Gait assessment in unicompartmental knee arthroplasty patients: Principal component modelling of gait waveforms and clinical status

Identification and interpretation of gait analysis features and foot conditions by explainable AI

Clinical gait analysis is a crucial step for identifying foot disorders and planning surgery. Automating this process is essential for efficiently assessing the substantial amount of gait data. In this study, we explored the potential of state-of-the-art machine learning (ML) and explainable artificial intelligence (XAI) algorithms to automate all various steps involved in gait analysis for six specific foot conditions. To address the complexity of gait data, we manually created new features, followed by recursive feature elimination using Support Vector Machines (SVM) and Random Forests (RF) to eliminate low-variance features. SVM, RF, K-nearest Neighbor (KNN), and Logistic Regression (LREGR) were compared for classification, with a Majority Voting (MV) model combining trained models. KNN and MV achieved mean balanced accuracy, recall, precision, and F1 score of 0.87. All models were interpreted using Local Interpretable Model-agnostic Explanation (LIME) method and the five most relevant features were identified for each foot condition. High success scores indicate a strong relationship between selected features and foot conditions, potentially indicating clinical relevance. The proposed ML pipeline, adaptable for other foot conditions, showcases its potential in aiding experts in foot condition identification and planning surgeries.

Principal Component Analysis of Knee Joint Differences Between Bilateral and Unilateral Total Knee Replacement Patients During Level Walking

Many unilateral total knee replacement (TKR) patients will need a contralateral TKR. Differences in knee joint biomechanics between bilateral patients and unilateral patients are not well established. The purpose of this study was to examine knee joint differences in level walking between bilateral and unilateral patients, and asymptomatic controls, using principal component analysis. Knee joints of 1st replaced limbs of 15 bilateral patients (69.40 ± 5.04 years), 15 replaced limbs of unilateral patients (66.47 ± 6.15 years), and 15 asymptomatic controls (63.53 ± 9.50 years) were analyzed during level walking. Principal component analysis examined knee joint sagittal- and frontal-plane kinematics and moments, and vertical ground reaction force (GRF). A one-way analysis of variance analyzed differences between principal component scores of each group. TKR patients exhibited more flexed and abducted knees throughout stance, decreased sagittal knee range of motion (ROM), increased early-stance adduction ROM, decreased loading-response knee extension and push-off knee flexion moments, decreased loading-response and push-off peak knee abduction moment (KAbM), increased KAbM at midstance, increased midstance vertical GRF, and decreased loading-response and push-off vertical GRF. Additionally, bilateral patients exhibited reduced sagittal knee ROM, increased adduction ROM, decreased sagittal knee moments throughout stance, decreased KAbM throughout stance, an earlier loading-response peak vertical GRF, and a decreased push-off vertical GRF, compared to unilateral patients. TKR patients, especially bilateral patients had stiff knee motion in the sagittal-plane, increased frontal-plane joint laxity, and a quadriceps avoidance gait.

Principal Component Analysis of the Running Ground Reaction Forces With Different Speeds

Ground reaction force (GRF) is a key metric in biomechanical research, including parameters of loading rate (LR), first impact peak, second impact peak, and transient between first and second impact peaks in heel strike runners. The GRFs vary over time during stance. This study was aimed to investigate the variances of GRFs in rearfoot striking runners across incremental speeds. Thirty female and male runners joined the running tests on the instrumented treadmill with speeds of 2.7, 3.0, 3.3, and 3.7 m/s. The discrete parameters of vertical average loading rate in the current study are consistent with the literature findings. The principal component analysis was modeled to investigate the main variances (95%) in the GRFs over stance. The females varied in the magnitude of braking and propulsive forces (PC1, 84.93%), whereas the male runners varied in the timing of propulsion (PC1, 53.38%). The female runners dominantly varied in the transient between the first and second peaks of vertical GRF (PC1, 36.52%) and LR (PC2, 33.76%), whereas the males variated in the LR and second peak of vertical GRF (PC1, 78.69%). Knowledge reported in the current study suggested the difference of the magnitude and patterns of GRF between male and female runners across different speeds. These findings may have implications for the prevention of sex-specific running-related injuries and could be integrated with wearable signals for the in-field prediction and estimation of impact loadings and GRFs.

