Search

Premature ventricular contraction (PVC) is one of the most common arrhythmias which can cause palpitation, cardiac arrest, and other symptoms affecting the work and rest activities of a patient. However, patients hardly decipher their own feelings to determine the severity of the disease thus, requiring a professional medical diagnosis. This study proposes a novel method based on image processing and convolutional neural network (CNN) to extract electrocardiography (ECG) curves from scanned ECG images derived from clinical ECG reports, and segment and classify heartbeats in the absence of a digital ECG data. The ECG curve is extracted using a comprehensive algorithm that combines the OTSU algorithm with erosion and dilation. This algorithm can efficiently and accurately separate the ECG curve from the ECG background grid. The performance of the classification model was evaluated and optimized using hundreds of clinical ECG data collected from Fujian Provincial Hospital. Additionally, thousands of clinical ECG reports were scanned to digital images as the test set to confirm the accuracy of the algorithm for practical application. Results showed that the average sensitivity, specificity, positive predictive value, and accuracy of the proposed model on the MIT-BIH dataset were 95.47%, 97.72%, 98.75%, and 98.25%, respectively. The classification average sensitivity, specificity, positive predictive value, and accuracy based on clinical scanned ECG images can reach to 97.24%, 81.6%, 83.8%, and 89.33%, respectively, and the clinical feasibility is high. Overall, the proposed method can extract ECG curves from scanned ECG images efficiently and accurately. Furthermore, it performs well on heartbeat classification of normal (N) and ventricular premature heartbeat.

An interpatient classification method using the second lead for the Association for the Advancement of Medical Instrumentation (AAMI) standard is introduced. This study includes three parts. In the first part, the fuzzy matching algorithm is used to locate the key waveform points of Electrocardiogram (ECG) data. In the second part, the feature engineering algorithm is used to filter the extracted data sets. In the third part, the random forest model is carried out to realize the five classifications of heart disease by the selected features. The final precision, recall, and F1-score are 91%, 89%, and 90%, respectively.

Wearable ECG collection system for home detection has attracted increasing attention. Such devices are designed to be portable with fewer leads. Reconstruction from three-lead ECG to 12-lead ECG can provide useful auxiliary information for telemedicine dynamic ECG detection. In this study, two reconstruction algorithms are discussed, and the correlation coefficients of the reconstruction results are analyzed. The average correlation coefficient in the randomized experiment reaches 0.901.

This study proposed the design of a multi-lead ECG wearable device based on Bluetooth microcontroller. The main parts include the ECG sensing electrodes and the ECG acquisition hardware circuit. ECG signals from three leads of I, II, and V1 can be recorded by the sensing electrodes completely and steadily. The hardware circuit is designed to implement signal conversion and data transmission. Two sampling rates in software program: 100Hz and 400Hz, are available for this device. This device satisfies the real requirement of small size, hand-held and real-time acquisition. The experimental results show, the device can provide good-quality ECG signals.

This study proposes a noninvasive end-to-end blood pressure (BP) classification model based on electrocardiography (ECG), dividing BP into three categories, namely, normotension (N), prehypertension (Pre), and hypertension (H). This study includes three parts: (1) preprocessing ECG signals, including cutting signals into fixed-length segments and denoising; (2) building a proper ResNet model with a 1D convolutional neural network (1D-CNN) to train data; and (3) evaluating the performance of the classification model. The classification algorithm can identify three types of BP, and the accuracy rate of the classification algorithm over the test set is 87.89%.

This study proposes a personal healthcare system to monitor, identify ectopic heartbeats, and provide services to patients with heart diseases. The proposed system includes three subsystems: (1) an electrocardiography (ECG) signal acquisition device to acquire ECG signals accurately; (2) an Android-based smartphone to display the actual ECG waveform and upload the signal to the Web platform for analysis of cardiac disorder events; and (3) a Web platform for feature selection and arrhythmia classification with the support vector machine algorithm. The classification algorithm can identify three types of heartbeats, namely, normal, atrial premature beat, and premature ventricular contraction. The average sensitivity, specificity, and accuracy of the classification algorithm over the training set are 98.5%, 96.7%, and 97.7%, respectively.

