Editorial: Machine Learning Approaches to Human Movement Analysis

A Review of Kinematic Theories and Practices Compiled for Biomechanics Students and Researchers

The topic of kinematics is fundamental to engineering and has a significant bearing on clinical evaluations of human movement. For those studying biomechanics, this topic is often overlooked in importance. The degree to which kinematic fundamentals are included in Biomedical engineering (BmE) curriculums is not consistent across programs and often foundational understandings are gained only after reading literature if a research or development project requires that knowledge. The purpose of this paper is to present the important theories and methods of kinematic analysis and synthesis that should be in the "toolbox" of students of biomechanics. Each topic is briefly presented accompanied by an example or two. Deeper learning of each topic is left to the reader, with the help of some sample references to begin that journey.

An IMU-Based Wearable System for Respiratory Rate Estimation in Static and Dynamic Conditions

PURPOSE

Breathing parameters change with activity and posture, but currently available solutions can perform measurements only during static conditions.

METHODS

This article presents an innovative wearable sensor system constituted by three inertial measurement units to simultaneously estimate respiratory rate (RR) in static and dynamic conditions and perform human activity recognition (HAR) with the same sensing principle. Two units are aimed at detecting chest wall breathing-related movements (one on the thorax, one on the abdomen); the third is on the lower back. All units compute the quaternions describing the subject's movement and send data continuously with the ANT transmission protocol to an app. The 20 healthy subjects involved in the research (9 men, 11 women) were between 23 and 54 years old, with mean age 26.8, mean height 172.5 cm and mean weight 66.9 kg. Data from these subjects during different postures or activities were collected and analyzed to extract RR.

RESULTS

Statistically significant differences between dynamic activities ("walking slow", "walking fast", "running" and "cycling") and static postures were detected (p < 0.05), confirming the obtained measurements are in line with physiology even during dynamic activities. Data from the reference unit only and from all three units were used as inputs to artificial intelligence methods for HAR. When the data from the reference unit were used, the Gated Recurrent Unit was the best performing method (97% accuracy). With three units, a 1D Convolutional Neural Network was the best performing (99% accuracy).

CONCLUSION

Overall, the proposed solution shows it is possible to perform simultaneous HAR and RR measurements in static and dynamic conditions with the same sensor system.

Machine learning aided jump height estimate democratization through smartphone measures

Introduction

The peak height reached in a countermovement jump is a well established performance parameter. Its estimate is often entrusted to force platforms or body-worn inertial sensors. To date, smartphones may possibly be used as an alternative for estimating jump height, since they natively embed inertial sensors.

Methods

For this purpose, 43 participants performed 4 countermovement jumps (172 in total) on two force platforms (gold standard). While jumping, participants held a smartphone in their hands, whose inertial sensor measures were recorded. After peak height was computed for both instrumentations, twenty-nine features were extracted, related to jump biomechanics and to signal time-frequency characteristics, as potential descriptors of soft tissues or involuntary arm swing artifacts. A training set (129 jumps - 75%) was created by randomly selecting elements from the initial dataset, the remaining ones being assigned to the test set (43 jumps - 25%). On the training set only, a Lasso regularization was applied to reduce the number of features, avoiding possible multicollinearity. A multi-layer perceptron with one hidden layer was trained for estimating the jump height from the reduced feature set. Hyperparameters optimization was performed on the multi-layer perceptron using a grid search approach with 5-fold cross validation. The best model was chosen according to the minimum negative mean absolute error.

Results

The multi-layer perceptron greatly improved the accuracy (4 cm) and precision (4 cm) of the estimates on the test set with respect to the raw smartphone measures estimates (18 and 16 cm, respectively). Permutation feature importance was performed on the trained model in order to establish the influence that each feature had on the outcome. The peak acceleration and the braking phase duration resulted the most influential features in the final model. Despite not being accurate enough, the height computed through raw smartphone measures was still among the most influential features.

Discussion

The study, implementing a smartphone-based method for jump height estimates, paves the way to method release to a broader audience, pursuing a democratization attempt.

Measuring posture change to detect emotional conditions for workers: A scoping review

BACKGROUND: The emotional management of workers can not only increase the efficiency of work, but also contribute to the improvement of the productivity of a company. OBJECTIVE: This scoping review surveyed the literature to identify the relationship between postural expression and emotion during sedentary tasks. METHODS: We searched relevant literature published up to December 1, 2019 using seven electronic databases (PubMed, CINAHL, Embase, Web of science, PsycINFO, IEEE Xplore, and Medline complete). RESULTS: A total of 14 publications were included in this scoping review. It was found that the application of pressure sensor and camera-based measurement equipment was effective. Additionally, it was proposed to predict the emotional state of the worker by using forward and backward movements as the main variable as opposed to left and right movements. The information-based analysis technique was able to further increase the accuracy of workers’ emotion prediction. CONCLUSIONS: The emotion prediction of workers based on sitting posture could be confirmed for certain movements, and the information-based technical method could further increase the accuracy of prediction. Expansion of information-based technical research will further increase the possibility of predicting the emotions of workers based on posture, and this will in turn promote safer and more efficient work performance.

Always Pay Attention to Which Model of Motor Learning You Are Using

This critical review considers the epistemological and historical background of the theoretical construct of motor learning for a more differentiated understanding. More than simply reflecting critically on the models that are used to solve problems-whether they are applied in therapy, physical education, or training practice-this review seeks to respond constructively to the recent discussion caused by the replication crisis in life sciences. To this end, an in-depth review of contemporary motor learning approaches is provided, with a pragmatism-oriented clarification of the researcher's intentions on fundamentals (what?), subjects (for whom?), time intervals (when?), and purpose (for what?). The complexity in which the processes of movement acquisition, learning, and refinement take place removes their predictable and linear character and therefore, from an applied point of view, invites a great deal of caution when trying to make generalization claims. Particularly when we attempt to understand and study these phenomena in unpredictable and dynamic contexts, it is recommended that scientists and practitioners seek to better understand the central role that the individual and their situatedness plays in the system. In this way, we will be closer to making a meaningful and authentic contribution to the advancement of knowledge, and not merely for the sake of renaming inventions.

Artificial Neural Network Analyzing Wearable Device Gait Data for Identifying Patients With Stroke Unable to Return to Work

A potential dramatic effect of long-term disability due to stroke is the inability to return to work. An accurate prognosis and the identification of the parameters inflating the possibility of return to work after neurorehabilitation are crucial. Many factors may influence it, such as mobility and, in particular, walking ability. In this pilot study, two emerging technologies have been combined with the aim of developing a prognostic tool for identifying patients able to return to work: a wearable inertial measurement unit for gait analysis and an artificial neural network (ANN). Compared with more conventional statistics, the ANN showed a higher accuracy in identifying patients with respect to healthy subjects (90.9 vs. 75.8%) and also in identifying the subjects unable to return to work (93.9 vs. 81.8%). In this last analysis, the duration of double support phase resulted the most important input of the ANN. The potentiality of the ANN, developed also in other fields such as marketing on social networks, could allow a powerful support for clinicians that today should manage a large amount of instrumentally recorded parameters in patients with stroke.