Gait Analysis Using a Support Vector Machine for Lumbar Spinal Stenosis

Walking stress-induced changes in gait patterns and muscle activity: Patients with lumbar spinal stenosis versus asymptomatic controls

INTRODUCTION

Patients with symptomatic lumbar spinal stenosis (sLSS) are often limited in their walking range because of worsening symptoms, which is thought to induce changes in the gait pattern. The aim of this study was to determine whether changes in gait pattern and muscle activity in these patients are elicited by a walking stress and differ from asymptomatic controls.

METHODS

Twenty patients with sLSS and 19 asymptomatic controls performed a 30-minute walking stress. Gait was assessed using seven inertial sensors and sagittal joint range of motion (ROM) was calculated during different phases of gait. Muscle activation of the gluteus medius, erector spinae and multifidus muscles was measured by surface electromyography (EMG) and integrated EMGs (normalized to the maximum during gait) were calculated. Differences between groups and time points (beginning and end) were assessed using mixed factorial analysis of variance.

RESULTS

Patients had less knee extension ROM in terminal stance, less knee flexion ROM in swing and less overall hip flexion/extension ROM than controls (p ≤ 0.03). There were no functionally relevant changes in these parameters during the walking stress. The integrated EMG was greater in all muscles in patients than in controls and increased in both groups during the walking stress in the paraspinal but not in the gluteus medius muscle. There was no interaction between group and time for any of the parameters.

CONCLUSION

Differences in gait pattern and muscle activity between patients with sLSS and controls are generally present, but are not amplified by a walking stress.

Artificial intelligence: a new cutting-edge tool in spine surgery

The purpose of this narrative review was to comprehensively elaborate the various components of artificial intelligence (AI), their applications in spine surgery, practical concerns, and future directions. Over the years, spine surgery has been continuously transformed in various aspects, including diagnostic strategies, surgical approaches, procedures, and instrumentation, to provide better-quality patient care. Surgeons have also augmented their surgical expertise with rapidly growing technological advancements. AI is an advancing field that has the potential to revolutionize many aspects of spine surgery. We performed a comprehensive narrative review of the various aspects of AI and machine learning in spine surgery. To elaborate on the current role of AI in spine surgery, a review of the literature was performed using PubMed and Google Scholar databases for articles published in English in the last 20 years. The initial search using the keywords "artificial intelligence" AND "spine," "machine learning" AND "spine," and "deep learning" AND "spine" extracted a total of 78, 60, and 37 articles and 11,500, 4,610, and 2,270 articles on PubMed and Google Scholar. After the initial screening and exclusion of unrelated articles, duplicates, and non-English articles, 405 articles were identified. After the second stage of screening, 93 articles were included in the review. Studies have shown that AI can be used to analyze patient data and provide personalized treatment recommendations in spine care. It also provides valuable insights for planning surgeries and assisting with precise surgical maneuvers and decisionmaking during the procedures. As more data become available and with further advancements, AI is likely to improve patient outcomes.

Classification of inertial sensor-based gait patterns of orthopaedic conditions using machine learning: A pilot study

Elderly patients often have more than one disease that affects walking behavior. An objective tool to identify which disease is the main cause of functional limitations may aid clinical decision making. Therefore, we investigated whether gait patterns could be used to identify degenerative diseases using machine learning. Data were extracted from a clinical database that included sagittal joint angles and spatiotemporal parameters measured using seven inertial sensors, and anthropometric data of patients with unilateral knee or hip osteoarthritis, lumbar or cervical spinal stenosis, and healthy controls. Various classification models were explored using the MATLAB Classification Learner app, and the optimizable Support Vector Machine was chosen as the best performing model. The accuracy of discrimination between healthy and pathologic gait was 82.3%, indicating that it is possible to distinguish pathological from healthy gait. The accuracy of discrimination between the different degenerative diseases was 51.4%, indicating the similarities in gait patterns between diseases need to be further explored. Overall, the differences between pathologic and healthy gait are distinct enough to classify using a classical machine learning model; however, routinely recorded gait characteristics and anthropometric data are not sufficient for successful discrimination of the degenerative diseases.

Kinematics and paraspinal muscle activation patterns during walking differ between patients with lumbar spinal stenosis and controls

BACKGROUND

The narrowing of the spinal canal due to degenerative processes may lead to symptomatic lumbar spinal stenosis (sLSS) and impairments in the patients' gait. Changes in lower extremity joint kinematics and trunk flexion angles have been reported, yet less is known about muscle activation patterns of paraspinal and gluteal muscles in patients with sLSS compared to healthy participants.

