Silvia Conforto

Classification of Motor Activities through Derivative Dynamic Time Warping applied on Accelerometer Data

In the context of tele-monitoring, great interest is presently devoted to physical activity, mainly of elderly or people with disabilities. In this context, many researchers studied the recognition of activities of daily living by using accelerometers. The present work proposes a novel algorithm for activity recognition that considers the variability in movement speed, by using dynamic programming. This objective is realized by means of a matching and recognition technique that determines the distance between the signal input and a set of previously defined templates. Two different approaches are here presented, one based on Dynamic Time Warping (DTW) and the other based on the Derivative Dynamic Time Warping (DDTW). The algorithm was applied to the recognition of gait, climbing and descending stairs, using a biaxial accelerometer placed on the shin. The results on DDTW, obtained by using only one sensor channel on the shin showed an average recognition score of 95%, higher than the values obtained with DTW (around 85%). Both DTW and DDTW consistently show higher classification rate than classical Linear Time Warping (LTW).

The sensitivity of posturographic parameters to acquisition settings

The objective of this study was to evaluate the sensitivity of posturographic parameters (PP) to changes in acquisition settings. A group of eight young adults underwent a set of typical orthostatic posture trials, and selected PP were then calculated from a set of centre of pressure (CoP) displacement time series obtained by applying different cut-off frequencies to the same set of raw data. Four PP out of 11 showed significant changes with respect to cut-off frequency. Statistical mechanics parameters exhibited smaller sensitivity than summary measures. On the basis of the results obtained, a proposal for a standard cut-off frequency and a sampling rate value is embodied in the paper together with some suggestions on measurement settings, with a view to standardized use of instrumentation for quantitative analysis in orthostatic posturography.

Optimal rejection of movement artefacts from myoelectric signals by means of a wavelet filtering procedure

In this work the problem of rejection of motion artefacts from surface myoelectric signals, recorded during dynamic contractions, is studied. In fact, the extraction of frequency parameters and the detection of muscular activation patterns can be detrimentally affected by artefacts due to the movement of the surface electrodes, particularly stressed by the dynamic conditions of the exercise performed during measurement. In order to overcome this difficulty, four different filtering procedures have been tested and compared: a high-pass filtering procedure, a moving average procedure, a moving median procedure and a new adaptive wavelet based procedure, expressly designed for this work. Orthogonal Meyer wavelets are used with the aim of obtaining both a good reconstruction and a decomposition of the signal into non-overlapping bands. The four procedures have been tested with a set of different proofs utilising both synthetic and experimentally recorded myoelectric signals. The results show that the wavelet procedure performs better than the other methods both in information preservation and in time-detection. Moreover, the features of user-independence and adaptivity to the noise level suggest a wider range of applications of the proposed algorithm.

Hemodynamics as a possible internal mechanical disturbance to balance

The postural control system is assessed by observing body sway while the subject involved aims at maintaining a specified up-right posture. Internal masses generate internal reaction forces that constitute an internal mechanical stimulus that may contribute to cause segmental displacements, i.e. body sway. Thus, gaining knowledge about the amplitude and direction of these reaction forces would contribute to gain insights into the mechanisms that influence the maintenance of balance and into its control. The 3-D force vector that acts on the body centre of mass (COM) and is associated with the transient blood movement at each cardiac cycle was assessed in a population sample of 20 young adults during the maintenance of a quiet up-right posture. Typical patterns of the three components of this force vector were identified. Relevant parameters were selected and submitted to sample statistics. For a number of them, linear correlation with subject-specific parameters was found. The antero-posterior force component was characterised by a triphasic major wave, the peaks of which had values up to 0.40 N. The vertical component showed a repeatable triphasic wave with peak-to-peak values in the range 1.3-3.0 N. The medio-lateral component showed relatively low peak-to-peak values (in the range 0.05-0.10 N). The resultant vector had an amplitude that underwent several oscillations during the cardiac cycle and reached its maximal value in the range 0.6-1.7 N.