Utilization of Time Series Tools in Life-sciences and Neuroscience

Time series tools are part and parcel of modern day research. Their usage in the biomedical field; specifically, in neuroscience, has not been previously quantified. A quantification of trends can tell about lacunae in the current uses and point towards future uses. We evaluated the principles and applications of few classical time series tools, such as Principal Component Analysis, Neural Networks, common Auto-regression Models, Markov Models, Hidden Markov Models, Fourier Analysis, Spectral Analysis, in addition to diverse work, generically lumped under time series category. We quantified the usage from two perspectives, one, information technology professionals', other, researchers utilizing these tools for biomedical and neuroscience research. For understanding trends from the information technology perspective, we evaluated two of the largest open source question and answer databases of Stack Overflow and Cross Validated. We quantified the trends in their application in the biomedical domain, and specifically neuroscience, by searching literature and application usage on PubMed. While the use of all the time series tools continues to gain popularity in general biomedical and life science research, and also neuroscience, and so have been the total number of questions asked on Stack overflow and Cross Validated, the total views to questions on these are on a decrease in recent years, indicating well established texts, algorithms, and libraries, resulting in engineers not looking for what used to be common questions a few years back. The use of these tools in neuroscience clearly leaves room for improvement.

Correlations between patient-perceived outcome and objectively-measured biomechanical change following Total Knee Replacement

BACKGROUND

Total Knee Replacement (TKR) surgery is being utilised in a younger, more active population with greater functional expectations. Understanding whether patient-perceived measures of function reflect objective biomechanical measures is critical in understanding whether functional limitations can be adequately captured within a clinical setting.

RESEARCH QUESTION

Do changes in objective gait biomechanics measures reflect patient-reported outcome measures at approximately 12 months following TKR surgery?

METHODS

Three-dimensional gait analysis was performed on 41 patients with OA who were scheduled for TKR surgery, 22 of which have returned for a (9-24 month) follow-up assessment. Principal Component Analysis was used to define features of variation between OA subjects and an additional 31 non-pathological control subjects. These were used to train the Cardiff Classifier, an objective classification technique, and subsequently quantify changes following TKR surgery. Patient-perceived changes were also assessed using the Oxford Knee Score (OKS), Knee Outcome Survey (KOS), and Pain Audit Collection System scores (PACS). Pearson and Spearman correlation coefficients were calculated to establish the relationship between changes in objectively-measured and perceived outcome.

RESULTS

Objective measures of biomechanical change were strongly correlated to changes in OKS(r=-0.695, p < 0.001) and KOS(r=-.810, p < 0.001) assessed outcomes. Pain (PACS) was only related to biomechanical function post-operatively (r=-.623, p = 0.003).

SIGNIFICANCE

In this biomechanics study, the relationship between changes in objective function and patient-reported measures pre to post TKR surgery is stronger than in studies which did not include biomechanics metrics. Quality of movement may hold more significance for a patient's perception of improvement than functional measures which consider only the time taken or distance travelled during functional activities.

Comparison of gait kinetics in total and unicondylar knee replacement surgery

Introduction The aim of this study was to compare kinetical data from gait analysis of patients who have undergone total and uni-condylar knee replacement. Materials and methods Thirteen patients with unilateral total knee arthroplasty (TKA) and 13 unicondylar knee arthroplasty (UKA), were included, all performed by the same surgeon more than one year prior. The Vicon gait analysis system was used. Statistical power was calculated using SPSS. Results No significant difference was found in the spatiotemporal parameters of gait and survival years of the knee prosthesis between the two groups. The UKA group was found to have significantly larger moments than the TKA group in knee adduction on the operated side and knee flexion moment on the unoperated side during the loading phase. The maximum and minimum sagittal plane moments of the operated sides in the TKA group were significantly lower than the unoperated side. The difference was most significant at pre-swing. The maximum and minimum moments on the operated sides in the UKA group were significantly lower for the knee flexion and adduction moments when compared with the unoperated side and were most prevalent during the loading phase. Conclusions These results are relevant in terms of prosthesis wear. The TKA knees had smaller magnitude moments than the UKA knees in the sagittal and coronal planes. This could explain the higher revision rates for UKA. In both groups, the non-operated knees had significantly larger moments than the operated knees, which implies that after unilateral knee replacement of either type, the non-operated knee is being put under greater stress.

Optimisation of a machine learning algorithm in human locomotion using principal component and discriminant function analyses

Assessment methods in human locomotion often involve the description of normalised graphical profiles and/or the extraction of discrete variables. Whilst useful, these approaches may not represent the full complexity of gait data. Multivariate statistical methods, such as Principal Component Analysis (PCA) and Discriminant Function Analysis (DFA), have been adopted since they have the potential to overcome these data handling issues. The aim of the current study was to develop and optimise a specific machine learning algorithm for processing human locomotion data. Twenty participants ran at a self-selected speed across a 15m runway in barefoot and shod conditions. Ground reaction forces (BW) and kinematics were measured at 1000 Hz and 100 Hz, respectively from which joint angles (°), joint moments (N.m.kg-1) and joint powers (W.kg-1) for the hip, knee and ankle joints were calculated in all three anatomical planes. Using PCA and DFA, power spectra of the kinematic and kinetic variables were used as a training database for the development of a machine learning algorithm. All possible combinations of 10 out of 20 participants were explored to find the iteration of individuals that would optimise the machine learning algorithm. The results showed that the algorithm was able to successfully predict whether a participant ran shod or barefoot in 93.5% of cases. To the authors' knowledge, this is the first study to optimise the development of a machine learning algorithm.