RESEARCH QUESTION

Do muscle activation patterns together with sagittal joint kinematics differ between patients with sLSS and healthy controls and do these differences-quantified using gait scores-correlate with clinical scores?

METHODS

In 20 patients with sLSS scheduled for surgery and 19 healthy participants, gait was assessed using seven inertial sensors and muscle activation of gluteus medius, erector spinae and multifidus using wireless surface electromyography (EMG). Differences in joint kinematics and EMG patterns were assessed using statistical parametric mapping with non-parametric independent sample t tests (P < 0.05). Gait scores that describe the overall deviation in joint angles (mGPS) and muscle activation patterns (EMG-Profile Score) were calculated as root mean square distances between patients and healthy participants and their associations with clinical scores (pain, Oswestry Disability Score (ODI)) were analyzed using Spearman's correlation coefficients rho (P < 0.05).

RESULTS

Patients had larger mGPS (+1.9°) and EMG-Profile Scores (+50%) and walked on average slower (-0.26 m/s) than controls. EMG patterns revealed higher activation of multifidus, erector spinae and gluteus medius during midstance in patients compared to controls. Clinical scores (pain, ODI) did not correlate with mGPS or EMG-Profile Scores within patients.

SIGNIFICANCE

Observed differences in gait and muscle activation patterns and in the summary scores of gait and EMG deviations between patients with sLSS and healthy controls may represent additional functional outcomes reflecting the functional status of patients that can be measured using wearable sensors and hence is suitable for application in clinical practice.

Gait asymmetry of lower extremities reduced immediately after minimally invasive surgery among patients with lumbar disc herniation

BACKGROUND

Lumbar disc herniation patients with increased pain exhibit greater gait asymmetry in stance time, swing time and single support time. Percutaneous endoscopic lumbar discectomy, as a minimally invasive surgical procedure has been used to treat patients with lumbar disc herniation. The objective of this study was to evaluate the immediate impact of the percutaneous endoscopic lumbar discectomy on gait asymmetry in spatiotemporal and kinetic parameters among lumbar disc herniation patients.

METHODS

Marker trajectories and ground reaction forces were measured during walking among 67 lumbar disc herniation patients and 15 healthy controls. Spatiotemporal gait parameters were analyzed via Visual3D. Muscle force and joint contact force were calculated with OpenSim. Gait asymmetry of those parameters were assessed with asymmetry index.

FINDINGS

After surgery, gait asymmetry in gait cycle time, step length, peak biceps femoris long head, tensor fasciae latae and rectus femoris muscle forces, and peak hip and knee joint contact forces reduced immediately. Postoperatively, increased gait cycle time and decreased step length were found in the affected side. Moreover, decreased peak biceps femoris long head, tensor fasciae latae and rectus femoris muscle forces, and peak hip joint contact force were observed in the contralateral side.

INTERPRETATION

These results suggested compensation strategy that biceps femoris long head, tensor fasciae latae and rectus femoris in the contralateral side were mainly used to compensate the affected side preoperatively in lumbar disc herniation patients, with less compensation between lower limbs after surgery, which may provide an insight into postoperative rehabilitation.

Quantitative Gait Analysis of Patients with Severe Symptomatic Spinal Stenosis Utilizing the Gait Profile Score: An Observational Clinical Study

Lumbar spine stenosis (LSS) typically manifests with neurogenic claudication, altering patients’ gait. The use of optoelectronic systems has allowed clinicians to perform 3D quantitative gait analysis to quantify and understand these alterations. Although several authors have presented analysis of spatiotemporal gait parameters, data concerning kinematic parameters is lacking. Fifteen patients with LSS were matched with 15 healthy controls. Quantitative gait analysis utilizing optoelectronic techniques was performed for each pair of subjects in a specialized laboratory. Statistical comparison of patients and controls was performed to determine differences in spatiotemporal parameters and the Gait Profile Score (GPS). Statistically significant differences were found between patient and control groups for all spatiotemporal parameters. Patients had significantly different overall GPS (p = 0.004) and had limited internal/external pelvic rotation (p < 0.001) and cranial/caudal movement (p = 0.034), limited hip extension (p = 0.012) and abduction/adduction (p = 0.012) and limited ankle plantar flexion (p < 0.001). In conclusion, patients with LSS have significantly altered gait patterns in three regions (pelvis, hip and ankle) compared to healthy controls. Analysis of kinematic graphs has given insight into gait pathophysiology of patients with LSS and the use of GPS will allow us to quantify surgical results in the future.

