Epileptic seizure recognition using EEG wavelet decomposition based on nonlinear and statistical features with support vector machine classification.

Epileptic seizure (ES) is a neurological brain dysfunction. ES can be detected using the electroencephalogram (EEG) signal. However, visual inspection of ES using long-time EEG recordings is a difficult, time-consuming and a costly procedure. Thus, automatic epilepsy recognition is of primary importance. In this paper, a new method is proposed for automatic ES recognition using short-time EEG recordings. The method is based on first decomposing the EEG signals on sub-signals using discrete wavelet transform. Then, from the obtained sub-signals, different non-linear parameters such as approximate entropy (ApEn), largest Lyapunov exponents (LLE) and statistical parameters are determined. These parameters along with phase entropies, calculated through higher order spectrum analysis, are used as an input vector of a multi-class support vector machine (MSVM) for ES recognition. The proposed method is evaluated using the standard EEG database developed by the Department of Epileptology, University of Bonn, Germany. The evaluation is carried out through a large number of classification experiments. Different statistical metrics namely Sensitivity (Se), Specificity (Sp) and classification accuracy (Ac) are calculated and compared to those obtained in the scientific research literature. The obtained results show that the proposed method achieves high accuracies, which are as good as the best existing state-of-the-art methods studied using the same EEG database.

A new approach for blood pressure estimation based on phonocardiogram.

Continuous and non-invasive measurement of blood pressure (BP) is of great importance particularly for patients in critical state. To achieve continuous and cuffless BP monitoring, pulse transit time (PTT) has been reported as a potential parameter. Nevertheless, this approach remains very sensitive, cumbersome and disagreeable in ambulatory measurement. This paper proposes a new approach to estimate blood pressure through PCG signal by exploring the correlation between PTT and diastolic duration (S21). In this purpose, an artificial neural network was developed using as input data: (systolic duration, diastolic duration, heart rate, sex, height and weight). According to the NN decision, the mean blood pressure was measured and consequently the systolic and the diastolic pressures were estimated. The proposed method is evaluated on 37 subjects. The obtained results are satisfactory, where, the error in the estimation of the systolic and the diastolic pressures compared to the commercial blood pressure device was in the order of 6.48 ± 4.48  mmHg and 3.91 ± 2.58  mmHg, respectively, which are very close to the AAMI standard, 5 ± 8  mmHg. This shows the feasibility of estimating of blood pressure using PCG.

An ultrasound look at Korotkoff sounds: the role of pulse wave velocity and flow turbulence.

AIMS: The aim of this study was to analyze the temporal relationships between pressure, flow, and Korotkoff sounds, providing clues for their comprehensive interpretation. MATERIALS AND METHODS: When measuring blood pressure in a group of 23 volunteers, we used duplex Doppler ultrasonography to assess, under the arm-cuff, the brachial artery flow, diameter changes, and local pulse wave velocity (PWV), while recording Korotkoff sounds 10 cm downstream together with cuff pressure and ECG. RESULTS: The systolic (SBP) and diastolic (DBP) blood pressures were 118.8±17.7 and 65.4±10.4 mmHg, respectively (n=23). The brachial artery lumen started opening when cuff pressure decreased below the SBP and opened for an increasing length of time until cuff pressure reached the DBP, and then remained open but pulsatile. A high-energy low-frequency Doppler signal, starting a few milliseconds before flow, appeared and disappeared together with Korotkoff sounds at the SBP and DBP, respectively. Its median duration was 42.7 versus 41.1 ms for Korotkoff sounds (P=0.54; n=17). There was a 2.20±1.54 ms/mmHg decrement in the time delay between the ECG R-wave and the Korotkoff sounds during cuff deflation (n=18). The PWV was 10±4.48 m/s at null cuff pressure and showed a 0.62% decrement per mmHg when cuff pressure increased (n=13). CONCLUSION: Korotkoff sounds are associated with a high-energy low-frequency Doppler signal of identical duration, typically resulting from wall vibrations, followed by flow turbulence. Local arterial PWV decreases when cuff pressure increases. Exploiting these changes may help improve SBP assessment, which remains a challenge for oscillometric techniques.

A novel biometric authentication approach using ECG and EMG signals.

