Reduction, classification and ranking of motion analysis data: an application to osteoarthritic and normal knee function data

Assessing functional recovery following total knee replacement surgery using objective classification of level gait data and patient-reported outcome measures

BACKGROUND

Patient recovery can be quantified objectively, via gait analysis, or subjectively, using patient reported outcome measures. Association between these measures would explain the level of disability reported in patient reported outcome measures and could assist with therapeutic decisions.

METHODS

Total knee replacement outcome was assessed using objective classification and patient-reported outcome measures (Knee Outcome Survey and Oxford Knee Scores). A classifier was trained to distinguish between healthy and osteoarthritic characteristics using knee kinematics, ground reaction force and temporal gait data, combined with anthropometric data from 32 healthy and 32 osteoarthritis knees. For the osteoarthritic cohort, classification of 20 subjects quantified changes at up to 3 timepoints post-surgery.

FINDINGS

Osteoarthritic classification was reduced for 17 subjects when comparing pre- to post-operative assessments, however only 6 participants achieved non-pathological classification and only 4 of these were classified as non-pathological at 12 months. In 15 cases, the level of osteoarthritic classification did not decrease between every post-operative assessment. For an individual's recovery, classification outputs correlated (r > 0.5) with knee outcome survey for 75% of patients and oxford knee score for 78% of patients (based on 20 and 9 subjects respectively). Classifier outputs from all visits of the combined total knee replacement sample correlated moderately with knee outcome survey (r > 0.4) and strongly with oxford knee score (r > 0.6).

INTERPRETATION

Biomechanical deficits existed in most subjects despite improvements in Patient Reported Outcome Measures, with larger changes reported subjectively as compared to measured objectively. Objective Classification provides additional insight alongside Patient Reported Outcomes when reporting recovered outcomes.

Deep learning approach for prediction of impact peak appearance at ground reaction force signal of running activity

Protruding impact peak is one of the features of vertical ground reaction force (GRF) that is related to injury risk while running. The present research is dedicated to predicting GRF impact peak appearance by setting a binary classification problem. Kinematic data, namely a number of raw signals in the sagittal plane, collected by the Vicon motion capture system (Oxford Metrics Group, UK) were employed as predictors. Therefore, the input data for the predictive model are presented as a multi-channel time series. Deep learning techniques, namely five convolutional neural network (CNN) models were applied to the binary classification analysis, based on a Multi-Layer Perceptron (MLP) classifier, support vector machine (SVM), logistic regression, k-nearest neighbors (kNN), and random forest algorithms. SVM, logistic regression, and random forest classifiers demonstrated performances that do not statistically significantly differ. The best classification accuracy achieved is 81.09% ± 2.58%. Due to good performance of the models, this study serves as groundwork for further application of deep learning approaches to predicting kinetic information based on this kind of input data.

Knee Joint Biomechanical Gait Data Classification for Knee Pathology Assessment: A Literature Review

Background

The purpose of this study is to review the current literature on knee joint biomechanical gait data analysis for knee pathology classification. The review is prefaced by a presentation of the prerequisite knee joint biomechanics background and a description of biomechanical gait pattern recognition as a diagnostic tool. It is postfaced by discussions that highlight the current research findings and future directions.

Methods

The review is based on a literature search in PubMed, IEEE Xplore, Science Direct, and Google Scholar on April 2019. Inclusion criteria admitted articles, written in either English or French, on knee joint biomechanical gait data classification in general. We recorded the relevant information pertaining to the investigated knee joint pathologies, the participants' attributes, data acquisition, feature extraction, and selection used to represent the data, as well as the classification algorithms and validation of the results.

Results

Thirty-one studies met the inclusion criteria for review.

Conclusions

The review reveals that the importance of medical applications of knee joint biomechanical gait data classification and recent progress in data acquisition technology are fostering intense interest in the subject and giving a strong impetus to research. The review also reveals that biomechanical data during locomotion carry essential information on knee joint conditions to infer an early diagnosis. This survey paper can serve as a useful informative reference for research on the subject.

