Determinants of gait stability while walking on a treadmill: A machine learning approach

Use of subject-specific models to detect fatigue-related changes in running biomechanics: a random forest approach

Running biomechanics are affected by fatiguing or prolonged runs. However, no evidence to date has conclusively linked this effect to running-related injury (RRI) development or performance implications. Previous investigations using subject-specific models in running have demonstrated higher accuracy than group-based models, however, this has been infrequently applied to fatigue. In this study, two experiments were conducted to determine whether subject-specific models outperformed group-based models to classify running biomechanics during non-fatigued and fatigued conditions. In the first experiment, 16 participants performed four treadmill runs at or around the maximal lactate steady state. In the second experiment, nine participants performed five prolonged runs using commercial wearable devices. For each experiment, two segments were extracted from each trial from early and late in the run. For each participant, a random forest model was applied with a leave-one-run-out cross-validation to classify between the early (non-fatigued) and late (fatigued) segments. Additionally, group-based classifiers with a leave-one-subject-out cross validation were constructed. For experiment 1, mean classification accuracies for the single-subject and group-based classifiers were 68.2 ± 8.2% and 57.0 ± 8.9%, respectively. For experiment 2, mean classification accuracies for the single-subject and group-based classifiers were 68.9 ± 17.1% and 61.5 ± 11.7%, respectively. Variable importance rankings were consistent within participants, but these rankings differed from each participant to those of the group. Although the classification accuracies were relatively low, these findings highlight the advantage of subject-specific classifiers to detect changes in running biomechanics with fatigue and indicate the potential of using big data and wearable technology approaches in future research to determine possible connections between biomechanics and RRI.

Classification of Ataxic Gait

Gait disorders accompany a number of neurological and musculoskeletal disorders that significantly reduce the quality of life. Motion sensors enable high-quality modelling of gait stereotypes. However, they produce large volumes of data, the evaluation of which is a challenge. In this publication, we compare different data reduction methods and classification of reduced data for use in clinical practice. The best accuracy achieved between a group of healthy individuals and patients with ataxic gait extracted from the records of 43 participants (23 ataxic, 20 healthy), forming 418 segments of straight gait pattern, is 98% by random forest classifier preprocessed by t-distributed stochastic neighbour embedding.

Severity Classification of Parkinson’s Disease Based on Permutation-Variable Importance and Persistent Entropy

Parkinson’s disease (PD) is a neurodegenerative disease that causes chronic and progressive motor dysfunction. As PD progresses, patients show different symptoms at different stages of the disease. The severity assessment is inefficient and subjective when it comes to artificial diagnosis. However, abnormal gait was contingent and the subject selection was limited. Therefore, few-shot learning based on small sample sets is critical to solving the problem of insufficient sample data in PD patients. Using datasets from PhysioNet, this paper presents a method based on permutation-variable importance (PVI) and persistent entropy of topological imprints, and uses support vector machine (SVM) as a classifier to achieve the severity classification of PD patients. The method includes the following steps: (1) Take the data as gait cycles, and calculate the gait characteristics of each cycle. (2) Use the random forest (RF) method to obtain the leading factors differentiating the gait of patients at different severity levels. (3) Use time-delay embedding to map the data into a topological space, and use the topological data analysis based on permutation homology to obtain the persistent entropy. (4) Use the Borderline-SMOTE (BSM) method to balance the sample data. (5) Use the SVM to classify the samples for the severity levels of PD. An accuracy of 98.08% was achieved by 10-fold cross-validation, so our method can be used as an effective means of computer-aided diagnosis of PD, and has important practical value.

Deep Learning for Fall Risk Assessment With Inertial Sensors: Utilizing Domain Knowledge in Spatio-Temporal Gait Parameters

Fall risk assessment is essential to predict and prevent falls in geriatric populations, especially patients with life-long conditions like neurological disorders. Inertial sensor-based pervasive gait analysis systems have become viable means to facilitate continuous fall risk assessment in non-hospital settings. However, a gait analysis system is not sufficient to detect the characteristics leading to increased fall risk, and powerful inference models are required to detect the underlying factors specific to fall risk. Machine learning models and especially the recently proposed deep learning methods offer the needed predictive power. Deep neural networks have the potential to produce models that can operate directly on the raw data, thus alleviating the need for feature engineering. However, the domain knowledge inherent in the well-established spatio-temporal gait parameters are still valuable to help a model achieve high inference accuracies. In this study, we explore deep learning methods, specifically long short-term memory (LSTM) neural networks, for the problem of fall risk assessment. We utilize sequences of spatio-temporal gait parameters extracted by an inertial sensor-based gait analysis system as input features. To quantify the performance of the proposed approach, we compare it with more traditional machine learning methods. The proposed LSTM model, trained with a gait dataset collected from 60 neurological disorder patients, achieves a superior classification accuracy of 92.1% on a separate test dataset collected from 16 patients. This study serves as one of the first attempts to employ deep learning approaches in this domain and the results demonstrate their potential.

