An Adaptive Median Filter Based on Sampling Rate for R-Peak Detection and Major-Arrhythmia Analysis

Certain investigation on hybrid neural network method for classification of ECG signal with the suitable a FIR filter

The Electrocardiogram (ECG) records are crucial for predicting heart diseases and evaluating patient's health conditions. ECG signals provide essential peak values that reflect reliable health information. Analyzing ECG signals is a fundamental technique for computerized prediction with advancements in Very Large-Scale Integration (VLSI) technology and significantly impacts in biomedical signal processing. VLSI advancements focus on high-speed circuit functionality while minimizing power consumption and area occupancy. In ECG signal denoising, digital filters like Infinite Impulse Response (IIR) and Finite Impulse Response (FIR) are commonly used. The FIR filters are preferred for their higher-order performance and stability over IIR filters, especially in real-time applications. The Modified FIR (MFIR) blocks were reconstructed using the optimized adder-multiplier block for better noise reduction performance. The MIT-BIT database is used as reference where the noises are filtered by the MFIR based on Optimized Kogge Stone Adder (OKSA). Features are extracted and analyzed using Discrete wavelet transform (DWT) and Cross Correlation (CC). At this modern era, Hybrid methods of Machine Learning (HMLM) methods are preferred because of their combined performance which is better than non-fused methods. The accuracy of the Hybrid Neural Network (HNN) model reached 92.3%, surpassing other models such as Generalized Sequential Neural Networks (GSNN), Artificial Neural Networks (ANN), Support Vector Machine with linear kernel (SVM linear), and Support Vector Machine with Radial Basis Function kernel (SVM RBF) by margins of 3.3%, 5.3%, 23.3%, and 24.3%, respectively. While the precision of the HNN is 91.1%, it was slightly lower than GSNN and ANN but higher than both SVM linear and SVM -RBF. The HNN with various features are incorporated to improve the ECG classification. The accuracy of the HNN is switched to 95.99% when the DWT and CC are combined. Also, it improvises other parameters such as precision 93.88%, recall is 0.94, F1 score is 0.88, Kappa is 0.89, kurtosis is 1.54, skewness is 1.52 and error rate 0.076. These parameters are higher than recently developed models whose algorithms and methods accuracy is more than 90%.

Robust human locomotion and localization activity recognition over multisensory

Human activity recognition (HAR) plays a pivotal role in various domains, including healthcare, sports, robotics, and security. With the growing popularity of wearable devices, particularly Inertial Measurement Units (IMUs) and Ambient sensors, researchers and engineers have sought to take advantage of these advances to accurately and efficiently detect and classify human activities. This research paper presents an advanced methodology for human activity and localization recognition, utilizing smartphone IMU, Ambient, GPS, and Audio sensor data from two public benchmark datasets: the Opportunity dataset and the Extrasensory dataset. The Opportunity dataset was collected from 12 subjects participating in a range of daily activities, and it captures data from various body-worn and object-associated sensors. The Extrasensory dataset features data from 60 participants, including thousands of data samples from smartphone and smartwatch sensors, labeled with a wide array of human activities. Our study incorporates novel feature extraction techniques for signal, GPS, and audio sensor data. Specifically, for localization, GPS, audio, and IMU sensors are utilized, while IMU and Ambient sensors are employed for locomotion activity recognition. To achieve accurate activity classification, state-of-the-art deep learning techniques, such as convolutional neural networks (CNN) and long short-term memory (LSTM), have been explored. For indoor/outdoor activities, CNNs are applied, while LSTMs are utilized for locomotion activity recognition. The proposed system has been evaluated using the k-fold cross-validation method, achieving accuracy rates of 97% and 89% for locomotion activity over the Opportunity and Extrasensory datasets, respectively, and 96% for indoor/outdoor activity over the Extrasensory dataset. These results highlight the efficiency of our methodology in accurately detecting various human activities, showing its potential for real-world applications. Moreover, the research paper introduces a hybrid system that combines machine learning and deep learning features, enhancing activity recognition performance by leveraging the strengths of both approaches.

Multilayer Perceptron-Based Real-Time Intradialytic Hypotension Prediction Using Patient Baseline Information and Heart-Rate Variation

Intradialytic hypotension (IDH) is a common side effect that occurs during hemodialysis and poses a great risk for dialysis patients. Many studies have been conducted so far to predict IDH, but most of these could not be applied in real-time because they used only underlying patient information or static patient disease information. In this study, we propose a multilayer perceptron (MP)-based IDH prediction model using heart rate (HR) information corresponding to time-series information and static data of patients. This study aimed to validate whether HR differences and HR slope information affect real-time IDH prediction in patients undergoing hemodialysis. Clinical data were collected from 80 hemodialysis patients from 9 September to 17 October 2020, in the artificial kidney room at Yeungnam University Medical Center (YUMC), Daegu, South Korea. The patients typically underwent hemodialysis 12 times during this period, 1 to 2 h per session. Therefore, the HR difference and HR slope information within up to 1 h before IDH occurrence were used as time-series input data for the MP model. Among the MP models using the number and data length of different hidden layers, the model using 60 min of data before the occurrence of two layers and IDH showed maximum performance, with an accuracy of 81.5%, a true positive rate of 73.8%, and positive predictive value of 87.3%. This study aimed to predict IDH in real-time by continuously supplying HR information to MP models along with static data such as age, diabetes, hypertension, and ultrafiltration. The current MP model was implemented using relatively limited parameters; however, its performance may be further improved by adding additional parameters in the future, further enabling real-time IDH prediction to play a supporting role for medical staff.