An expert system feedback tool improves the reliability of clinical gait kinematics for older adults with lower limb osteoarthritis

Recently, an expert system was developed to provide feedback to examiners with the aim of improving reliability of marker-based gait analysis. The purpose of the current study was to evaluate the effectiveness of this novel feedback tool in improving the reliability of gait analysis for individuals with lower limb osteoarthritis. Three-dimensional gait analysis was conducted for n=27 individuals, at two different time points, and during each session the feedback tool was used to refine marker placement. Results for both discrete variables and support vector machine classifications demonstrated improved reliability of the data with the feedback tool.

Abnormal loading and functional deficits are present in both limbs before and after unilateral knee arthroplasty

Unilateral knee replacement is often followed by a contralateral replacement in time and the biomechanics of the other knee before and after knee replacement remains poorly understood. The aim of this paper is to distinguish the features of arthritic gait in the affected and unaffected legs relative to a normal population and to assess the objective recovery of gait function post-operatively, with the aim of defining patients at risk of poor post-operative function. Twenty patients with severe knee OA but no pain or deformity in any other lower limb joint were compared to twenty healthy subjects of the same age. Gait analysis was performed and quadriceps and hamstrings co-contraction was measured. Fifteen subjects returned 1year following knee arthroplasty. Moments and impulses were calculated, principal component analysis was used to analyse the waveforms and a classification technique (the Cardiff Classifier) was used to select the most discriminant data and define functional performance. Comparing pre-operative function to healthy function, classification accuracies for the affected and unaffected knees were 95% and 92.5% respectively. Post-operatively, the affected limb returned to the normal half of the classifier in 8 patients, and 7 of those patients returned to normal function in the unaffected limb. Recovery of normal gait could be correctly predicted 13 out of 15 times at the affected knee, and 12 out of 15 times at the unaffected knee based on pre-operative gait function. Focused rehabilitation prior to surgery may be beneficial to optimise outcomes and protect the other joints following knee arthroplasty.

Comparing multiple correspondence and principal component analyses with biomechanical signals. Example with turning the steering wheel

The purpose of this article is to compare Principal Component Analysis (PCA) and a much less used method, i.e. MCA (Multiple Correspondence Analysis) with data being first changed into membership values to fuzzy space windows. For such a comparison, data from an experimental study about turning the steering wheel is used. In a didactic perspective, this article only considers one multidimensional signal with 5 components: 3 linked to the steering wheel angle and hand positions and 2 to hand effort variables. A discussion weighs out the pros and the cons of both methods with criteria such as the possibility to show complex relational phenomena, the analysis/computing time or the information loss inherent to the averaging stage (in the perspective to analyze several hundreds of large multidimensional signals).

Principal component analysis in ground reaction forces and center of pressure gait waveforms of people with transfemoral amputation

BACKGROUND

The alterations in gait pattern of people with transfemoral amputation leave them more susceptible to musculoskeletal injury. Principal component analysis is a method that reduces the amount of gait data and allows analyzing the entire waveform.

OBJECTIVES

To use the principal component analysis to compare the ground reaction force and center of pressure displacement waveforms obtained during gait between able-bodied subjects and both limbs of individuals with transfemoral amputation.

STUDY DESIGN

This is a transversal study with a convenience sample.

METHODS

We used a force plate and pressure plate to record the anterior-posterior, medial-lateral and vertical ground reaction force, and anterior-posterior and medial-lateral center of pressure positions of 12 participants with transfemoral amputation and 20 able-bodied subjects during gait. The principal component analysis was performed to compare the gait waveforms between the participants with transfemoral amputation and the able-bodied individuals.

RESULTS

The principal component analysis model explained between 74% and 93% of the data variance. In all ground reaction force and center of pressure waveforms relevant portions were identified; and always at least one principal component presented scores statistically different (p < 0.05) between the groups of participants in these relevant portions.

CONCLUSION

Principal component analysis was able to discriminate many portions of the stance phase between both lower limbs of people with transfemoral amputation compared to the able-bodied participants.

