alignment influences the biomechanics of the lower limb. Cardiff University has unique access to a group of local patients who have relatively high frequency of poor alignment, and early failure. This provides a rare insight into how mal-alignment can affect patients from a clinical and biomechanical function to assist surgeons by determining the optimum alignment. The main aim was to compare patients with neutral (−2° to 2°), varus (≥2°) and valgus alignment (≤−2°) TKR with healthy volunteers (n = 29). Methods: 26 patient volunteers (12 males, 14 females; mean age 74, range 60–89; average BMI 31.85 ± 6.62) with 29 Kinemax (Stryker) TKR's and 29 healthy volunteers (17 females, 12 males; mean age 46, range 22–72; average BMI 24.70 ± 4.01) were recruited. Patients undertook gait analysis of level walking using 8 Qualysis Pro-Reflex cameras with an instrumented walkway (Bertec, USA) using a modified Helen Hayes marker set. Visual3D (C-motion, Inc.) was used to compute lower limb kinetics and kinematics. Hip-Knee-Ankle (HKA) angle was measured from long leg radiographs to determine long leg alignment in the frontal plane. 3D joint angles and kinetics were time normalized to 101 data points, and then Principal Component Analysis (PCA) was performed (Matlab, Mathworks). This reduced the joint kinetics and kinematics waveforms into new uncorrelated principle components via an orthogonal transformation. A one-way ANOVA was performed on the first 3 PC values (ranked by percentage variance and on average accounted for 90% of the variance of the waveform) for the 24 variables under analysis for the four different groups. Bonferroni post hoc test was used for parametric data and Games–Howell post hoc test for non-parametric variables to see if any groups were significantly different from healthy. Results: All three post TKR patient groups presented with statistically different (P < 0.05) PC values when compared to healthy controls (valgus group – 11 variables; neutral group – 6 variables; varus group – 8 variables). These differences were found at the hip, knee and ankle and across both kinetic and kinematic variables. Noticeable clustering was evident between healthy and a combined patient cohort in scatter plots (PC1 versus PC2). However, clustering between the three patient groups was less visible. For PC1 knee adduction–abduction angle, only the varus group was not statistically different to healthy compared with the other groups. In contrast to the findings of varus and valgus groups, the neutral group was not found to be different to the healthy cohort for three variables (PC1 Ankle internal–external angle; PC2 knee flexor–extensor moment; PC2 hip flexor–extensor moment). It was noted that several of the kinetic moment differences between healthy and valgus groups were representative of off loading/compensatory mechanisms by the valgus groups. Conclusions: PCA has revealed biomechanical differences between TKR patients and healthy controls, confirming a lack of healthy function in patients who undergo TKR, which has previously been reported. Whilst clustering between patient groups was limited, the one-way ANOVA revealed a number of biomechanical differences between each patient group and the healthy cohort that may help to understand the implications of knee mal-alignment on post-surgery hip, knee and ankle function.

Purpose: Over 90,000 osteoarthritis (OA) related TKR surgeries take place across the UK annually, with patients undergoing regular post-surgery physiotherapy that is reliant on home-based exercise rehabilitation and driven by personalised self-management. With poor patient adherence that is difficult to ascertain, clinicians who are challenged to optimise patient outcomes are unable to determine whether improvements (or lack of) can be attributed to an exercise intervention or (non) adherence. There is a clear need for enhanced forms of objectively monitoring patient adherence to home based exercise rehabilitation, providing valuable biomechanical knowledge to clinicians to guide personalised exercise prescription. This could provide rigorous adherence measurements, optimise the rehabilitation process, reduce NHS burden and improve patient satisfaction. This research aims to determine whether the performance of 4 rehabilitation exercises, routinely prescribed to OA patients following TKR, can be objectively distinguished using inertial measurement sensors (IMU's) placed on the lower limbs. Methods: 5 healthy participants (4 males, 1 female; mean age 32.6 ± 11.1 years, height 1.79 ± 0.14 m and mass 82.88 ± 15.93 kg) performed a battery of early phase knee rehabilitation exercises based on the Taxonomy for RehAbilitation of Knee conditions (TRAK). Data was collected for multiple exercises with participants wearing a range of time synchronised biomechanical measurement systems. This study focused on the performance of 1) Knee Flexion in sitting 2) Knee Extension 3) Single Step Down and 4) Sit to Stand, with each participant performing 4 repetitions per exercise and the data collected using lower body IMU sensors. These were placed on the pelvis and bilateral thigh, shank and feet (Xsens, Holland; sampling at 60 Hz). Anthropometric measurements for each participant were combined with IMU data during a static calibration to define the biomechanical model (MVN Studio). 3D hip, knee and ankle joint angles were calculated using the Euler sequence ZXY using the ISB based coordinate system. Joint angle data were processed in Python, with exercise repetitions defined using a detect peaks algorithm. 3D angle data were formatted in Excel, time normalised to 101 points and then Principal Component Analysis (PCA) was performed (Matlab, Mathworks), reducing all joint angle waveforms into new uncorrelated principal components via an orthogonal transformation. Scatterplots of PC1 versus PC2 were used to visually inspect for clustering between the PC values for the 4 exercise groups. A one way ANOVA (SPSS, IBM) was performed on the first 3 PC values (ranked by percentage variance accounted for) for the 9 variables under analysis, with an a priori alpha level of significance set at 0.05. Games-Howell post hoc tests identified variables that were significantly different between exercises. Results: The PC scatterplot representing the hip flexion-extension waveforms produced the most prominent clustering, with all 4 exercise groups easily distinguishable (Fig.1). Whilst multiple statistically significant differences were found between pairs of exercises for individual PC values, only one PC value was statistically different across all exercise pairings (PC1, knee flexion-extension waveform). Conclusions: This study demonstrates the potential to objectively distinguish between different knee rehabilitation exercises using IMU sensors and PCA. It would appear that flexion-extension angles at the hip and knee are most suited for accurate exercise classification and require further investigation. Future work will focus on increasing the healthy cohort sample size and generating a post TKR patient cohort to identify whether similar differentiation between exercises can be established in a pathological cohort, and whether there are functional difference between healthy and post-TKR patients that could be used to map patient progress.