Gait analysis of young male patients diagnosed with primary bladder neck obstruction

Primary bladder neck obstruction (PBNO) represents an inappropriate or inadequate relaxation of the bladder neck during micturition. Based on the observation of an increased rate of postural imbalances in male patients with PBNO, we hypothesized a possible role of an unbalanced biomechanics of the pelvis on urethral sphincters activity. Our aim was to identify kinematic imbalances, usually disregarded in PBNO patients, and which could eventually be involved in the etiopathogenesis of the disease. Seven male adult patients (39.6±7.1years) were recruited; in all patients, PBNO was suspected at bladder diary and uroflowmetry, and was endoscopically confirmed with urethroscopy. Participants gait was recorded with a motion capture system (BTS Spa, Italy) to obtain three-dimensional joint angles and gait parameters. Multivariate statistics based on a Principal Component model allowed to assess the similarity of patients' gait patterns with respect to control subjects. The main finding is that patients with PBNO showed significant discordance in the observations at the ankle and pelvis level. Additionally, 6/7 patients demonstrated altered trunk positions compared to normal curves. We suggest that the identified postural imbalances could represent the cause for an anomalous activation of pelvic floor muscles (hypertonia). The consequent urinary sphincters hypercontraction may be responsible for the development of voiding dysfunction in male patients with no significant morphological alterations. Results reinforced the hypothesis of an etiopathogenetic role of postural imbalances on primary bladder neck obstruction in male patients.

Gait comparison of unicompartmental and total knee arthroplasties with healthy controls

Aims

To compare the gait of unicompartmental knee arthroplasty (UKA) and total knee arthroplasty (TKA) patients with healthy controls, using a machine-learning approach.

Patients and Methods

145 participants (121 healthy controls, 12 patients with cruciate-retaining TKA, and 12 with mobile-bearing medial UKA) were recruited. The TKA and UKA patients were a minimum of 12 months post-operative, and matched for pattern and severity of arthrosis, age, and body mass index.

Participants walked on an instrumented treadmill until their maximum walking speed was reached. Temporospatial gait parameters, and vertical ground reaction force data, were captured at each speed. Oxford knee scores (OKS) were also collected. An ensemble of trees algorithm was used to analyse the data: 27 gait variables were used to train classification trees for each speed, with a binary output prediction of whether these variables were derived from a UKA or TKA patient. Healthy control gait data was then tested by the decision trees at each speed and a final classification (UKA or TKA) reached for each subject in a majority voting manner over all gait cycles and speeds. Top walking speed was also recorded.

Results

92% of the healthy controls were classified by the decision tree as a UKA, 5% as a TKA, and 3% were unclassified. There was no significant difference in OKS between the UKA and TKA patients (p = 0.077). Top walking speed in TKA patients (1.6 m/s; 1.3 to 2.1) was significantly lower than that of both the UKA group (2.2 m/s; 1.8 to 2.7) and healthy controls (2.2 m/s; 1.5 to 2.7; p < 0.001).

Conclusion

UKA results in a more physiological gait compared with TKA, and a higher top walking speed. This difference in function was not detected by the OKS.

Cite this article: Bone Joint J 2016;98-B(10 Suppl B):16–21.

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.

The effects of a semi-rigid soled shoe compared to walking barefoot on knee adduction moment

BACKGROUND/PURPOSE

On a background of literature suggesting that certain rigid soled shoes may increase the knee adduction moment during gait this study was performed to look at specific postoperative shoe - the Medishoe. This shoe is used on a daily basis in a district general hospital orthopaedic department for patients post-operatively to protect wounds and fixations.

METHODS

Using force plates and an opto-electronic motion capture system with retroreflective markers the knee adduction moment was estimated in ten healthy subject both with and without the shoe during normal gait. The angle at which the ground reaction acted with respect to the ground in the coronal plane as well as the tibiofemoral angle were also calculated using the Qualsys software - both with and without the Medishoe.

RESULTS

Two-tailed paired t-tests using a 95% confidence interval showed that there was no significant difference between the two groups in the estimated knee adduction moment (p = 0.238), tibiofemoral angle (p = 0.4952) and the angle of the ground reaction force to the ground (p = 0.059).

CONCLUSION

There was no significant difference in the estimated knee adduction moment between the two groups, although there was a statistical trend to an alteration in the angle of the ground reaction force. Further work involving a greater number of subjects and a three dimensional model would further answer the question as to whether these or other post-operative shoes have a significant effect on the knee adduction moment.

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.