A neural-based remote eye gaze tracker under natural head motion

A novel approach to view-based eye gaze tracking for human computer interface (HCI) is presented. The proposed method combines different techniques to address the problems of head motion, illumination and usability in the framework of low cost applications. Feature detection and tracking algorithms have been designed to obtain an automatic setup and strengthen the robustness to light conditions. An extensive analysis of neural solutions has been performed to deal with the non-linearity associated with gaze mapping under free-head conditions. No specific hardware, such as infrared illumination or high-resolution cameras, is needed, rather a simple commercial webcam working in visible light spectrum suffices. The system is able to classify the gaze direction of the user over a 15-zone graphical interface, with a success rate of 95% and a global accuracy of around 2 degrees , comparable with the vast majority of existing remote gaze trackers.

Automatic detection of surface EMG activation timing using a wavelet transform based method

The problem of the identification of the muscle contraction timing by using surface electromyographic signal is addressed. The timing detection of the muscular activation in dynamic conditions has a real clinical diagnostic impact. Widely used single threshold methods still rely on the experience of the operator in manually setting that threshold. A new approach to detect the muscular activation intervals, that is based on discontinuities detection in the wavelet domain, is proposed. Accuracy and precision of the algorithm were assessed by using a set of simulated signals obtaining values lower than 11.0 and 8.7 ms for biases and standard deviations of the estimation, respectively. Moreover an experimental application of the algorithm was carried out recruiting a population of 10 able-bodied subjects and processing the myoelectric signals recorded from the lower limb during an isokinetic exercise. The algorithm was able to reveal correctly the timing of muscular activation with performance comparable to the state-of-the-art methods. The detection algorithm is automatic and user-independent, it manages the detection of both onset and offset activation, it can be fruitfully applied even in presence of noise and, therefore, it can be used also by unskilled operators.

Early recognition of upper limb motor tasks through accelerometers: real-time implementation of a DTW-based algorithm

A new real-time implementation of a Dynamic Time Warping (DTW)-based classification scheme is presented here, and its performance evaluated on experimental data. Nine young adults were requested to perform instances of eight different purposeful movements described in the Wolf Motor Function Test, while wearing a three-axis accelerometer sensor placed on the inner forearm. Results include the correct recognition percentage, as compared to a classification scheme based on the traditional DTW measure, and the recognition percentage as a function of the time elapsed from the beginning of the performed movements. The Real-Time DTW basically performs with the same accuracy of the traditional DTW-based classification scheme (91.5% of correct recognition percentage), a figure that increases to 96.5% if the multidimensional scheme is adopted. Moreover, more than 60% of movements are correctly recognized before their end, thus setting the way for applications in rehabilitation and assistive technologies, where a real-time control scheme is able to interact with the user while the movement is being performed.

Markerless Human Motion Analysis in Gauss–Laguerre Transform Domain: An Application to Sit-To-Stand in Young and Elderly People

A markerless computer vision technique specifically designed to track natural elements on the human body surface is presented. The method implements the estimate of translation, rotation, and scaling by means of a maximum likelihood approach carried out in the Gauss-Laguerre transform domain. The approach is particularly suitable for human movement analysis in clinical contexts, where kinematics is at present performed by means of marker-based systems. Specific drawbacks of these latter systems, such as the burden of time for marker placement and the intrinsic intrusive nature, would be removed by the proposed method. Experimental results in terms of tracking performance are obtained by analyzing video sequences capturing the execution of the sit-to-stand task in two groups of young and elderly volunteers. The results are compared with clinical studies that used marker-based systems, and are particularly encouraging for a future extension of the approach to other motor tasks and to predict scores obtained from the physical performance batteries that are widely and regularly used by clinicians and physical therapists.

The development of postural strategies in children: a factorial design study.

