Abstract
Gait performance is an important marker of motor and cognitive decline in older adults. An instrumented gait analysis resorting to inertial sensors allows the complete evaluation of spatiotemporal gait parameters, offering an alternative to laboratory-based assessments. To estimate gait parameters, foot trajectories are typically obtained by integrating acceleration two times. However, to deal with cumulative integration errors, additional error handling strategies are required. In this study, we propose an alternative approach based on a deep recurrent neural network to estimate heel and toe trajectories. We propose a coordinate frame transformation for stride trajectories that eliminates the dependency from previous strides and external inputs. Predicted trajectories are used to estimate an extensive set of spatiotemporal gait parameters. We evaluate the results in a dataset comprising foot-worn inertial sensor data acquired from a group of young adults, using an optical motion capture system as a reference. Heel and toe trajectories are predicted with low errors, in line with reference trajectories. A good agreement is also achieved between the reference and estimated gait parameters, in particular when turning strides are excluded from the analysis. The performance of the method is shown to be robust to imperfect sensor-foot alignment conditions.

Abstract
This study presents principal component analysis (PCA) intra-subject variability of lower limb surface electromyography (sEMG) at different muscle stretch-shortening cycle (SSC). Several key steps are presented on the research of muscle force production for human in-vivo and noninvasive studies as well as on SSC contribution at gait, run, and jump with the need for separation of muscle and tendon behavior. Complexity and unpredicted multiple muscle actuation are highlighted with the need for extraction of PCA components from muscle stretch-shortening cycle sEMG, namely on lower limb stereotyped muscle patterns assessed on standard maximum vertical jump (MVJ). The purpose of this study is to apply PCA to sEMG linear envelopes of lower limb selected muscles at different MVJ, to detect lower number of components explaining maximum sEMG variability, representative of low dimensional signal control on muscles synergies. Different MVJ were assessed with subject specific PCA of lower limb sEMG during Counter Movement Jump (CMJ), Drop Jump (DJ), and Squat Jump (SJ). Intra-subject variability of sEMG PCA allowed the detection of two components explaining maximum variability with different profiles and muscle grouping at CMJ, DJ, and SJ. First component (PC1), representing larger signal variability, presented higher value at SJ and DJ than CMJ, with the need for a higher number of PC's to explain the same cumulative percentual variance at CMJ than DJ and SJ. Comparison with intra-subject linear (r) and cross-correlation (CCr) presented higher r and CCr at SJ and DJ than CMJ, with higher paired correlations at the muscles grouped on the same component. Comparison of intra-subject analysis with previous study on same subject single trial allowed subject-specific generalization of the preceding results.

Abstract
This study presents and applies in vivo lower limb frequency response analysis during standard maximum vertical jump (MVJ) with long and short counter movement (CM) and corresponding muscle stretch shortening cycle (SSC) for comparison without CM and SSC condition. The study makes use of algebraic relation at the frequency domain to obtain the response function from the input and output signals. Single-input/single-output (SI/SO) constant parameter linear system (CPLS) was applied with vertical ground reaction force (GRFz) input and center of gravity (CG) vertical displacement (Delta z) output, obtaining lower limb frequency response function during MVJ impulse phase. Piecewise linearity and limited input-output range of experimentally acquired GRFz and CG Delta z during MVJ impulse phase were assessed to confirm assumptions for CPLS application. Piecewise stationarity of the input and output signal was ensured by acquiring those signals on each MVJ type at similar conditions, guaranteeing experimental repetitions under statistical similar conditions on each CM. Different CM condition on each MVJ type were compared as regards to maximum vertical height, time period of the impulse phase, fundamental harmonic frequencies, convergence of the GRFz input and CG Delta z output Fourier series, their autospectral and cross-spectral density, as well as its input-output coherence, cross-spectrum gain factor, and phase of the frequency response function. Several differences were detected among CM condition, potentially contributing to explain differences on achieved performances at each CM and SSC.