CLINICAL RELEVANCE

Principal component analysis reduced the amount of data, allowed analyzing the whole waveform, and identified specific sub-phases of gait that were different between the groups. Therefore, this approach seems to be a powerful tool to be used in gait evaluation and following the rehabilitation status of people with transfemoral amputation.

Relationship between knee adduction moment patterns extracted using principal component analysis and discrete measures with different amplitude normalizations: Implications for knee osteoarthritis progression studies

BACKGROUND

Knee adduction moment discrete features (peaks and impulses) are commonly reported in knee osteoarthritis gait studies, but they do not necessarily capture loading patterns. Principal component analysis extracts dynamic patterns, but can be difficult to interpret. This methodological study determined relationships between external knee adduction moment discrete measures and principal component analysis features, and examined whether amplitude-normalization methods influenced differences in those with knee osteoarthritis who progressed to surgery versus those that did not.

METHODS

54 knee osteoarthritis patients had three-dimensional biomechanical measures assessed during walking. Knee adduction moments were calculated and non-normalized and amplitude-normalized waveforms using two common methods were calculated. Patterns were extracted using principal component analysis. Knee adduction moment peak and impulse were calculated. Correlation coefficients were determined between two knee adduction moment patterns extracted and peak and impulse. T-tests evaluated between-group differences.

FINDINGS

An overall magnitude pattern was correlated with peak (r=0.88-0.90, p<0.05) and impulse (r=0.93, p<0.05). A pattern capturing a difference between early and mid/late -stance knee adduction moment was significantly correlated with peak (r=0.27-0.40, p<0.05), but explained minimal variance. Between-group peak differences were only affected by amplitude-normalization method.

INTERPRETATION

Findings suggest that the overall magnitude knee adduction moment principal pattern does not provide unique information from peak and impulse measures. However, low correlations and minimal variance explained between the pattern capturing ability to unload the joint during mid-stance and the two discrete measures, suggests that this pattern captured a unique waveform feature.

Assessing changes in subjective and objective function from pre- to post-knee arthroplasty using the Cardiff Dempster-Shafer theory classifier

The purpose of this study is to assess changes in subjective and objective function from pre- to post-knee arthroplasty (KA) using a combined classifier technique. Twenty healthy adults (50-80 years) and 31 KA patients (39-81 years) were studied (4 weeks pre- and 6 months post-KA). Questionnaire measures of subjective pain, joint stability, activity and function were collected. Objective functional assessment included goniometry, ultrasound imaging and 3-D motion analysis/inverse modelling of gait and sit-stand. An optimal set of variables were used to classify function using the Cardiff Dempster-Shafer theory (DST) method. Out of sample accuracy of the classifiers ranged between 90% and 94% for segregating healthy individuals and pre-KA patients. Post-KA subjective function improved with 74% classified as healthy. However, there was minimal improvement in objective measures (23% classified as healthy). The novel use of Cardiff DST segregated KA patients from healthy individuals and estimated changes in function from pre- to post-surgery. KA patients had improved pain and function post-operation but objective knee joint measures remained different to healthy individuals.

Influence of wedges on lower limbs' kinematics and net joint moments during healthy elderly gait using principal component analysis

The elderly are susceptible to many disorders that alter the gait pattern and could lead to falls and reduction of mobility. One of the most applied therapeutical approaches to correct altered gait patterns is the insertion of insoles. Principal Component Analysis (PCA) is a powerful method used to reduce redundant information and it allows the comparison of the complete waveform. The purpose of this study was to verify the influence of wedges on lower limbs' net joint moment and range of motion (ROM) during the gait of healthy elderly participants using PCA. In addition, discrete values of lower limbs' peak net moment and ROM were also evaluated. 20 subjects walked with no wedges (control condition) and wearing six different wedges. The variables analyzed were the Principal Components from joint net moments and ROM in the sagittal plane in the ankle and knee and joint net moments in frontal plane in the knee. The discrete variables were peak joint net moments and ROM in sagittal plane in knee and ankle. The results showed the influence of the wedges to be clearer by analyzing through PCA methods than to use discrete parameters of gait curves, where the differences between conditions could be hidden.