Cervical Myelopathy Screening with Machine Learning Algorithm Focusing on Finger Motion Using Noncontact Sensor

STUDY DESIGN

Cross-sectional study.

OBJECTIVE

To develop a binary classification model for cervical myelopathy (CM) screening based on a machine learning algorithm using Leap Motion (Leap Motion, San Francisco, CA), a novel noncontact sensor device.

SUMMARY OF BACKGROUND DATA

Progress of CM symptoms are gradual and cannot be easily identified by the patients themselves. Therefore, screening methods should be developed for patients of CM before deterioration of myelopathy. Although some studies have been conducted to objectively evaluate hand movements specific to myelopathy using cameras or wearable sensors, their methods are unsuitable for simple screening outside hospitals because of the difficulty in obtaining and installing their equipment and the long examination time.

METHODS

In total, 50 and 28 participants in the CM and control groups were recruited, respectively. The diagnosis of CM was made by spine surgeons. We developed a desktop system using Leap Motion that recorded 35 parameters of fingertip movements while participants gripped and released their fingers as rapidly as possible. A support vector machine was used to develop the binary classification model, and a multiple linear regression analysis was performed to create regression models to estimate the total Japanese Orthopaedic Association (JOA) score and the JOA score of the motor function of the upper extremity (MU-JOA score).

RESULTS

The binary classification model indexes were as follows: sensitivity, 84.0%; specificity, 60.7%; accuracy, 75.6%; area under the curve, 0.85. The Spearman rank correlation coefficient between the estimated score and the total JOA score was 0.44 and that between the estimated score and the MU-JOA score was 0.51.

CONCLUSION

Our binary classification model using a machine learning algorithm and Leap Motion could classify CM with high sensitivity and would be useful for CM screening in daily life before consulting doctors and telemedicine.Level of Evidence: 3.

Changes in kinematics, kinetics, and muscle activity in patients with lumbar spinal stenosis during gait: systematic review

BACKGROUND CONTEXT

Lumbar spinal stenosis (LSS) is one of the most common orthopaedic conditions and affects more than half a million people over the age of 65 in the US. Patients with LSS have gait dysfunction and movement deficits due to pain and symptoms caused by compression of the nerve roots within a narrowed spinal canal.

PURPOSE

The purpose of the current systematic review was to summarize existing literature reporting biomechanical changes in gait function that occur with LSS, and identify knowledge gaps that merit future investigation in this important patient population.

STUDY DESIGN/SETTING

This study is a systematic literature review.

OUTCOME MEASURES

The current study included biomechanical variables (e.g., kinematic, kinetic, and muscle activity parameters).

METHODS

Relevant articles were selected through MEDLINE, Scopus, Embase, and Web of Science. Articles were included if they: 1) included participants with LSS or LSS surgery, 2) utilized kinematic, kinetic, or muscle activity variables as the primary outcome measure, 3) evaluated walking or gait tasks, and 4) were written in English.

RESULTS

A total of 11 articles were included in the current systematic review. The patients with LSS exhibited altered gait function as compared to healthy controls. Improvements in some biomechanical variables were found up to one year after surgery, but most gait changes were found within one month after surgery.

CONCLUSIONS

Although numerous studies have investigated gait function in patients with LSS, gait alterations in joint kinetics and muscle activity over time remain largely unknown. In addition, there are limited findings of spinal kinematics in patients with LSS during gait. Thus, future investigations are needed to investigate longer-term gait changes with regard to spinal kinematics, joint kinetics, and muscle activity beyond one month after LSS surgery.

Physical Abilities in Low Back Pain Patients: A Cross-Sectional Study with Exploratory Comparison of Patient Subgroups

An abundance of literature has investigated the association between low back pain (LBP) and physical ability or function. It has been shown that LBP patients display reduced range of motion, decreased balance ability, impaired proprioception, and lower strength compared to asymptomatic persons. The aim of this study was to investigate the differences between LBP patients and healthy controls in terms of several physical abilities. Based on the premised that different biomechanical and physiological causes and consequences could be related to different types of LBP, a secondary exploratory attempt of the study was to examine the differences between LBP subgroups based on the pain location (local or referred) or type of pathology (discogenic or degenerative) on the level of impairment of function and ability. Participants performed range of motion tests, trunk maximal voluntary contraction force tests, a sitting balance assessment, the timed up-and-go test, the chair rise test, and the trunk reposition error test. Compared to the control group, symptomatic patients on average showed 45.7% lower trunk extension (p < 0.001, η² = 0.33) and 27.7 % lower trunk flexion force (p < 0.001, η² = 0.37) during maximal voluntary contraction. LBP patients exhibited decreased sitting balance ability and lower scores in mobility tests (all p < 0.001). There were no differences between groups in Schober’s test and trunk repositioning error (p > 0.05). No differences were observed among the LBP subgroups. The exploratory analyses are limited by the sample size and uncertain validity of the diagnostic procedures within this study. Further studies with appropriate diagnostic procedures and perhaps a different subgrouping of the LBP patients are needed to elucidate if different types of LBP are related to altered biomechanics, physiology, and function.