Security biometrics is a secure alternative to traditional methods of identity verification of individuals, such as authentication systems based on user name and password. Recently, it has been found that the electrocardiogram (ECG) signal formed by five successive waves (P, Q, R, S and T) is unique to each individual. In fact, better than any other biometrics' measures, it delivers proof of subject's being alive as extra information which other biometrics cannot deliver. The main purpose of this work is to present a low-cost method for online acquisition and processing of ECG signals for person authentication and to study the possibility of providing additional information and retrieve personal data from an electrocardiogram signal to yield a reliable decision. This study explores the effectiveness of a novel biometric system resulting from the fusion of information and knowledge provided by ECG and EMG (Electromyogram) physiological recordings. It is shown that biometrics based on these ECG/EMG signals offers a novel way to robustly authenticate subjects. Five ECG databases (MIT-BIH, ST-T, NSR, PTB and ECG-ID) and several ECG signals collected in-house from volunteers were exploited. A palm-based ECG biometric system was developed where the signals are collected from the palm of the subject through a minimally intrusive one-lead ECG set-up. A total of 3750 ECG beats were used in this work. Feature extraction was performed on ECG signals using Fourier descriptors (spectral coefficients). Optimum-Path Forest classifier was used to calculate the degree of similarity between individuals. The obtained results from the proposed approach look promising for individuals' authentication.

A new approach based on the median filter to T-wave detection in ECG signal.

The electrocardiogram (ECG) is one of the most used signals in the diagnosis of heart disease. It contains different waves which directly correlate to heart activity. Different methods have been used in order to detect these waves and consequently lead to heart activity diagnosis. This paper is interested more particularly to the detection of the T-wave. Such a wave represents the re-polarization state of the heart activity. The proposed approach is based on the algorithm procedure which allows the detection of the T-wave using a lot of filter including mean and median filter. The proposed algorithm is implemented and tested on a set of ECG recordings taken from, respectively, the European STT, MITBIH and MITBIH ST databases. The results are found to be very satisfactory in terms of sensitivity, predictivity and error compared to other works in the field.

Analysis of ECG-trunk muscle signal amplitude and heart rate relationship.

The objective of this paper is to investigate if bioelectrical signals, generated from trunk muscles identified in an electrocardiogram (ECG) signal presented in this paper as ECG-Trunk Muscles Signals amplitude (Ecg-TMSA) are correlated with Heart rate (HR) during different levels of physical activity and also if Ecg-TMSA is not influenced by mental activity. HR and Ecg-TMSA were derived from ECG in 14 subjects when walking and jogging at different treadmill velocities from 4-10 (km h(-1)). The mean relationship for all 14 subjects was HR = (42.3 ± 0.2) + (45.3 ± 2.8) Ecg-TMSA, r(2 )= 0.91. The result of one individual data points example for a 21 min experiment was (r(2 )= 0.93, p < 0.0001, n = 336). The obtained results show a linear relationship between Ecg-TMSA and HR. Moreover, the Ecg-TMSA was not affected by mental activity.

Detection of the valvular split within the second heart sound using the reassigned smoothed pseudo Wigner-Ville distribution.

BACKGROUND: In this paper, we developed a novel algorithm to detect the valvular split between the aortic and pulmonary components in the second heart sound which is a valuable medical information. METHODS: The algorithm is based on the Reassigned smoothed pseudo Wigner-Ville distribution which is a modified time-frequency distribution of the Wigner-Ville distribution. A preprocessing amplitude recovery procedure is carried out on the analysed heart sound to improve the readability of the time-frequency representation. The simulated S2 heart sounds were generated by an overlapping frequency modulated chirp-based model at different valvular split durations. RESULTS: Simulated and real heart sounds are processed to highlight the performance of the proposed approach. The algorithm is also validated on real heart sounds of the LGB-IRCM (Laboratoire de Génie biomédical-Institut de recherches cliniques de Montréal) cardiac valve database. The A2-P2 valvular split is accurately detected by processing the obtained RSPWVD representations for both simulated and real data.

Automatic classification of heartbeats using wavelet neural network.

The electrocardiogram (ECG) signal is widely employed as one of the most important tools in clinical practice in order to assess the cardiac status of patients. The classification of the ECG into different pathologic disease categories is a complex pattern recognition task. In this paper, we propose a method for ECG heartbeat pattern recognition using wavelet neural network (WNN). To achieve this objective, an algorithm for QRS detection is first implemented, then a WNN Classifier is developed. The experimental results obtained by testing the proposed approach on ECG data from the MIT-BIH arrhythmia database demonstrate the efficiency of such an approach when compared with other methods existing in the literature.