Estimating the Standing Long Jump Length from Smartphone Inertial Sensors through Machine Learning Algorithms.

The length of the standing long jump (SLJ) is widely recognized as an indicator of developmental motor competence or sports conditional performance. This work aims at defining a methodology to allow athletes/coaches to easily measure it using the inertial measurement units embedded on a smartphone. A sample group of 114 trained young participants was recruited and asked to perform the instrumented SLJ task. A set of features was identified based on biomechanical knowledge, then Lasso regression allowed the identification of a subset of predictors of the SLJ length that was used as input of different optimized machine learning architectures. Results obtained from the use of the proposed configuration allow an estimate of the SLJ length with a Gaussian Process Regression model with a RMSE of 0.122 m in the test phase, Kendall's τ < 0.1. The proposed models give homoscedastic results, meaning that the error of the models does not depend on the estimated quantity. This study proved the feasibility of using low-cost smartphone sensors to provide an automatic and objective estimate of SLJ performance in ecological settings.

Impact of Gait Events Identification through Wearable Inertial Sensors on Clinical Gait Analysis of Children with Idiopathic Toe Walking.

Idiopathic toe walking (ITW) is a gait deviation characterized by forefoot contact with the ground and excessive ankle plantarflexion over the entire gait cycle observed in otherwise-typical developing children. The clinical evaluation of ITW is usually performed using optoelectronic systems analyzing the sagittal component of ankle kinematics and kinetics. However, in standardized laboratory contexts, these children can adopt a typical walking pattern instead of a toe walk, thus hindering the laboratory-based clinical evaluation. With these premises, measuring gait in a more ecological environment may be crucial in this population. As a first step towards adopting wearable clinical protocols embedding magneto-inertial sensors and pressure insoles, this study analyzed the performance of three algorithms for gait events identification based on shank and/or foot sensors. Foot strike and foot off were estimated from gait measurements taken from children with ITW walking barefoot and while wearing a foot orthosis. Although no single algorithm stands out as best from all perspectives, preferable algorithms were devised for event identification, temporal parameters estimate and heel and forefoot rocker identification, depending on the barefoot/shoed condition. Errors more often led to an erroneous characterization of the heel rocker, especially in shoed condition. The ITW gait specificity may cause errors in the identification of the foot strike which, in turn, influences the characterization of the heel rocker and, therefore, of the pathologic ITW behavior.

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.

A functional calibration protocol for ankle plantar-dorsiflexion estimate using magnetic and inertial measurement units: Repeatability and reliability assessment.

The ankle joint complex presents a tangled functional anatomy, which understanding is fundamental to effectively estimate its kinematics on the sagittal plane. Protocols based on the use of magnetic and inertial measurement units (MIMUs) currently do not take in due account this factor. To this aim, a joint coordinate system for the ankle joint complex is proposed, along with a protocol to perform its anatomical calibration using MIMUs, consisting in a combination of anatomical functional calibrations of the tibiotalar axis and static acquisitions. Protocol repeatability and reliability were tested according to the metrics proposed in Schwartz et al. (2004) involving three different operators performing the protocol three times on ten participants, undergoing instrumented gait analysis through both stereophotogrammetry and MIMUs. Instrumental reliability was evaluated comparing the MIMU-derived kinematic traces with the stereophotogrammetric ones, obtained with the same protocol, through the linear fit method. A total of 270 gait cycles were considered. Results showed that the protocol was repeatable and reliable for what concerned the operators (0.4 ± 0.4 deg and 0.8 ± 0.5 deg, respectively). Instrumental reliability analysis showed a mean RMSD of 3.0 ± 1.3 deg, a mean offset of 9.4 ± 8.4 deg and a mean linear relationship strength of R2 = 0.88 ± 0.08. With due caution, the protocol can be considered both repeatable and reliable. Further studies should pay attention to the other ankle degrees of freedom as well as on the angular convention to compute them.

Intra-subject approach for gait-event prediction by neural network interpretation of EMG signals.

BACKGROUND: Machine learning models were satisfactorily implemented for estimating gait events from surface electromyographic (sEMG) signals during walking. Most of them are based on inter-subject approaches for data preparation. Aim of the study is to propose an intra-subject approach for binary classifying gait phases and predicting gait events based on neural network interpretation of sEMG signals and to test the hypothesis that the intra-subject approach is able to achieve better performances compared to an inter-subject one. To this aim, sEMG signals were acquired from 10 leg muscles in about 10.000 strides from 23 healthy adults, during ground walking, and a multi-layer perceptron (MLP) architecture was implemented. RESULTS: Classification/prediction accuracy was tested vs. the ground truth, represented by the foot-floor-contact signal provided by three foot-switches, through samples not used during training phase. Average classification accuracy of 96.1 ± 1.9% and mean absolute value (MAE) of 14.4 ± 4.7 ms and 23.7 ± 11.3 ms in predicting heel-strike (HS) and toe-off (TO) timing were provided. Performances of the proposed approach were tested by a direct comparison with performances provided by the inter-subject approach in the same population. Comparison results showed 1.4% improvement of mean classification accuracy and a significant (p < 0.05) decrease of MAE in predicting HS and TO timing (23% and 33% reduction, respectively). CONCLUSIONS: The study developed an accurate methodology for classification and prediction of gait events, based on neural network interpretation of intra-subject sEMG data, able to outperform more typical inter-subject approaches. The clinically useful contribution consists in predicting gait events from only EMG signals from a single subject, contributing to remove the need of further sensors for the direct measurement of temporal data.

Dofetilide-Induced Microvolt T-Wave Alternans.

Dofetilide is an antiarrhythmic drug that selectively inhibits the rapid component of the delayed rectifier potassium current. The administration of dofetilide may cause ventricular arrhythmias and torsade de pointes. Electrocardiographic (ECG) microvolt T-wave alternans (TWA), an electrophysiologic phenomenon consisting in the beat-to-beat alternation of the T-wave amplitude requiring computerized algorithms to be detected, has also been associated to malignant ventricular arrhythmias. Aim of the present study was to evaluate if dofetilide induces TWA during the 24 hours following administration. The study population consisted of 22 healthy subjects ("ECG Effects of Ranolazine, Dofetilide, Verapamil, and Quinidine in Healthy Subjects" database by Physionet) to whom a 500 µg-dose of dofetilide was administered. For each subject, 10 s ECG were acquired at baseline (0.5 hour before dofetilide administration) and at 15 time points during the 24 hours following the drug administration. ECG were then processed for automatic TWA detection by correlation method. In 21 subjects out of 22, after dofetilide administration, TWA significantly increased to a peak value (median TWA values went from 6 µV at baseline to a max 32 µV; p<; 0.05), on average after 5 hours, to then come back to values closer to baseline. Thus, in healthy subjects, dofetilide increases occurrence and levels (6 times baseline value on average) of TWA in the hours following its administration.