Principal component modeling of isokinetic moment curves for discriminating between the injured and healthy knees of unilateral ACL deficient patients

Bilateral knee strength evaluations of unilateral anterior cruciate ligament (ACL) deficient patients using isokinetic dynamometry are commonly performed in rehabilitation settings. The most frequently-used outcome measure is the peak moment value attained by the knee extensor and flexor muscle groups. However, other strength curve features may also be of clinical interest and utility. The purpose of this investigation was to identify, using Principal Component Analysis (PCA), strength curve features that explain the majority of variation between the injured and uninjured knee, and to assess the capabilities of these features to detect the presence of injury. A mixed gender cohort of 43 unilateral ACL deficient patients performed 6 continuous concentric knee extension and flexion repetitions bilaterally at 60°s(-1) and 180°s(-1) within a 90° range of motion. Moment waveforms were analyzed using PCA, and binary logistic regression was used to develop a discriminatory decision rule. For all directions and speeds, a statistically significant overall reduction in strength was noted for the involved knee in comparison to the uninvolved knee. The discriminatory decision rule yielded a specificity and sensitivity of 60.5% and 60.5%, respectively, corresponding to an accuracy of ∼62%. As such, the curve features extracted using PCA enabled only limited clinical usefulness in discerning between the ACL deficient and contra lateral, healthy knee. Improvement in discrimination capabilities may perhaps be achieved by consideration of different testing speeds and contraction modes, as well as utilization of other data analysis techniques.

Three-dimensional knee joint moments during performance of the bodyweight squat: effects of stance width and foot rotation

The purpose of this investigation was to assess the effects of stance width and foot rotation angle on three-dimensional knee joint moments during bodyweight squat performance. Twenty-eight participants performed 8 repetitions in 4 conditions differing in stance or foot rotation positions. Knee joint moment waveforms were subjected to principal component analysis. Results indicated that increasing stance width resulted in a larger knee flexion moment magnitude, as well as larger and phase-shifted adduction moment waveforms. The knee's internal rotation moment magnitude was significantly reduced with external foot rotation only under the wide stance condition. Moreover, squat performance with a wide stance and externally rotated feet resulted in a flattening of the internal rotation moment waveform during the middle portion of the movement. However, it is speculated that the differences observed across conditions are not of clinical relevance for young, healthy participants.

Principal component analysis of knee kinematics and kinetics after anterior cruciate ligament reconstruction

This study compared the gait of 10 subjects with unilateral anterior cruciate ligament (ACL) reconstruction to a group of 12 height- and weight-matched control subjects. The analysis was based on knee flexion, adduction, and internal rotation angles and moments. The objective was to use principal component analysis (PCA) to identify knees of the ACL reconstructed subjects that fell outside normal ranges as determined by control subjects. Gait data were collected on all subjects in a motion analysis laboratory. Principal component (PC) models were developed for each gait measure based on the control subjects' data and used to assess gait waveforms of ACL reconstructed subjects. PCA allows analysis of entire gait waveforms for comparisons. In a sample of 10 ACL reconstructed subjects (7 years after surgery, on average), six of the ACL reconstructed knees had not returned to normal following surgery and eight of the contralateral knees functioned differently from controls. A majority of the differences were noted to occur in the abduction-adduction knee moment with corresponding infrequency in the differences seen in abduction-adduction rotation. PCA enabled us to identify subjects with abnormal gait waveforms as outliers relative to the normal control group.

Using principal component analysis to aid bayesian network development for prediction of critical care patient outcomes

BACKGROUND

Predicting an intensive care unit patient's outcome is highly desirable. An end goal is for computational techniques to provide updated, accurate predictions about changing patient condition using a manageable number of physiologic parameters.

METHODS

Principal component analysis was used to select input parameters for critical care patient outcome models. Vital signs and laboratory values from each patient's hospital stay along with outcomes ("Discharged" vs. "Deceased") were collected retrospectively at a Level I Trauma-Military Medical Center in the southwest; intensive care unit patients were included if they had been admitted for burn, infection, or hypovolemia during a 5-year period ending October 2007. Principal component analysis was used to determine which of the 24 parameters would serve as inputs in a bayesian network developed for outcome prediction.

RESULTS

Data for 581 patients were collected. Pulse pressure, heart rate, temperature, respiratory rate, sodium, and chloride were found to have statistically significant differences between Discharged and Deceased groups for "Hypovolemia" patients. For "Burn" patients, pulse pressure, hemoglobin, hematocrit, and potassium were found to have statistically significant differences. For a "Combined" group, heart rate, temperature, respiratory rate, sodium, and chloride had statistically significant differences. A bayesian network based on these results, developed for the Combined group, achieved an accuracy of 75% when predicting patient outcome.

CONCLUSIONS

Outcome prediction for critical care patients is possible. Future work should explore model development using additional temporal data and should include prospective validation. Such technology could serve as the basis of real-time intelligent monitoring systems for critical patients.

