PCA-based selection of distinctive stability criteria and classification of post-stroke pathological postural behaviour.

In this paper, we study the postural behaviour of two categories of people: Post-CVA subjects suffering from cerebrovascular accident syndromes and healthy individuals under several levels of anterior-posterior and medial-lateral sinusoidal disturbances (0.1-0.5 Hz). These perturbations were produced from an omnidirectional platform called Isiskate. Afterwards, we have quantified seventy postural parameters, they were combined of linear stabilometric parameters and non-linear time dependent stochastic parameters using stabilogram diffusion analysis and some spectral attributes using power spectral density. The aim of our analysis is to reduce data dimensionality using principal component analysis (PCA). Furthermore, we proposed a new PCA-related criterion named: criterion of contribution in order to evaluate the contribution of every variable in the resulted system structure, and thus to eliminate the redundant postural characteristics. Afterwards, we highlighted some interesting distinctive parameters. The selected parameters were used thereafter in comparison between the studied groups. Finally, we created a classification model using support vector machines to distinguish stroke patients. Our proposed techniques help in understanding the human postural dynamics and facilitate the diagnosis of pathologies related to equilibrium which can be used to improve the rehabilitation services.

Joint angle estimation with accelerometers for dynamic postural analysis.

This paper presents a new accelerometer based method for estimating the posture of a subject standing on a dynamic perturbation platform. The induced perturbation is used to study the control mechanisms as well as the balance requirements that regulate the upright standing. These perturbations are translated into different intensity levels of speed and acceleration along longitudinal and lateral directions of motion. In our method, the human posture is modeled by a tridimensional, three-segment inverted pendulum which simultaneously takes into account both the anterior-posterior and medio-lateral strategies of hip and ankle. Four tri-axial accelerometers are used her, one accelerometer is placed on the platform, and the other three are attached to a human subject. Based on the results, the joint angle estimated compare closely to measurements from magnetic encoders placed on an articulated arm joint. The results were also comparable to those found when using a high-end optical motion capture system coupled with advanced biomechanical simulation software. This paper presents the comparisons of our accelerometer-based method with encoder and optical marker based method of the estimated joint angles under different dynamics perturbations.

Design of multiple axis robotic platform for postural stability analysis.

This paper presents the design and implementation of IsiMove, a new dynamic posturography platform. It allows the evaluation of the static and dynamic balance of a human placed on a force plate. IsiMove is a robotic platform open kinematic with four degrees of freedom: anteroposterior tilt, mediolateral tilt, vertical rotation, and horizontal translation. It is capable of measuring the displacement of the center of pressure over time, with a resolution of 0.1 mm for each foot and support a human of about 120 kg. IsiMove can generate various types of balance perturbations based on parameters such as direction, amplitude, frequency and shape. In this paper, we will give a description of the mechanisms that constitute our platform. First, the technical specifications of the hardware and software architecture will be presented. Then, we will provide details related to extensive experimental evaluations of the platform in both static and dynamic condition as well as result of postural stability analysis with healthy subjects and stroke patients.

Bench test evaluation of adaptive servoventilation devices for sleep apnea treatment.

RATIONALE: Adaptive servoventilation devices are marketed to overcome sleep disordered breathing with apneas and hypopneas of both central and obstructive mechanisms often experienced by patients with chronic heart failure. The clinical efficacy of these devices is still questioned. STUDY OBJECTIVES: This study challenged the detection and treatment capabilities of the three commercially available adaptive servoventilation devices in response to sleep disordered breathing events reproduced on an innovative bench test. METHODS: The bench test consisted of a computer-controlled piston and a Starling resistor. The three devices were subjected to a flow sequence composed of central and obstructive apneas and hypopneas including Cheyne-Stokes respiration derived from a patient. The responses of the devices were separately evaluated with the maximum and the clinical settings (titrated expiratory positive airway pressure), and the detected events were compared to the bench-scored values. RESULTS: The three devices responded similarly to central events, by increasing pressure support to raise airflow. All central apneas were eliminated, whereas hypopneas remained. The three devices responded differently to the obstructive events with the maximum settings. These obstructive events could be normalized with clinical settings. The residual events of all the devices were scored lower than bench test values with the maximum settings, but were in agreement with the clinical settings. However, their mechanisms were misclassified. CONCLUSION: The tested devices reacted as expected to the disordered breathing events, but not sufficiently to normalize the breathing flow. The device-scored results should be used with caution to judge efficacy, as their validity depends upon the initial settings.