Postural control in healthy adults: Determinants of trunk sway assessed with a chest-worn accelerometer in 12 quiet standing tasks

Many diseases and conditions decrease the ability to control balance. In clinical settings, there is therefore a major interest in the assessment of postural control. Trunk accelerometry is an easy, low-cost method used for balance testing and constitutes an alternative method to the posturography using force platforms. The objective was to assess the responsiveness of accelerometry in a battery of 12 quiet standing tasks. We evaluated the balance of 100 healthy adults with an accelerometer fixed onto the sternum. We used the average amplitude of acceleration as an indirect measure of postural sways. The tasks of increased difficulty were realized with or without vision. The battery of tasks was repeated four times on two different days to assess reliability. We analyzed the extent to which the task difficulty and the absence of vision affected the trunk sway. The influence of individual characteristics (age, height, mass, sex, and physical activity level) was also assessed. The reliability analysis revealed that four repetitions of the battery of tasks are needed to reach a high accuracy level (mean ICC = 0.85). The results showed that task difficulty had a very large effect on trunk sways and that the removal of vision further increased sways. Concerning the effects of individual characteristics, we observed that women tended to oscillate more than men did in tasks of low difficulty. Age and physical activity level also had significant effects, whereas height and mass did not. In conclusion, age, sex, and physical fitness are confounders that should be considered when assessing patients’ balance. A battery of simple postural tasks measured by upper-trunk accelerometry can be a useful method for simple balance evaluation in clinical settings.

Subject-specific and group-based running pattern classification using a single wearable sensor

The objective of this study was to determine whether subject-specific or group-based models provided better classification accuracy to identify changes in biomechanical running gait patterns across different inclination conditions. The classification process was based on measurements from a single wearable sensor using a total of 41,780 strides from eleven recreational runners while running in real-world and uncontrolled environment. Biomechanical variables included pelvic drop, ground contact time, braking, vertical oscillation of pelvis, pelvic rotation, and cadence were recorded during running on three inclination grades: downhill, -2° to -7°; level, -0.2° to +0.2°; and uphill, +2° to +7°. An ensemble and non-linear machine learning algorithm, random forest (RF), was used to classify inclination condition and determine the importance of each of the biomechanical variables. Classification accuracy was determined for subject-specific and group-based RF models. The mean classification accuracy of all subject-specific RF models was 86.29%, while group-based classification accuracy was 76.17%. Braking was identified as the most important variable for all the runners using the group-based model and for most of the runners based on a subject-specific models. In addition, individual runners used different strategies across different inclination conditions and the ranked order of variable importance was unique for each runner. These results demonstrate that subject-specific models can better characterize changes in gait biomechanical patterns compared to a more traditional group-based approach.

Using wearable sensors to classify subject-specific running biomechanical gait patterns based on changes in environmental weather conditions

Running-related overuse injuries can result from a combination of various intrinsic (e.g., gait biomechanics) and extrinsic (e.g., running surface) risk factors. However, it is unknown how changes in environmental weather conditions affect running gait biomechanical patterns since these data cannot be collected in a laboratory setting. Therefore, the purpose of this study was to develop a classification model based on subject-specific changes in biomechanical running patterns across two different environmental weather conditions using data obtained from wearable sensors in real-world environments. Running gait data were recorded during winter and spring sessions, with recorded average air temperatures of -10° C and +6° C, respectively. Classification was performed based on measurements of pelvic drop, ground contact time, braking, vertical oscillation of pelvis, pelvic rotation, and cadence obtained from 66,370 strides (~11,000/runner) from a group of recreational runners. A non-linear and ensemble machine learning algorithm, random forest (RF), was used to classify and compute a heuristic for determining the importance of each variable in the prediction model. To validate the developed subject-specific model, two cross-validation methods (one-against-another and partitioning datasets) were used to obtain experimental mean classification accuracies of 87.18% and 95.42%, respectively, indicating an excellent discriminatory ability of the RF-based model. Additionally, the ranked order of variable importance differed across the individual runners. The results from the RF-based machine-learning algorithm demonstrates that processing gait biomechanical signals from a single wearable sensor can successfully detect changes to an individual's running patterns based on data obtained in real-world environments.

The Maximum Lyapunov Exponent During Walking and Running: Reliability Assessment of Different Marker-Sets

The maximum Lyapunov exponent (MLE) has often been suggested as the prominent measure for evaluation of dynamic stability of locomotion in pathological and healthy population. Although the popularity of the MLE has increased in the last years, there is scarce information on the reliability of the method, especially during running. The purpose of the current study was, thus, to examine the reliability of the MLE during both walking and running. Sixteen participants walked and ran on a treadmill completing two measurement blocks (i.e., two trials per day for three consecutive days per block) separated by 2 months on average. Six different marker-sets on the trunk were analyzed. Intraday, interday and between blocks reliability was assessed using the intraclass correlation coefficient (ICC) and the root mean square difference (RMSD). The MLE was on average significantly higher (p < 0.001) in running (1.836 ± 0.080) compared to walking (1.386 ± 0.207). All marker-sets showed excellent ICCs (>0.90) during walking and mostly good ICCs (>0.75) during running. The RMSD ranged from 0.023 to 0.047 for walking and from 0.018 to 0.050 for running. The reliability was better when comparing MLE values between blocks (ICCs: 0.965–0.991 and 0.768–0.961; RMSD: 0.023–0.034 and 0.018–0.027 for walking and running respectively), and worse when considering trials of the same day (ICCs: 0.946–0.980 and 0.739–0.844; RMSD: 0.042–0.047 and 0.045–0.050 for walking and running respectively). Further, different marker-sets affect the reliability of the MLE in both walking and running. Our findings provide evidence that the assessment of dynamic stability using the MLE is reliable in both walking and running. More trials spread over more than 1 day should be considered in study designs with increased demands of accuracy independent of the locomotion condition.