Analysis of tremor in multiple sclerosis using Hilbert-Huang Transform.

Tremor is frequently described in patients with multiple sclerosis (MS) but remains poorly characterized using neurophysiological techniques. Accelerometric (ACC) and electromyographic (EMG) recordings were performed in 26 MS patients complaining of clumsiness, associated (n = 16) or not associated (n = 10) with visible tremor. Seventeen healthy subjects with physiological tremor (PT) and eight patients with essential tremor (ET) served as controls. Signals were analyzed using non-linear Empirical Mode Decomposition (EMD) and related Hilbert-Huang Transform (HHT), compared to the standard linear spectral analysis using Fast Fourier Transform (FFT). The presence of cerebellar signs and motor deficit was assessed on clinical examination. Using FFT, tremor was found in all patients with ET and 12% of subjects with PT, but in none of the MS patients, even in the presence of visible tremor. In contrast, EMD-HHT analysis of ACC-EMG coupling showed common frequency peaks characterizing tremor related to a central generator in 62.5% of MS patients with visible tremor, 40% of MS patients without visible tremor, 29% of subjects with PT, and all patients with ET. In EMD-HHT analysis, tremor characteristics were similar in subjects with PT and MS patients, regardless of the presence of a visible tremor, but these characteristics clearly differed in patients with ET. A visible tremor in MS patients was associated with more frequent cerebellar signs and less motor deficit at the upper limb. The low-frequency tremor observed in MS patients could therefore originate in lesions of the brainstem (midbrain) or cerebellothalamic circuits, or may correspond to an enhanced PT, partly favored by cerebellar dysfunction and being more visible during movement execution in the absence of concomitant motor deficit.

Distinction between essential and physiological tremor using Hilbert-Huang transform.

Neurophysiological investigation, based on accelerometric (ACC) and electromyographic (EMG) recordings, is an essential step in the diagnosis of tremor. Among various signal analysis methods, fast Fourier transform (FFT) is the most frequently used in this domain. However, FFT has several limitations: first, it assumes that tremor is a periodic and linear signal, which is not true; second, it cannot distinguish between different types of tremor, when their frequency overlap in similar range, such as essential tremor (ET) and physiological tremor (PT). Therefore, we decided to apply a non-linear method of signal analysis based on empirical mode decomposition (EMD) and Hilbert Huang transform (HHT), according to various procedures and compared to a more classical FFT approach. A first group of 8 healthy subjects with PT and a second group of 8 patients with ET were included in this study. At individual level, FFT was effective to highlight ET in the 8 patients, but PT in only 2 subjects. The EMD-HHT procedures performed better than FFT, revealing a common peak of PT in all subjects. Moreover, at group level, our EMD-HHT method allowed to clearly differentiate the two groups, especially by giving evidence for the existence of low frequency oscillations (around 4Hz) in subjects with PT. Although their physiological origin remains largely unknown, such slow oscillations seem to be of great importance to highlight PT and they have been much underestimated in the literature. Our original EMD-HHT approach is able to provide substantial improvement in the neurophysiological characterisation of the different types of tremor, especially for diagnostic application.

Long short-term memory for apnea detection based on Heart Rate Variability.

The main drive force in apnea current diagnostic is to reduce overwhelming number of sleep disorders candidates by means of very simple-to-use, comfortable and cheap methodology. The proposed framework is based only on automatic analysis of electrocardiogram signal. The feature extraction stage was performed using methods of Heart Rate Variability and Detrended Fluctuation analysis. Feature-spaces formed using these two methods were used as input to a Long Short-Term Memory Artificial Neural Network chosen for its capability to find temporally dependencies in the data. The framework was evaluated on Challenge 2000 Physionet database yielding successful rate 82.1%, sensitivity 85.5% and specificity 80.1%.

Speech recognition methods applied to biomedical signals processing.

The paper focuses on processing of long biological signals used during monitoring procedures like in the case of portable Holter device for arrythmia analysis (ECG), intracranial pressure monitoring (ICP) in intensive care unit or overnight electroencephalogram monitoring (EEG) for sleep apnea detection. Two methods taken from speech processing are proposed: dynamic time warping (DTW) and hidden Markov models (HMM). The unsupervised analysis of ECG and ICP beats is carried out using hierarchical clustering approach. In case of EEG, first the estimation of sleep stages is performed and next the different breathing events are detected by HMM by means of Viterbi inference. We show that for the first two problems DTW outperforms HMM while in the third case the HMM inference capability makes HMM suitable for sleep apnea diagnosis.

Distribution of calcium absorbed by all or part of the root system of beans.

When either the whole root system or a single root of the bean plant (Phaseolus vulgaris L.), at different stages of maturity, was fed (45)Ca for a period of 6 hr, the leaves exhibited a definite pattern of (45)Ca acquisition with maximum acquisition for the primary, first, second, and third trifoliolate leaves at 12, 16, 18, and 22 days of growth, respectively. This maximum acquisition occurred when a leaf approached full expansion. A liquid scintillation spectrometer was employed for (45)Ca analysis.Autoradiograms indicated that when (45)Ca was applied to the terminal 5 cm portion of specific lateral roots for a period of 6 hr, specific areas of the shoot were preferentially supplied with (45)Ca. The amount of (45)Ca transported upward, when the terminal 5 cm portion of a single lateral root was treated, greatly decreased in plants older than 20 days, and the growth rate of these roots was also reduced at about the same time. When the whole lateral root was involved in (45)Ca uptake, the amount of (45)Ca transported to the shoot increased with increasing plant age until after day 22. A continuous increase in (45)Ca transport to the shoot occurred when the whole root system was subjected to (45)Ca treatment. One factor involved in the decreased movement of (45)Ca into a particular leaf may be a decreased absorptive capacity of that part of the root which preferentially supplies Ca to this leaf. Other possible factors are discussed.