Abstract
This study presents subject specific lower limb joint angular kinematic and dynamic analysis at time and frequency domain as well as joint mechanical work, power and dynamic stiffness assessment during normal gait, stiff knee gait and slow running for indicative range operation of personalized active gait aid device. Gait aid devices present increasing interest for the generalization of gait rehabilitation, as an answer to the growth demand of population with gait rehabilitation need, as well as the insufficient health care personnel. Nevertheless, the large costs and standardized equipment leave out many patients without gait rehabilitation, with the need for low cost, personalized gait rehabilitation equipment, based on subject-specific analysis. In vivo and noninvasive case study was assessed of a healthy male subject 70 kg mass and 1.86 m height on human gait lab. Reflective adhesive marks were applied at skin surface of lower limb selected anatomical points and images captured with eight 100 Hz camera Qualisys along with ground reaction forces and force moments acquired at 2000 Hz with two AMTI force plates during foot contact with the ground on normal gait (NG) at comfortable auto-selected velocity, stiff knee gait (SKG) with lower knee flexion and slow running (SR) at minimum run velocity on stiff knee condition. Inverse kinematics and dynamics were performed using AnyGait with TLEM model and lower limb joint angular signal analyzed. Indicative range operation from lower limb joint mechanical assessment were obtained at complementary domain for subject specific gait aid device selection and parametrization.

Abstract
This article proposes a new method of identity recognition in sanitary facilities based on electrocardiography (ECG) signals. Our team previously proposed a novel approach of invisible ECG at the thighs using polymeric electrodes, leading to the creation of a proof-of-concept system integrated into a toilet seat. In this work, a biometrics pipeline was devised, which tested four different classifiers, varying the population from 2 to 17 subjects and simulating a residential environment. However, for this approach to be industrially viable, further optimization is required, particularly regarding electrode materials that are compatible with industrial processes. As such, we also explore the use of a conductive silicone material as electrodes, aiming at the industrial-scale production of a toilet seat capable of recording ECG data, without the need for body-worn devices. A desirable aspect when using such a system is matching the recorded data with the monitored user, ideally using a minimal sensor set, further reinforcing the relevance of user identification through ECG signals collected at the thighs. Our approach was evaluated against a reference device for a population of 17 healthy and pathological individuals, covering a wide age range (24-70 years). With the silicone composite, we were able to acquire signals in 100% of the sessions, with a mean heart rate deviation between a reference system and our experimental device of 2.82 +/- 1.99 beats per minute (BPM). In terms of ECG waveform morphology, the best cases showed a Pearson correlation coefficient of 0.91 +/- 0.06. For biometric detection, the best classifier was the Binary Convolutional Neural Network (BCNN), with an accuracy of 100% for a population of up to four individuals.

Abstract
Purpose To provide a systematic review with meta-analysis providing evidence of the current diagnostic test accuracy (DTA) of pedicle screw electrical stimulation. Methods A systematic database search on PubMed, Scopus and Web of Science was performed according to the PRISMA-DTA guidelines, and eligibility criteria applied to reduce the results to: (1) only journal articles reporting electrical stimulation of the pedicle screw head, (2) screw position confirmation by imaging techniques, and (3) enough information allowing the calculation of a 2 x 2 contingency table. Sample characteristics, image confirmation method, electrical current threshold and stimulation results were retrieved and analyzed using according to appropriate DTA analysis methods, and allowing the calculation of specificity, sensitivity for pedicle screws insertion at the lumbar and thoracic levels. Results Lumbar screw stimulation presents a higher sensitivity (0.586 [0.336, 0.798] and specificity (0.984 [0.958, 0.994]) than thoracic screws (sensitivity: 0.270 [0.096; 0.562]; specificity: 0.958 [0.931, 0.975]). The same is observed in terms of the diagnostic odds ratio for lumbar (88.32 [32.136, 242.962]) and thoracic (8.460 [2.139, 33.469]) levels. When performing a sub-group analysis, it is possible to divide the lumbar stimulation threshold as 8 and 10-12 mA, and the thoracic threshold as 6 and 9-12 mA. A threshold of 8 mA at the lumbar level provides higher sensitivity and specificity. Increasing the threshold results in higher specificity but not sensitivity. In fact, at the range of 10-12 mA, the diagnostic validity is too low to confer this technique any robust diagnostic validity. Similarly, at the thoracic level, lower threshold currents are associated with increased sensitivity, but their diagnostic validity is very low. Conclusion Electrical stimulation of the pedicle screw can be used as an adequate diagnostic capability at the lumbar level with a threshold of 8 mA. However, thoracic stimulation is currently not reliable, with very low sensitivity and diagnostic validity at 6 mA or higher.