Use of gait summary measures with lower limb amputees

Gait summary measures have been developed as a convenient method to communicate overall gait pathology. These measures are primarily used in the context of paediatric cerebral palsy and their use remains largely untested in other disability groups. This study assessed the suitability of gait summary measures for use with lower limb amputees. Modified (m) versions of three published gait summary measures were investigated - the Gillette Gait Index (mGGI), the Gait Deviation Index (mGDI) and the Gait Profile Score (mGPS) in conjunction with the Movement Analysis Profile (MAP). Twenty unilateral lower limb amputees underwent three-dimensional gait analysis. All measures reported significant differences between levels of amputation on the prosthetic limb. The mGGI and mGPS detected significant differences between the levels of amputation on the intact side, but the mGDI did not. All gait summary measures were moderately to strongly correlated with leg-length normalised self-selected walking speed and strong correlations were reported between all measures. The MAP exposed common strategies in amputee gait and showed that sagittal hip and knee kinematics contributed predominantly to overall gait deviation in this population group. The mGGI, mGDI and mGPS identified, quantified and stratified gait pathology, indicating that any of the gait measures investigated in this study can be applied as outcome measures in research and case management in lower limb amputees.

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.

Differentiation of young and older adult stair climbing gait using principal component analysis

INTRODUCTION

Principal component analysis (PCA) has been used to reduce the volume of gait data and can also be used to identify the differences between populations. This approach has not been used on stair climbing gait data. Our objective was to use PCA to compare the gait patterns between young and older adults during stair climbing.

METHODS

The knee joint mechanics of 30 healthy young adults (23.9 + or - 2.6 years) and 32 healthy older adults (65.5 + or - 5.2 years) were analyzed while they ascended a custom 4-step staircase. The three-dimensional net knee joint forces, moments, and angles were calculated using typical inverse dynamics. PCA models were created for the knee joint forces, moments and angles about the three axes. The principal component scores (PC scores) generated from the model were analyzed for group differences using independent samples t-tests. A stepwise discriminant procedure determined which principal components (PCs) were most successful in differentiating the two groups.

RESULTS

The number of PCs retained for analysis was chosen using a 90% trace criterion. Of the scores generated from the PCA models nine were statistically different (p < .0019) between the two groups, four of the nine PC scores could be used to correctly classify 95% of the original group.

CONCLUSIONS

The PCA and discriminant function analysis applied in this investigation identified gait pattern differences between young and older adults. Identification of stair gait pattern differences between young and older adults could help in understanding age-related changes associated with the performance of the locomotor task of stair climbing.

An application of principal component analysis for lower body kinematics between loaded and unloaded walking

Load carriage is a very common daily activity at home and in the workplace. Generally, the load is in the form of an external load carried by an individual, it could also be the excessive body mass carried by an overweight individual. To quantify the effects of carrying extra weight, whether in the form of an external load or excess body mass, motion capture data were generated for a diverse subject set. This consisted of twenty-three subjects generating one hundred fifteen trials for each loading condition. This study applied principal component analysis (PCA) to motion capture data in order to analyze the lower body gait patterns for four loading conditions: normal weight unloaded, normal weight loaded, overweight unloaded and overweight loaded. PCA has been shown to be a powerful tool for analyzing complex gait data. In this analysis, it is shown that in order to quantify the effects of external loads and/or for both normal weight and overweight subjects, the first principal component (PC1) is needed. For the work in this paper, PCs were generated from lower body joint angle data. The PC1 of the hip angle and PC1 of the ankle angle are shown to be an indicator of external load and BMI effects on temporal gait data.

Gait analysis to classify external load conditions using linear discriminant analysis

There are many instances where it is desirable to determine, at a distance, whether a subject is carrying a hidden load. Automated detection systems based on gait analysis have been proposed to detect subjects that carry hidden loads. However, very little baseline gait kinematic analysis has been performed to determine the load carriage effect while ambulating with evenly distributed (front to back) loads on human gait. The work in this paper establishes, via high resolution motion capture trials, the baseline separability of load carriage conditions into loaded and unloaded categories using several standard lower body kinematic parameters. A total of 23 participants (19 for training and 4 for testing) were studied. Satisfactory classification of participants into the correct loading condition was achieved by employing linear discriminant analysis (LDA). Six lower body kinematic parameters including ranges of motion and path lengths from the phase portraits were used to train the LDA to discriminate loaded and unloaded walking conditions. Baseline performance from 4 participants who were not included in training data sets show that the use of LDA provides a 92.5% correct classification over two loaded and unloaded walking conditions. The results suggest that there are gait pattern changes due to external loads, and LDA could be applied successfully to classify the gait patterns with an unknown load condition.

