A markerless system based on smartphones and webcam for the measure of step length, width and duration on treadmill.

BACKGROUND AND OBJECTIVE: A markerless low cost prototype has been developed for the determination of some spatio-temporal parameters of human gait: step-length, step-width and cadence have been considered. Only a smartphone and a high-definition webcam have been used. METHODS: The signals obtained by the accelerometer embedded in the smartphone are used to recognize the heel strike events, while the feet positions are calculated through image processing of the webcam stream. Step length and width are computed during gait trials on a treadmill at various speeds (3, 4 and 5 km/h). RESULTS: Six subjects have been tested for a total of 504 steps. Results were compared with those obtained by a stereo-photogrammetric system (Elite, BTS Engineering). The maximum average errors were 3.7 cm (5.36%) for the right step length and 1.63 cm (15.16%) for the right step width at 5 km/h. The maximum average error for step duration was 0.02 s (1.69%) at 5 km/h for the right steps. CONCLUSION: The system is characterized by a very high level of automation that allows its use by non-expert users in non-structured environments. A low cost system able to automatically provide a reliable and repeatable evaluation of some gait events and parameters during treadmill walking, is relevant also from a clinical point of view because it allows the analysis of hundreds of steps and consequently an analysis of their variability.

Accuracy evaluation of the Kinect v2 sensor during dynamic movements in a rehabilitation scenario.

In this paper, the accuracy evaluation of the Kinect v2 sensor is investigated in a rehabilitation scenario. The accuracy analysis is provided in terms of joint positions and angles during dynamic postures used in low-back pain rehabilitation. Although other studies have focused on the validation of the accuracy in terms of joint angles and positions, they present results only considering static postures whereas the rehabilitation exercise monitoring involves to consider dynamic movements with a wide range of motion and issues related to the joints tracking. In this work, joint positions and angles represent clinical features, chosen by medical staff, used to evaluate the subject's movements. The spatial and temporal accuracy is investigated with respect to the gold standard, represented by a stereophotogrammetric system, characterized by 6 infrared cameras. The results provide salient information for evaluating the reliability of Kinect v2 sensor for dynamic postures.

A tool for home-based rehabilitation allowing for clinical evaluation in a visual markerless scenario.

This work deals with the design of an interactive monitoring tool for home-based physical rehabilitation. The software platform includes a video processing stage and the exercise performance evaluation. Image features are extracted by a Kinect v2 sensor and elaborated to return the exercises score. Furthermore the tool provides to physiotherapists a quantitative exercise evaluation of subject's performances. The proposed tool for home rehabilitation has been tested on 5 subjects and 5 different exercises and results are presented. In particular both exercises and relative evaluation indexes were selected by specialists in neurorehabilitation.

Analysis of ground reaction forces by means of wavelet transform.

OBJECTIVE: To identify details of clinical relevance in ground reaction forces by means of wavelet transform. DESIGN: A retrospective analysis of gait tests performed by total knee replacement patients and a control population has been performed. METHODS: The ground reaction forces have been processed by means of wavelet transform. Results of the wavelet analysis are represented, in a time-frequency plane, by tiles. These are coloured in different grey levels associated to the values of a suitable energy function of the expansion coefficients, resulting from the wavelet transform. RESULTS: The high frequency tiles revealed the presence of irregularities with clinical significance in the first part of the stance phase. These transients in ground reaction forces are described in a quantitative manner allowing to monitor their evolution during the patient observation time. The heel strike transient has been identified by the tile representation of the vertical component of ground reaction forces and confirmed by its correlation with corresponding irregularities in the other components. CONCLUSIONS: The tile representation allows to detect and to quantify details not easily perceivable by the examiner through traditional techniques. The wavelet transform seems particularly appealing for clinical applications such as outcome assessment or treatment evaluation and can assist in the definition of normative models of ground reaction forces. The effectiveness of the procedure suggest to try to automate it. RELEVANCE: The tile representation allows to identify and to keep the records of ground reaction forces clinically significant details, such as heel strike transient.