Self-Reporting Technique-Based Clinical-Trial Service Platform for Real-Time Arrhythmia Detection

The analysis of the electrocardiogram (ECG) is critical for the diagnosis of arrhythmias. Recent advances in information and communications technology (ICT) have led to the development of wearable ECG devices and arrhythmia-detection algorithms. This study aimed to develop an ICT-based clinical trial service platform using a self-reporting technique for real-time arrhythmia detection. To establish a clinical-trial service platform, a mobile application (app), a demilitarized zone (DMZ), an internal network, and Amazon web services virtual private cloud (AWS-VPC) were developed. The ECG data acquired by a wearable device were transmitted to the mobile app, which collected the participants’ self-reported information. The mobile app transmitted raw ECG and self-reported data to the AWS-VPC and DMZ, respectively. In the AWS-VPC, the live-streaming and playback-reviewer services were operational to display the currently and previously acquired ECG data to clinicians through the web client. All the measured data were transmitted to the internal network, in which the arrhythmia-detection algorithm was executed and all the data were saved. The self-reporting technique and arrhythmia-detection algorithm are the key elements of this platform. In particular, subjective information of participants can be easily collected using a self-reporting technique. These features are particularly of critical importance for treating painless, sparsely occurring arrhythmias.

[Heartbeat-aware convolutional neural network for R-peak detection of wearable device ECG data]

OBJECTIVE

To develop a method for R-peak detection of ECG data from wearable devices to allow accurate estimation of the physiological parameters including heart rate and heart rate variability.

METHODS

A fully convolutional neural network was applied to predict the R-peak heatmap of ECG data and locate the R-peak positions. The heartbeat-aware (HA) module was introduced to enable the model to learn to predict the heartbeat number and R-peak heatmap simultaneously, thereby improving the capability of the model for extraction of the global context. The R-R interval estimated by the predicted heartbeat number was adopted to calculate the minimum horizontal distance for peak positioning. To achieve real-time R-peak detection on mobile devices, the deep separable convolution was adopted to reduce the number of parameters and the computational complexity of the model.

RESULTS

The proposed model was trained only with ECG data from wearable devices. At a tolerance window interval of 150 ms, the proposed method achieved R peak detection sensitivities of 100% for both wearable device ECG dataset and a public dataset (i.e. LUDB), and the true positivity rates exceeded 99.9%. As for the ECG signal of a 10 s duration, the CPU time of the proposed method for R-peak detection was about 23.2 ms.

CONCLUSION

The proposed method has good performance for R-peak detection of both wearable device ECG data and routine ECG data and also allows real-time R-peak detection of the ECG data.

Medical Data Acquisition and Internet of Things Technology-Based Cerebral Stroke Disease Prevention and Rehabilitation Nursing Mobile Medical Management System

This research was aimed at exploring the application value of a mobile medical management system based on Internet of Things technology and medical data collection in stroke disease prevention and rehabilitation nursing. In this study, on the basis of radio frequency identification (RFID) technology, the signals collected by the sensor were filtered by the optimized median filtering algorithm, and a rehabilitation nursing evaluation model was established based on the backpropagation (BP) neural network. The performance of the medical management system was verified in 32 rehabilitation patients with hemiplegia after stroke and 6 healthy medical staff in the rehabilitation medical center of the hospital. The results showed that the mean square error (MSE) and peak signal-to-noise ratio (PSNR) of the median filtering algorithm after optimization were significantly higher than those before optimization (P < 0.05). When the number of neurons was 23, the prediction accuracy of the test set reached a maximum of 89.83%. Using traingda as the training function, the model had the lowest training time and root mean squared error (RMSE) value of 2.5 s and 0.29, respectively, which were significantly lower than the traingd and traingdm functions (P < 0.01). The error percentage and RMSE of the model reached a minimum of 7.56% and 0.25, respectively, when the transfer functions of both the hidden and input layers were tansig. The prediction accuracy in stages III~VI was 90.63%. It indicated that the mobile medical management system established based on Internet of Things technology and medical data collection has certain application value for the prevention and rehabilitation nursing of stroke patients, which provides a new idea for the diagnosis, treatment, and rehabilitation of stroke patients.