Application of principal component analysis in vertical ground reaction force to discriminate normal and abnormal gait

Discrete parameters from ground reaction force (GRF) are been considered in gait analysis studies. However, principal component analysis (PCA) may provide additional insight into gait analysis for considering the complete pattern of GRF. This study aimed at testing the application of PCA to discriminate the vertical GRF pattern between control group (CG) and patients with lower limb fractures (FG), as well as proposing a score to quantify the abnormality of gait. Thirty-eight healthy subjects participated of CG and 13 subjects in FG, five subjects from FG were also evaluated after physiotherapeutic treatment (FGA). The GRF was measured by an instrumented treadmill. Principal component coefficients (PCCs) were obtained by singular value decomposition using GRF of complete stride. Two, four and six PCCs were used to obtain the standard distance (D). The classification between groups was mainly given by the first PC, which indicated higher loading factors during push off of affected side and heel strike of unaffected side. The classification performance achieved 92.2% accuracy with two PCCs, 94.1% with four PCCs and 96.1% with six PCCs. Four subjects reached normal boundary after treatment, with all FGA subjects presenting decreased D. This study demonstrates that PCA is an adequate method for discriminating normal and abnormal gait and D allows an objective evaluation of the progress and effectiveness of rehabilitation treatment.

The Gait Deviation Index: a new comprehensive index of gait pathology

This article describes a new multivariate measure of overall gait pathology called the Gait Deviation Index (GDI). The first step in developing the GDI was to use kinematic data from a large number of walking strides to derive a set of mutually independent joint rotation patterns that efficiently describe gait. These patterns are called gait features. Linear combinations of the first 15 gait features produced a 98% faithful reconstruction of both the data from which they were derived and 1000 validation strides not used in the derivation. The GDI was then defined as a scaled distance between the 15 gait feature scores for a subject and the average of the same 15 gait feature scores for a control group of typically developing (TD) children. Concurrent and face validity data for the GDI are presented through comparisons with the Gillette Gait Index (GGI), Gillette Functional Assessment Questionnaire Walking Scale (FAQ), and topographic classifications within the diagnosis of Cerebral Palsy (CP). The GDI and GGI are strongly correlated (r(2)=0.56). The GDI scales with FAQ level, distinguishes levels from one another, and is normally distributed across FAQ levels six to ten and among TD children. The GDI also scales with respect to clinical involvement based on topographic CP classification in Hemiplegia Types I-IV, Diplegia, Triplegia and Quadriplegia. The GDI offers an alternative to the GGI as a comprehensive quantitative gait pathology index, and can be readily computed using the electronic addendum provided with this article.

Analysis of multiple waveforms by means of functional principal component analysis: normal versus pathological patterns in sit-to-stand movement

This paper presents an application of functional principal component analysis (FPCA) to describe the inter-subject variability of multiple waveforms. This technique was applied to the study of sit-to-stand movement in two groups of people, osteoarthritic patients and healthy subjects. Although STS movement has not been extensively applied to the study of knee osteoarthritis, it can provide relevant information about the effect of osteoarthritis on knee joint function. Two waveforms, knee flexion angle and flexion moment, were analysed simultaneously. Instead of using the common multivariate approach we used the functional one, which allows working with continuous functions with neither discretization nor time-scale normalization. The results show that time-scale normalization can alter the FPCA solution. Furthermore, FPCA presents better discriminatory power compared with the classical multivariate approach. This technique can, therefore, be applied as a functional assessment tool, allowing the identification of relevant variables to discriminate heterogeneous groups such as healthy and pathological subjects.

A new approach for quantitative analysis of inter-joint coordination during gait

A new method for quantitative analysis of interjoint coordination at various walking speeds is presented. The model imposed a parametric relationship among lower limb joint motions (hips and knees) using the least number of parameters. An integration of different analysis tools such as harmonic analysis, principal component analysis, and artificial neural networks helped overcome high-dimensionality, temporal dependence, and nonlinear relationships of the gait patterns. The trained model was fed only two control parameters (cadence and stride length) for each gait cycle and predicted the corresponding gait waveforms. Based on the differences between predicted and actual gait waveforms, a coordination score, which ranged from 0 to 10, was defined at various walking speeds. The model was applied to eight patients with knee arthroplasty at different follow-ups as well as to eight healthy subjects, walking at three different speeds. The results showed that knee replacement and rehabilitation programs improved the coordination score. The technique provides an analytical tool that can be used as a routine test in the clinical evaluation of human gait abnormalities.

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

There are certain major obstacles to using motion analysis as an aid to clinical decision making. These include: the difficulty in comprehending large amounts of both corroborating and conflicting information; the subjectivity of data interpretation; the need for visualization; and the quantitative comparison of temporal waveform data. This paper seeks to overcome these obstacles by applying a hybrid approach to the analysis of motion analysis data using principal component analysis (PCA), the Dempster-Shafer (DS) theory of evidence and simplex plots. Specifically, the approach is used to characterise the differences between osteoarthritic (OA) and normal (NL) knee function data and to produce a hierarchy of those variables that are most discriminatory in the classification process. Comparisons of the results obtained with the hybrid approach are made with results from artificial neural network analyses.