Kinematic biomarkers of chronic neck pain measured during gait: A data-driven classification approach

People with chronic neck pain (CNP) often present with altered gait kinematics. This paper investigates, combines, and compares the kinematic features from linear and nonlinear walking trajectories to design supervised machine learning models which differentiate asymptomatic individuals from those with CNP. For this, 126 features were extracted from seven body segments of 20 asymptomatic subjects and 20 individuals with non-specific CNP. Neighbourhood Component Analysis (NCA) was used to identify body segments and the corresponding significant features which have the maximum discriminative power for conducting classification. We assessed the efficacy of NCA combined with K- Nearest Neighbour (K-NN), Support Vector Machine and Linear Discriminant Analysis. By applying NCA, all classifiers increased their performance for both linear and nonlinear walking trajectories. Notably, features selected by NCA which magnify the classification power of the computational model were solely from the head, trunk and pelvis kinematics. Our results revealed that the nonlinear trajectory provides the best classification performance through the NCA-K-NN algorithms with an accuracy of 90%, specificity of 100% and sensitivity of 83.3%. The selected features by NCA are introduced as key biomarkers of gait kinematics for classifying non-specific CNP. This paper provides insight into changes in gait kinematics which are present in people with non-specific CNP which can be exploited for classification purposes. The result highlights the importance of curvilinear gait kinematic features which potentially could be utilized in future research to predict recurrent episodes of neck pain.

Feature Analysis of Smart Shoe Sensors for Classification of Gait Patterns

Gait analysis is commonly used to detect foot disorders and abnormalities such as supination, pronation, unstable left foot and unstable right foot. Early detection of these abnormalities could help us to correct the walking posture and avoid getting injuries. This paper presents extensive feature analyses on smart shoes sensor data, including pressure sensors, accelerometer and gyroscope signals, to obtain the optimum combination of the sensors for gait classification, which is crucial to implement a power-efficient mobile smart shoes system. In addition, we investigated the optimal length of data segmentation based on the gait cycle parameters, reduction of the feature dimensions and feature selection for the classification of the gait patterns. Benchmark tests among several machine learning algorithms were conducted using random forest, k-nearest neighbor (KNN), logistic regression and support vector machine (SVM) algorithms for the classification task. Our experiments demonstrated the combination of accelerometer and gyroscope sensor features with SVM achieved the best performance with 89.36% accuracy, 89.76% precision and 88.44% recall. This research suggests a new state-of-the-art gait classification approach, specifically on detecting human gait abnormalities.

Analysis of Dynamic Plantar Pressure before and after the Occurrence of Neurogenic Intermittent Claudication in Patients with Lumbar Spinal Stenosis: An Observational Study

Lumbar spinal stenosis (LSS) is a common disease in the elderly population; it has been reported that patients with LSS have an abnormal gait pattern due to symptom such as neurogenic intermittent claudication (NIC); however, no detailed reports exist on the plantar pressure distributions in LSS patients with NIC. To analysis the plantar pressure characteristics of LSS patients, the Footscan® pressure system was used to perform dynamic plantar pressure measurements in 20 LSS patients (age, 69.5 ± 7.2 years) before and after the occurrence of NIC. The contact time (CT), foot progression angle (FPA), pressure-time integral (PTI), and contact area (CA) were collected and compared between the LSS patients and age-matched healthy subjects in each measurement. The LSS group showed an increase in forefoot CT%, PTI, and CA% in both measurements compared with those in the control group. After the occurrence of NIC in the LSS group, CT%, PTI, and CA% of the forefoot increased further compared with those before the occurrence of NIC. In addition, after the occurrence of NIC, the PTI and CA% of the forefoot shifted from the medial foot to the lateral foot. The results suggested that the plantar pressure distributions of patients with LSS differs from normal subjects due to the posture of waking with lumbar forward flexion, and the forefoot bears a higher relative load. In addition, the occurrence of NIC could affect the plantar pressure distribution of the patients with LSS, predicting the patient's risk of falling to the anterior direction and to the symptomatic side.

