12 December 2019 in IEEE Access
By classifying the heart sound signals, it can provide very favorable clinical information to the diagnosis of cardiovascular diseases. According to the characteristics of heart sound signals which are complex and difficult to classify and recognize, a new method of feature extraction and classification about heart sound signal is proposed by a combination of wavelet scattering transform and twin support vector machine in this paper. The method is as follows: The heart sound signal data set is firstly divided into two parts, one as a training set and the other as a testing set. Then the wavelet scattering transform is applied to the heart sound signals in the training set and the testing set. The scattering transform is a new timefrequency analysis method. It overcomes the shortcomings of the traditional wavelet transform which has the time-shift changes. It has the advantages of translation invariance and elastic deformation stability. Thus obtain the scattering feature matrix of the heart sound signal. Due to the large dimension of scattering feature matrix, this paper uses multidimensional scaling (MDS) method to reduce the dimension. This method is compared with the classical dimension reduction method—principal component analysis (PCA). Finally, the dimensionality-reduced feature matrix is input into the twin support vector machine (TWSVM) for training. After training the classifier to get the optimal parameters, the dimensionality-reduced scattering feature matrix of the testing signal is input into the classifier for testing. Experimental results show that the classification accuracy of the proposed method can reach 98% or more, and the running time is greatly reduced compared with support vector machine (SVM).
21 June 2023 in Biomedical Signal Processing and Control
2 May 2022 in IEEE Transactions on Instrumentation and Measurement
Background: Cerebral blood flow (CBF) auto-regulation function is essential for evaluating CBF status. This article provides an evaluation method of CBF using bioimpedance. Method: A mathematical model of cerebral vascular compliance is established to analyze the effect of autoregulation ability changes on the impedance of CBF. The synchronization signals of impedance and blood pressure are measured, and the peak phase shift is extracted to evaluate the cerebral blood autoregulation function. Results: The experiment results show that the peak phase shift can express the cerebral autoregulation change process. The frontal and occipital region’s peak phase shift changes are 10.32%–37.6% and 11.18%–36.86%, respectively. Significant: The proposed approach could build the relationship between CBF autoregulation and impedance, and this method is noninvasive, continuous, and easy to execute. This study provides a basis for the health assessment of the CBF system using bioimpedance.
23 October 2024 in IEEE Transactions on Instrumentation and Measurement
Background: Surface electromyography-based gesture recognition and prosthetic hand control using deep learning have become increasingly significant in the field of human-computer interaction. Objective: This study aims to enhance the control of prosthetic hands driven by complex gestures, addressing the challenge of low-resolution gesture differentiation caused by the coupling and superposition of surface electromyography signals in deep learning models. Method: We propose a deep learning-based framework for the recognition of complex surface electromyography signals, utilizing a multi-pathway approach to acquire raw surface electromyography signals, process them in the time-frequency domain, and extract features using multi-scale convolutional networks. The processed surface electromyography features are then analyzed in parallel to enhance accuracy. Results: This method effectively processes multiple signals concurrently and extracts diverse feature sets. By collecting data from six channels, it achieves an 88.56% recognition rate for 16 complex hand gestures, enabling control of 10 distinct prosthetic hand movements. Conclusion: By leveraging multi-dimensional continuous surface electromyography images, we have developed a feature model that resolves the issues of signal coupling and superposition in multi-channel surface electromyography data, allowing for precise control of prosthetic hand movements.