Objective classification of scapular kinematics in participants with movement faults of the scapula on clinical assessment

The aim of this study was to assess the potential of employing a classification tool to objectively classify participants with clinically assessed movement faults (MFs) of the scapula. Six participants with a history of shoulder pain with MFs of the scapula and 12 healthy participants with no movement faults (NMFs) performed a flexion movement control test of the scapula, while scapular kinematic data were collected. Principal component scores and discrete kinematic variables were used as input into a classifier. Five out of the six participants with a history of pain were successfully classified as having scapular MFs with an accuracy of 72%. Variables related to the upward rotation of the scapula had the most influence on the classification. The results of the study demonstrate the potential of adopting a multivariate approach in objective classification of participants with altered scapular kinematics in pathological groups.

A CaRBS analysis of hip replacement approaches and non-pathology

This study investigates the differences in hip biomechanics for subjects following a total hip arthroplasty (THA), through the lateral approach (LA) and posterior approach (PA), to those with no pathology (NP). The principal component analysis was performed on two kinematic and two kinetic waveforms (subject-based characteristics) from level gait to identify salient portions of the waveforms for comparison between the subject cohorts. These were classified to identify the differences between post-THA and non-pathological cohorts. The primary technique exposited in the THA analysis is classification and ranking belief simplex (CaRBS). Within the analysis, from the configuration of a CaRBS model, there is discussion on the model fit and contribution of the subject-based characteristics. Where appropriate, comparisons to the CaRBS model are made with the results from a logistic regression (LR) analysis. In terms of model fit, using CaRBS, 24 out of 27 LA/PA subjects (88.89%) and 13 out of 16 NP subjects (81.25%) were correctly classified as exhibiting either post-THA or NP hip functional characteristics during level gait, combining to 86.05% classification accuracy, compared with 81.40% classification accuracy when using LR.

Feature selection using a principal component analysis of the kinematics of the pivot shift phenomenon

The pivot shift test reproduces a complex instability of the knee joint following rupture of the anterior cruciate ligament. The grade of the pivot shift test has been shown to correlate to subjective criteria of knee joint function, return to physical activity and long-term outcome. This severity is represented by a grade that is attributed by a clinician in a subjective manner, rendering the pivot shift test poorly reliable. The purpose of this study was to unveil the kinematic parameters that are evaluated by clinicians when they establish a pivot shift grade. To do so, eight orthopaedic surgeons performed a total of 127 pivot shift examinations on 70 subjects presenting various degrees of knee joint instability. The knee joint kinematics were recorded using electromagnetic sensors and principal component analysis was used to determine which features explain most of the variability between recordings. Four principal components were found to account for most of this variability (69%), with only the first showing a correlation to the pivot shift grade (r = 0.55). Acceleration and velocity of tibial translation were found to be the features that best correlate to the first principal component, meaning they are the most useful for distinguishing different recordings. The magnitudes of the tibial translation and rotation were amongst those that accounted for the least variability. These results indicate that future efforts to quantify the pivot shift should focus more on the velocity and acceleration of tibial translation and less on the traditionally accepted parameters that are the magnitudes of posterior translation and external tibial rotation.

Challenges Facing Quantification of Rat Locomotion along Beams of Varying Widths

Optoelectronic motion capture systems have been widely used to investigate temporal gait parameters in humans and animals in order to understand function and behavioural attributes of different pathologies, e.g. Parkinson's disease (PD). The aim of the present paper was to investigate the practicality of utilising this system to investigate the effects of a unilateral 6-hydroxydopamine (6-OHDA) lesion on rat locomotion while walking on beams of varying widths (graduated, narrow, and wide). Temporal gait parameters of ten male Lister Hooded rats (five controls and five hemiparkinsonian) were observed using passive markers placed in locations that were representative of their four limbs and their body axis. The results demonstrate that marker-based motion capture can provide an effective and simple approach to quantifying temporal gait parameters for rat models of PD. They also reveal how the width of the path affects the locomotion in both experimental cohorts. Such measurements can be compared with human motion analysis to explore correlations between the animal model and human behaviour, which is an important step for translational medicine.