An extended index to quantify normality of gait in children

Clinical gait analysis aims to quantify and assess the mechanics of walking and identify deviations from 'normal' movement patterns. To facilitate the use of clinical equipment, protocols are required to process data and produce a few meaningful summary measurements which can, in turn, be used to flag gait abnormalities. Earlier work produced a one-dimensional index of gait, calculated from sagittal hip, knee and ankle rotation angle patterns. The objective of this study was to extend the original index, incorporating kinematic and kinetic data from multiple planes, while allowing for correlations between component measures. A one-dimensional index of normal gait was developed, based on normative gait data (N=45 children, aged 3-13 years). The new one-dimensional index was calculated using correlation patterns between seven component indices, each of which has diagnostic interpretation. The effectiveness of the new index was tested using immature normative data (N=14) and hypotonic data (N=10). Approximately 85% of immature normative children and 100% of hypotonic children were classified as either unusual or extreme by the one-dimensional index. These data reduction protocols improve objective gait analyses in the clinical setting.

Biomechanical features of gait waveform data associated with knee osteoarthritis: an application of principal component analysis

This study compared the gait of 50 patients with end-stage knee osteoarthritis to a group of 63 age-matched asymptomatic control subjects. The analysis focused on three gait waveform measures that were selected based on previous literature demonstrating their relevance to knee osteoarthritis (OA): the knee flexion angle, flexion moment, and adduction moment. The objective was to determine the biomechanical features of these gait measures related to knee osteoarthritis. Principal component analysis was used as a data reduction tool, as well as a preliminary step for further analysis to determine gait pattern differences between the OA and the control groups. These further analyses included statistical hypothesis testing to detect group differences, and discriminant analysis to quantify overall group separation and to establish a hierarchy of discriminatory ability among the gait waveform features. The two groups were separated with a misclassification rate (estimated by cross-validation) of 8%. The discriminatory features of the gait waveforms were, in order of their discriminatory ability: the amplitude of the flexion moment, the range of motion of the flexion angle, the magnitude of the flexion moment during early stance, and the magnitude of the adduction moment during stance.

Knee biomechanics of moderate OA patients measured during gait at a self-selected and fast walking speed

Osteoarthritis (OA) is a chronic disorder resulting in degenerative changes to the knee joint. Three-dimensional gait analysis provides a unique method of measuring knee dynamics during activities of daily living such as walking. The purpose of this study was to identify biomechanical features characterizing the gait of patients with mild-to-moderate knee OA and to determine if the biomechanical differences become more pronounced as the locomotor system is stressed by walking faster. Principal component analysis was used to compare the gait patterns of a moderate knee OA group (n=41) and a control group (n=43). The subjects walked at their self-selected speed as well as at 150% of that speed. The two subject groups did not differ in knee joint angles, stride length, and stride time or walking speed. Differences in the magnitude and shape of the knee joint moment waveforms were found between the two groups. The OA group had larger adduction moment magnitudes during stance and this higher magnitude was sustained for a longer portion of the gait cycle. The OA group also had a reduced flexion moment and a reduced external rotation moment during early stance. Increasing speed was associated with an increase in the magnitude of all joint moments. The fast walks did not, however, increase or bring out any biomechanical differences between the OA and control groups that did not exist at the self-selected walks.

Automated feature assessment in instrumented gait analysis

A methodological modular framework is presented for automated assessment of gait patterns. The processing steps of data selection, gait parameter calculation and evaluation are not limited to a specific field of application and are largely independent of case-based clinical expert knowledge. For these steps, a variety of mathematical methods was used and the validity of the approach to assess gait parameters tested by applying it to the clinical problem of Botulinum Toxin A (BTX-A) treatment of the spastic equinus foot. A set of 3670 parameters was ranked by relevance for classification of a group of 42 diplegic cerebral palsy (CP) patients and an age-matched reference group. The same procedure was performed for pre- and post-therapeutic data sets of these patients. Gait parameters of high relevance coincided well with results of previous studies based on partly manual and more subjective parameter selection. A norm distance measure is introduced to facilitate the quantification of deviations from a normal walking pattern and can be used as an overall scalar measure to evaluate differences in gait patterns or as a set of measures attributing each joint angle separately.

Principal component analysis of vertical ground reaction force: a powerful method to discriminate normal and abnormal gait and assess treatment

This study aims at testing the application of principal component analysis (PCA) in the ground reaction force (GRF) in discriminating the gait pattern between normal and abnormal subjects, and assessing the rehabilitation treatment. The sample was composed by 31 subjects, organized into two groups: a control group (CG) of 25 normal and a group (FG) of six patients with lower limb fractures, which was considered before (FGB) and after (FGA) a treadmill physiotherapeutic treatment. The vertical component of GRF data was collected with an instrumentized treadmill. PCA method was applied and the first two coefficients (PCC) were obtained for the three groups. The region of CG values was separated in the PCC plane with the elliptical area of displacement and with a linear threshold between CG and FGB obtained by stepwise logistic regression. Results show that all values of FGA moved towards CG region from the corresponding FGB position, indicating the potential power of PCA in discriminating between normal and abnormal gait and objectively evaluating the effects of rehabilitation treatment.