This study proposed a classification method of premature ventricular contraction based on single lead electrocardiograph (ECG), and compared the classification results of AlexNet and convolutional neural network (CNN) model based on AlexNet. The results showed that the sensitivity, specificity, accuracy and positive predictive value of AlexNet on MIT-BIH dataset are 94.88%, 94.66%, 97.76% and 95.87%, respectively. The average specificity and positive predictive value of the improved model were 99.55% and 99.27%, respectively.

Power efficiency and stability are critical for wearable ECG application. The paper presents a Pulse-Width-Modulation (PWM) voltage type DC-DC converter, which has a small area of 0.32 mm2 on active chip, output voltage range of 0.8V to 3.5V, output power up to 7.5 mW, and peak efficiency of 95.8%. The simulation results show that the output voltage of the system is 1.8V, the output current is 4mA, the ripple voltage is 13mV and the efficiency is 86.1%, which meets the requirements of a presented system. The circuit will be fabricated by the SMIC in the near future with $0.18\ \mu\mathrm{m}$ Bipolar-CMOS-DMOS (BCD) process.

In this study, a dynamic respiratory signal acquisition method based on portable ECG acquisition device was proposed. Most EDR algorithms are suitable for static signal acquisition. ECG signal acquisition and EDR signal analysis in motion state has affected by motion artifact, resulting in low accuracy. The ambulate ECG signal analyzer proposed in this study uses a digital algorithm to remove the motion artifacts in the acquisition process, so that the stability of the ECG signal is close to that of the static signal, thus obtaining the dynamic respiratory signal with high accuracy.

Atrial fibrillation (AF) is the most common cardiovascular disease (CVD), and most existing algorithms are usually designed for the diagnosis (i.e., feature classification) or prediction of AF. Artificial intelligence (AI) algorithms integrate the diagnosis of AF electrocardiogram (ECG) and predict the possibility that AF will occur in the future. In this paper, we utilized the MIT-BIH AF Database (AFDB), which is composed of data from normal people and patients with AF and onset characteristics, and the AFPDB database (i.e., PAF Prediction Challenge Database), which consists of data from patients with Paroxysmal AF (PAF, the records contain the ECG preceding an episode of PAF), and subjects who do not have documented AF. We extracted the respective characteristics of the databases and used them in modeling diagnosis and prediction. In the aspect of model construction, we regarded diagnosis and prediction as two classification problems, adopted the traditional support vector machine (SVM) algorithm, and combined them. The improved quantum particle swarm optimization support vector machine (IQPSO-SVM) algorithm was used to speed the training time. During the verification process, the clinical FZU-FPH database created by Fuzhou University and Fujian Provincial Hospital was used for hybrid model testing. The data were obtained from the Holter monitor of the hospital and encrypted. We proposed an algorithm for transforming the PDF ECG waveform images of hospital examination reports into digital data. For the diagnosis model and prediction model trained using the training set of the AFDB and AFPDB databases, the sensitivity, specificity, and accuracy measures were 99.2% and 99.2%, 99.2% and 93.3%, and 91.7% and 92.5% for the test set of the AFDB and AFPDB databases, respectively. Moreover, the sensitivity, specificity, and accuracy were 94.2%, 79.7%, and 87.0%, respectively, when tested using the FZU-FPH database with 138 samples of the ECG composed of two labels. The composite classification and prediction model using a new water-fall ensemble method had a total accuracy of approximately 91% for the test set of the FZU-FPH database with 80 samples with 120 segments of ECG with three labels.