Usefulness of a combined approach of DIERS Formetric 4D® and QUINTIC gait analysis system to evaluate the clinical effects of different spinal diseases on spinal-pelvic-lower limb motor function

BACKGROUND

To investigate the alterations in body movement and their compensatory characteristics under different spinal diseases through an objective and quantitative analysis of the spinal-pelvic-lower limb motor function.

METHODS

A total of 120 subjects were recruited from October 2016 to April 2017. The patients were classified into 2 groups in which 65 patients with cervical spondylotic myelopathy (CSM) and 25 patients with idiopathic scoliosis (IS). The former group was evaluated with JOA score while those in the IS group underwent Lenke classification. A control group was set up with 30 healthy subjects. All the subjects were instructed to walk at a constant speed for one minute on a treadmill, and their spinal-pelvic-lower limb motions were monitored simultaneously with a DIERS Formetric 4D® grating system and a QUINTIC gait analysis system.

RESULTS

The rotation angle of thoracic and lumbar vertebrae in IS group were larger than those in the control group (P < 0.05), and the knee joint angle A in the CSM group and IS group were larger than the control group (P < 0.05). In the CSM group, the knee joint angular velocity and angular acceleration were both greater than the control group (P < 0.05). And there was a negative linear correlation between the JOA score for the lower extremity of CSM patients and their knee joint angular acceleration.

CONCLUSION

IS patients tend to demonstrate increased swing amplitude of the trunk. Those with CSM will also have larger knee joint angular velocity and angular acceleration.

Artificial intelligence and machine learning in spine research

Artificial intelligence (AI) and machine learning (ML) techniques are revolutionizing several industrial and research fields like computer vision, autonomous driving, natural language processing, and speech recognition. These novel tools are already having a major impact in radiology, diagnostics, and many other fields in which the availability of automated solution may benefit the accuracy and repeatability of the execution of critical tasks. In this narrative review, we first present a brief description of the various techniques that are being developed nowadays, with special focus on those used in spine research. Then, we describe the applications of AI and ML to problems related to the spine which have been published so far, including the localization of vertebrae and discs in radiological images, image segmentation, computer-aided diagnosis, prediction of clinical outcomes and complications, decision support systems, content-based image retrieval, biomechanics, and motion analysis. Finally, we briefly discuss major ethical issues related to the use of AI in healthcare, namely, accountability, risk of biased decisions as well as data privacy and security, which are nowadays being debated in the scientific community and by regulatory agencies.

An automatic non-invasive method for Parkinson's disease classification

BACKGROUND AND OBJECTIVE

The automatic noninvasive identification of Parkinson's disease (PD) is attractive to clinicians and neuroscientist. Various analysis and classification approaches using spatiotemporal gait variables have been presented earlier in classifying Parkinson's gait. In this paper, we present a wavelet transform based representation of spatiotemporal gait variables to explore the potential of such representation in the identification of Parkinson's gait.

METHODS

Here, we present wavelet analysis as an alternate method and show that wavelet analysis combined with support vector machine (SVM) can produce efficient classification accuracy. Computationally simplified features are extracted from the wavelet transformation and are fed to support vector machine for Parkinson's gait identification. We have assessed various gait parameters namely stride interval, swing interval, and stance interval (from both legs) to observe the best single parameter for such classification.

RESULTS

By employing wavelet decomposition of the gait variables as an alternate method for the identification of Parkinson's subjects, the classification accuracy of 90.32% (Confidence Interval; 74.2%-97.9%) has been achieved, at par to recently reported accuracy, using only one gait parameter. Left stance interval performed equally good to Right swing interval showing classification accuracy of 90.32%. The classification accuracy improved to 100% when all the gait parameters from left leg were put together to form a larger feature vector. We have shown that Haar wavelet performed significantly better than db2 wavelet (p = 0.05) for certain gait variables e.g., right stride time series. The results show that wavelet analysis is a promising approach in reducing down the required number of gait variables, however at the cost of increased computations in wavelet analysis.

CONCLUSIONS

In this work a wavelet transform approach is explored to classify Parkinson's subjects and healthy subjects using their gait cycle variables. The results show that the proposed method can efficiently extract relevant features from the different levels of the wavelet towards the classification of Parkinson's and healthy subjects and thus, the present work is a potential candidate for the automatic noninvasive neurodegenerative disease classification.