Project

Cardiac Pathology and Signal Processing

Updates
0 new
0
Recommendations
0 new
0
Followers
0 new
2
Reads
0 new
43

Project log

L.N. Sharma
added a research item
Pulse transit time (PTT) has been widely used for cuffless blood pressure (BP) measurement. But, it requires more than one cardiovascular signals involving more than one sensing device. In this paper, we propose a method for cuffless continuous blood pressure measurement with the help of left ventricular ejection time (LVET). A MEMS-based accelerometric sensor acquires a seismocardiogram (SCG) signal at the chest surface, and then, the LVET information is extracted. Both systolic and diastolic blood pressures are estimated by calibrating the system with the original arterial blood pressures of the subjects. The performance evaluation is done using different statistical quantitative measures for the proposed method. The performance is also compared with two earlier approaches, where PTT intervals are measured from electrocardiogram (ECG)-photoplethysmogram (PPG) and SCG-PPG pairs, respectively. The performance results clearly show that the proposed method is comparable with the state-of-the-art methods. Also, the estimated blood pressure is compared with the original one, measured through a reference system. It gives the mean errors of the systolic and diastolic BPs within the range of -0.197±3.332 mmHg and -1.299±2.578 mmHg, respectively. The BPs estimation errors satisfy the requirements of the IEEE standard 5±8 mmHg deviation, and thus, our method may be used for ubiquitous continuous blood pressure monitoring.
L.N. Sharma
added 2 research items
In this work, multiscale covariance analysis is proposed for multilead electrocardiogram signals to detect myocardial infarction (MI). Due to multiresolution decomposition, diagnostically important clinical components are grossly segmented at different scales. If multiscale multivariate matrices are formed using all ECG leads and subjected to covariance analysis at wavelet scales, covariances change from normal as MI evolves. This is due to the underlying pathology which is seen in few ECG leads. To capture the changes that occur during infarction, multiscale multivariate distortion metric is applied on covariance structures. To evaluate the proposed method, data sets are adopted from PTB diagnostic ECG database. This includes healthy control (HC), myocardial infarction in early stage (MIES), and acute myocardial infarction (AMI). The results show that the proposed method can detect the pathological MI subjects. For MI detection, the accuracy, the sensitivity, and the specificity is found to be 80, 76, and 84 %, respectively. The proposed method is simple and can be easily implemented for offline analysis for diagnosis of infarction using multiple leads.
In this paper a heart sound segmentation algorithm in multiresolution domain is proposed. Wavelet decomposition of heart sound signal grossly segments its components into different subbands. If multiscale Hilbert envelope is computed on reconstructed signals at different scales, it provides suitable markers for first and second heart sound boundaries. The selection of wavelet subband for marker generation is based on analysis of heart sound spectra and energy contribution of wavelet subbands. The heart sound boundaries for S1 and S2 are decided by markers derived from second derivative of Hilbert envelope of the reconstructed subband signal. The proposed method is evaluated using heart sound signal available in the web site of the Department of Medicine, Washington University. The performance of the proposed method is found satisfactory.