This study presents a low-power multi-lead wearable electrocardiogram (ECG) signal sensor system design that can simultaneously acquire the electrocardiograms from three leads, I, II, and V1. The sensor system includes two parts, an ECG test clothing with five electrode patches and an acquisition device. Compared with the traditional 12-lead wired ECG detection instrument, which limits patient mobility and needs medical staff assistance to acquire the ECG signal, the proposed vest-type ECG acquisition system is very comfortable and easy to use by patients themselves anytime and anywhere, especially for the elderly. The proposed study incorporates three methods to reduce the power consumption of the system by optimizing the micro control unit (MCU) working mode, adjusting the radio frequency (RF) parameters, and compressing the transmitted data. In addition, Huffman lossless coding is used to compress the transmitted data in order to increase the sampling rate of the acquisition system. It makes the whole system operate continuously for a long period of time and acquire abundant ECG information, which is helpful for clinical diagnosis. Finally, a series of tests were performed on the designed wearable ECG device. The results have demonstrated that the multi-lead wearable ECG device can collect, process, and transmit ECG data through Bluetooth technology. The ECG waveforms collected by the device are clear, complete, and can be displayed in real-time on a mobile phone. The sampling rate of the proposed wearable sensor system is 250 Hz per lead, which is dependent on the lossless compression scheme. The device achieves a compression ratio of 2.31. By implementing a low power design on the device, the resulting overall operational current of the device is reduced by 37.6% to 9.87 mA under a supply voltage of 2.1 V. The proposed vest-type multi-lead ECG acquisition device can be easily employed by medical staff for clinical diagnosis and is a suitable wearable device in monitoring and nursing the off-ward patients.

The study proposed a switched-capacitor delta-sigma modulator with single-loop and third-order structures for wearable electrocardiogram acquisition device for low-power and high-precision applications. Several techniques, including chain of integrators with weighted feedforward structure, gain-enhanced current-mirror push-pull operational transconductance amplifier, and dynamic comparator with a set-reset latch are adopted to improve the signal-to-noise ratio (SNR) and reduce the power consumption. Simulation results shows that the SNR and the corresponding effective number of bits are 104.2 dB and 17.02 bits, respectively, which fulfill the requirements of an presented system. The circuit will be fabricated by the SMIC using a 0.18 μm general purpose process in the near future.

Motion artifacts can cause baseline shifts and disrupt the waveform of the ECG signal. There will be a serious impact on the diagnosis when motion artifacts not removed. In this paper, a two-stage adaptive filter is proposed. In the first stage, the reference ECG signals are measured under static status, and the noise signals also be obtained. The obtained noise signals become reference signals for the second stage, and finally the desired ECG signal obtained. The experimental results show that the two-stage adaptive filter can filter out the noise of 0. $1\sim 10\mathrm{Hz}$ in the ECG signal which containing motion artifacts well.

In this study, a multi-lead wearable ECG signal device is designed, which can simultaneously acquire the electrocardiograms of three leads (I, II, and V1) and be placed on the human body by ECG test clothing. To meet the low power requirements of wearable devices, the hardware and software are designed at the same time. Results show that the ECG waveforms collected by this device are clear, and each waveform is complete. When the low power design on the device is implemented, the power consumption of the entire device decreases by 31%.

The current paper presents an outdoor monitoring system for elderly people, which can transmit information on physiological signals and falling events to a healthcare center at any time and from any place. To detect simultaneously the occurrence of any falling event, as well as the relative electrocardiogram (ECG) signal of the user, a multi-thread method is proposed with the objective of enhancing the response time and the accuracy of detection. A healthcare box is used to determine the relative position of the patient through a global positioning system for fall detection; moreover, an ECG signal acquisition thread is adopted to increase the precision of the fall detection system. Integrating a precise map into the monitoring system facilitates understanding of a client user of the correct location and surrounding environment using the portable display. According to experimental results based on 4,000 samples, successful detection time with the multi-thread method was reduced by 38%, thereby increasing rescue opportunities for elderly patients who are at risk1.