Browsing Health by Department
Now showing items 1-4 of 4
Classifying NIR spectra of textile products with kernel methodsThis paper describes the use of kernel methods to classify tissue samples using near-infrared spectra in order to discriminate between samples, either with or without elastane. The aim of this real-world study is to identify an alternative method to classify textile products using near-infrared (NIR) spectroscopy in order to improve quality control, and to aid in the detection of counterfeit garments. The principles behind support vector machines (SVMs), of which the main idea is to linearly separate data, are recalled progressively in order to demonstrate that the decision function obtained is a global optimal solution of a quadratic programming problem. Generally, this solution is found after embedding data in another space F with a higher dimension by the means of a specific non-linear function, the kernel. For a selected kernel, one of the most important and difficult subjects concerning SVM is the determination of tuning parameters. Generally, different combinations of these parameters are tested in order to obtain a machine with adequate classification ability. With the kernel alignment method used in this paper, the most appropriate kernel parameters are identified rapidly. Since in many cases, data are embedded in F, a linear principal component (PC) analysis (PCA) can be considered and studied. The main properties and the algorithm of k-PCA are described here. This paper compares the results obtained in prediction for a linear classifier built in the initial space with the PCs from a PCA and those obtained in F with non-linear PCs from a k-PCA. In the present study, even if there are potentially discriminating wavelengths seen on the NIR spectra, linear discriminant analysis and soft independent modelling of class analogy results show that these wavelengths are not sufficient to build a machine with correct generalisation ability. The use of a non-linear method, such as SVM and its corollary methods, kernel alignment and k-PCA, is then justified.
Evolution in impedance at the electrode-skin interface of two types of surface EMG electrodes during long-term recordingsThe evolution in impedance at the electrode-skin interface of Beckman and Red Dot electrodes was assessed during long-term recordings. Impedance was measured between each pair of electrodes, arranged in a bipolar configuration on tibialis anterior (n=13). A waveform constructed of sinusoids of known frequencies, evenly distributed on a log scale from 1–16,384 Hz, was applied through the electrodes, and the result recorded by a DAQ system. SEMG signals were recorded at 1000 Hz during isometric dorsiflexion contractions of 30 s, performed every 15 min for 2 h. Impedance data were acquired at 65,536 Hz immediately before and after SEMG recordings. Large individual differences in impedance levels were observed at low frequencies. At high frequencies, impedance values depended only on the electrode type. Impedance decreased steadily with time for Beckman electrodes (p < 0.05), but did not decrease significantly for Red Dot electrodes. The magnitude of the reduction over time varied widely between individuals, and was related to the initial impedance values. The impedance-bandwidth product remained constant for each electrode type (95% confidence intervals 146.2–148.2 and 126.1–127.8 for Beckman and Red Dot electrodes respectively). When skin impedance is electrically modelled with a simple network containing a resistor and a capacitor, the capacitance varies with the properties of the electrode used, whereas resistance is dependent on the subject. Furthermore, the EMG spectrum is unaffected by impedance provided skin preparation is sufficient to reduce the impedance below 55 kΩ.
Intrinsic Mode Entropy for postural steadiness analysisPostural balance during quiet standing is maintained by complex interactions of many sensory systems, including visual, vestibular, and proprioceptive systems. It has been demonstrated that applying vibration to the tibialis anterior tendon when subjects are in a static upright position creates an illusion of body inclination, thus decreasing postural stability. Postural balance was evaluated using centre of pressure (COP) displacements measured using a force plate. Recently, Intrinsic Mode Entropy (IMEn) has been proposed to quantify the degree of regularity and complexity in nonlinear signals. IMEn can be considered as an extension of Sample Entropy (SampEn) to deal with different oscillation levels. The first step of IMEn consists of extracting the Intrinsic Mode Functions (IMFs) of a time-series using Empirical Mode Decomposition (EMD). The IMEn is then obtained by computing the SampEN of the cumulative sums of the IMFs.
Techniques d’évaluation à domicile de la qualité de l’équilibre et de la force de préhension chez la personne âgée en perte d’autonomie ; Devices analysing balance quality and autonomy levelPerforming a movement from an initial stable posture requires the person to create disequilibrium. The forces of gravity that the person is subjected to would tend to make them fall. To counteract these problems, it is necessary to develop mechanisms of balance in order to move about. The degeneration of mechanisms of balance control has been largely measured in elderly subjects. The balance decline, the appearance of fear of falling, and the resulting loss of autonomy, constitute a major problem for public health. The IDéAS research group (UTT) addresses these critical issues by specialising in the development of innovative devices that enable the capacity of elderly to live autonomous to be evaluated in their own homes. The devices used have been designed in order to perform frequent evaluation of the quality of balance and grip strength. These technologies consist of a balance quality tester and the Grip-ball.