Changes in frontal plane dynamics and the loading response phase of the gait cycle are characteristic of severe knee osteoarthritis application of a multidimensional analysis technique

BACKGROUND

Osteoarthritis of the knee is related to many correlated mechanical factors that can be measured with gait analysis. Gait analysis results in large data sets. The analysis of these data is difficult due to the correlated, multidimensional nature of the measures.

METHODS

A multidimensional model that uses two multivariate statistical techniques, principal component analysis and discriminant analysis, was used to discriminate between the gait patterns of the normal subject group and the osteoarthritis subject group. Nine time varying gait measures and eight discrete measures were included in the analysis. All interrelationships between and within the measures were retained in the analysis.

FINDINGS

The multidimensional analysis technique successfully separated the gait patterns of normal and knee osteoarthritis subjects with a misclassification error rate of <6%. The most discriminatory feature described a static and dynamic alignment factor. The second most discriminatory feature described a gait pattern change during the loading response phase of the gait cycle.

INTERPRETATION

The interrelationships between gait measures and between the time instants of the gait cycle can provide insight into the mechanical mechanisms of pathologies such as knee osteoarthritis. These results suggest that changes in frontal plane loading and alignment and the loading response phase of the gait cycle are characteristic of severe knee osteoarthritis gait patterns. Subsequent investigations earlier in the disease process may suggest the importance of these factors to the progression of knee osteoarthritis.

PCA in studying coordination and variability: a tutorial

OBJECTIVE

To explain and underscore the use of principal component analysis in clinical biomechanics as an expedient, unbiased means for reducing high-dimensional data sets to a small number of modes or structures, as well as for teasing apart structural (invariant) and variable components in such data sets.

DESIGN

The method is explained formally and then applied to both simulated and real (kinematic and electromyographic) data for didactical purposes, thus illustrating possible applications (and pitfalls) in the study of coordinated movement.

BACKGROUND

In the sciences at large, principal component analysis is a well-known method to remove redundant information in multidimensional data sets by means of mode reduction. At present, principal component analysis is starting to penetrate the fundamental and clinical study of human movement, which amplifies the need for an accessible explanation of the method and its possibilities and limitations. Besides mode reduction, we discuss principal component analysis in its capacity as a data-driven filter, allowing for a separation of invariant and variant properties of coordination, which, arguably, is essential in studies of motor variability.

METHODS

Principal component analysis is applied to kinematic and electromyographic time series obtained during treadmill walking by healthy humans.

RESULTS

Common signal structures or modes are identified in the time series that turn out to be readily interpretable. In addition, the identified coherent modes are eliminated from the data, leaving a filtered, residual pattern from which useful information may be gleaned regarding motor variability.

CONCLUSIONS

Principal component analysis allows for the detection of modes (information reduction) in both kinematic and electromyographic data sets, as well as for the separation of invariant structure and variance in those data sets.

RELEVANCE

Principal component analysis can be successfully applied to movement data, both as feature extractor and as data-driven filter. Its potential for the (clinical) study of human movement sciences (e.g., diagnostics and evaluation of interventions) is evident but still largely untapped.

A review of analytical techniques for gait data. Part 2: neural network and wavelet methods

Multivariate gait data have traditionally been challenging to analyze. Part 1 of this review explored applications of fuzzy, multivariate statistical and fractal methods to gait data analysis. Part 2 extends this critical review to the applications of artificial neural networks and wavelets to gait data analysis. The review concludes with a practical guide to the selection of alternative gait data analysis methods. Neural networks are found to be the most prevalent non-traditional methodology for gait data analysis in the last 10 years. Interpretation of multiple gait signal interactions and quantitative comparisons of gait waveforms are identified as important data analysis topics in need of further research.

A review of analytical techniques for gait data. Part 1: Fuzzy, statistical and fractal methods

In recent years, several new approaches to gait data analysis have been explored, including fuzzy systems, multivariate statistical techniques and fractal dynamics. Through a critical survey of recent gait studies, this paper reviews the potential of these methods to strengthen the gait laboratory's analytical arsenal. It is found that time-honoured multivariate statistical methods are the most widely applied and understood. Although initially promising, fuzzy and fractal analyses of gait data remain largely unknown and their full potential is yet to be realized. The trend towards fusing multiple techniques in a given analysis means that additional research into the application of these two methods will benefit gait data analysis.