BackgroundThe present study investigates balance control mechanisms, their variations with the absence of visual input, and their development in children from 7 to 11 years old, in order to provide insights on the development of balance control in the pediatric population.MethodsPosturographic data were recorded during 60 s trials administered on a sample population of 148 primary school children while stepping and then quietly standing on a force plate in two different vision conditions: eyes closed and eyes open. The extraction of posturographic parameters on the quiet standing phase of the experiment was preceded by the implementation of an algorithm to identify the settling time after stepping on the force plate. The effect of different conditions on posturographic parameters was tested with a two-way ANOVA (Age × Vision), and the corresponding eyes-closed/eyes-open (Romberg) Ratios underwent a one-way ANOVA.ResultsSeveral posturographic measures were found to be sensitive to testing condition (eyes closed vs. eyes open) and some of them to age and anthropometric parameters. The latter relationship did not explain all the data variability with age. An evident modification of postural strategy was observed between 7 and 11 years old children.ConclusionSimple measures extracted from posturographic signals resulted sensitive to vision and age: data acquired from force plate made it possible to confirm the hypothesis of the development of postural strategies in children as a more mature selection and re-weighting of proprioceptive inputs to postural control in absence of visual input.

A neural tracking and motor control approach to improve rehabilitation of upper limb movements

BackgroundRestoration of upper limb movements in subjects recovering from stroke is an essential keystone in rehabilitative practices. Rehabilitation of arm movements, in fact, is usually a far more difficult one as compared to that of lower extremities. For these reasons, researchers are developing new methods and technologies so that the rehabilitative process could be more accurate, rapid and easily accepted by the patient. This paper introduces the proof of concept for a new non-invasive FES-assisted rehabilitation system for the upper limb, called smartFES (sFES), where the electrical stimulation is controlled by a biologically inspired neural inverse dynamics model, fed by the kinematic information associated with the execution of a planar goal-oriented movement. More specifically, this work details two steps of the proposed system: an ad hoc markerless motion analysis algorithm for the estimation of kinematics, and a neural controller that drives a synthetic arm. The vision of the entire system is to acquire kinematics from the analysis of video sequences during planar arm movements and to use it together with a neural inverse dynamics model able to provide the patient with the electrical stimulation patterns needed to perform the movement with the assisted limb.MethodsThe markerless motion tracking system aims at localizing and monitoring the arm movement by tracking its silhouette. It uses a specifically designed motion estimation method, that we named Neural Snakes, which predicts the arm contour deformation as a first step for a silhouette extraction algorithm. The starting and ending points of the arm movement feed an Artificial Neural Controller, enclosing the muscular Hills model, which solves the inverse dynamics to obtain the FES patterns needed to move a simulated arm from the starting point to the desired point. Both position error with respect to the requested arm trajectory and comparison between curvature factors have been calculated in order to determine the accuracy of the system.ResultsThe proposed method has been tested on real data acquired during the execution of planar goal-oriented arm movements. Main results concern the capability of the system to accurately recreate the movement task by providing a synthetic arm model with the stimulation patterns estimated by the inverse dynamics model. In the simulation of movements with a length of ± 20 cm, the model has shown an unbiased angular error, and a mean (absolute) position error of about 1.5 cm, thus confirming the ability of the system to reliably drive the model to the desired targets. Moreover, the curvature factors of the factual human movements and of the reconstructed ones are similar, thus encouraging future developments of the system in terms of reproducibility of the desired movements.ConclusionA novel FES-assisted rehabilitation system for the upper limb is presented and two parts of it have been designed and tested. The system includes a markerless motion estimation algorithm, and a biologically inspired neural controller that drives a biomechanical arm model and provides the stimulation patterns that, in a future development, could be used to drive a smart Functional Electrical Stimulation system (sFES). The system is envisioned to help in the rehabilitation of post stroke hemiparetic patients, by assisting the movement of the paretic upper limb, once trained with a set of movements performed by the therapist or in virtual reality. Future work will include the application and testing of the stimulation patterns in real conditions.