Lessons from a multicenter clinical trial with an approved wearable electrocardiogram: issues and practical considerations

Although wearable electrocardiograms (ECGs) are being increasingly applied in clinical settings, validation methods have not been standardized. As an exploratory evaluation, we performed a multicenter clinical trial implementing an approved wearable patch ECG. Healthy male adults were enrolled in 2 study centers. The approved ECGs were deployed for 6 hours, and pulse rates were measured independently with conventional pulse oximetry at selected time points for correlation analyses. The transmission status of the data was evaluated by heart rates and classified into valid, invalid, and missing. A total of 55 subjects (40 in center 1 and 15 in center 2) completed the study. Overall, 77.40% of heart rates were within the valid range. Invalid and missing data accounted for 1.42% and 21.23%, respectively. There were significant differences in valid and missing data between centers. The proportion of missing data in center 1 (24.77%) was more than twice center 2 (11.77%). Heart rates measured by the wearable ECG and conventional pulse oximetry showed a poor correlation (intraclass correlation coefficient = 0.0454). In conclusion, we evaluated the multicenter feasibility of implementing wearable ECGs. The results suggest that systems to mitigate multicenter discrepancies and remove artifacts should be implemented prior to performing a clinical trial.

The Comparison Features of ECG Signal with Different Sampling Frequencies and Filter Methods for Real-Time Measurement

Electrocardiogram (ECG) signals have been used to monitor and diagnose signs of cardiovascular disease and abnormal signals about the human body. ECG signals are typically characterized by the PR, QRS, QT interval, ST-segment, and heart rate (HR) parameters. ECG devices are widely used for many applications, especially for the elderly. However, ECG signals are often affected by noises from the environment. There are mainly two types of noises that affect the ECG signals: low frequencies from muscle activity and 50/60 Hz from the electrical grid. Removing these noises is important for improving the quality of the ECG signal. A clear ECG signal makes it easy to diagnose cardiovascular problems. ECG signals with high sampling frequency are more accurate. However, the noises in the signal will be more obvious and it will be difficult to remove these noises with filters. We analyzed the symmetrical correlation between the sampling frequency of the signal and the parameters of the signal such as signal to noise ratio (SNR) and signal amplitude. This study will compare characterization of ECG signals performed at different sampling frequencies before and after applying infinite impulse response (IIR) and symmetric finite impulse response (FIR) filters. Therefore, it is critical that the sampling frequency is consistent at the same frequency of the ECG signal for accurate diagnosis. Furthermore, the approach can be also important for the device to help reduce the device’s computing power and hardware resources. Our results were tested with the MIT/ BIH database at 360 Hz sampling frequency with 11-bit resolution. We also experimented with the device operating in real-time with a sampling frequency from 100 Hz to 2133 Hz and a 24-bit resolution. The test results show the advantages of the symmetric FIR filter over IIR when applied to the filtering of ECG signals. The study’s conclusions can be applied to real-world devices to improve the quality of ECG signals.

Integration of Internet of Things with Quantum Dots: A State-of-the-art of Medicine

Internet of Things (IoT) emerges as disruptive innovation and development in the fields of drug delivery and biomedical sciences using on-target active transportation, sensors, wearable devices, real-time diagnostics, etc. Semiconducting fluorescence emitting material, quantum dots on integration with IoT displayed interesting results in the healthcare sector, especially in hospitals and pathological laboratories. Presently, the integrated system is used to improve productivity without the interference of human and offer a cost-effective system. This integrated system can be used for the detection of various diseases like epilepsy, cancer, diabetes, etc., and various biomedical applications like energy storage, lights, sensor technology, light filters, etc. The integrated technology is implemented into the field of medicine for simplifying the approaches in therapeutics and diagnostic applications. The collected and analyzed data are further useful for healthcare professionals to find patient-centric solutions. Artificial Intelligence-aided IoT emerges as a novel technology for transmitting and securing health data. Despite some of the limitations like e-waste and the risk of hacking, an IoT-based QD system will be considered as a modern healthcare provider with life-saving products for enriching the medical quality and real-time accessibility.

Automatic Premature Ventricular Contraction Detection Using Deep Metric Learning and KNN

Premature ventricular contractions (PVCs), common in the general and patient population, are irregular heartbeats that indicate potential heart diseases. Clinically, long-term electrocardiograms (ECG) collected from the wearable device is a non-invasive and inexpensive tool widely used to diagnose PVCs by physicians. However, analyzing these long-term ECG is time-consuming and labor-intensive for cardiologists. Therefore, this paper proposed a simplistic but powerful approach to detect PVC from long-term ECG. The suggested method utilized deep metric learning to extract features, with compact intra-product variance and separated inter-product differences, from the heartbeat. Subsequently, the k-nearest neighbors (KNN) classifier calculated the distance between samples based on these features to detect PVC. Unlike previous systems used to detect PVC, the proposed process can intelligently and automatically extract features by supervised deep metric learning, which can avoid the bias caused by manual feature engineering. As a generally available set of standard test material, the MIT-BIH (Massachusetts Institute of Technology-Beth Israel Hospital) Arrhythmia Database is used to evaluate the proposed method, and the experiment takes 99.7% accuracy, 97.45% sensitivity, and 99.87% specificity. The simulation events show that it is reliable to use deep metric learning and KNN for PVC recognition. More importantly, the overall way does not rely on complicated and cumbersome preprocessing.