Portable Electrocardiograph Research Articles

Design a Simple Portable Electrocardiograph Based on the IOT Concept

Abstract There are many cases of heart disease with death associated. The symptoms are not always clearly visible and require examination using an electrocardiogram (ECG) device to be detected. Recording human electrical activity can be done in various organs of the human body, one of which is in the heart. An ECG is a medical test to detect electrical activity signals produced by the heart by exchanging the signals on a monitor or graph on paper. This study aimed to design, build, test, and analyze a portable ECG device that is compact and easy to use. This portable electrocardiogram helps monitor human heart rate to check heart rate periodically. Tool design is done by making tools equipped with a monitoring system. The proposed system uses the AD8232 sensor to detect a person’s cardiac activity. Arduino Nano is a microcontroller connected to ESP32 with a WiFi feature. Heart activity monitors can be done anywhere by applying the Internet of Things concept, namely with the Ubidots application. A portable tool using Li-Ion batteries as a tool voltage source and BMS 3S as a charger, this tool is also equipped with a 3.2-inch TFT screen as a graph viewer of the results of monitoring heart activity. The results showed that the system could monitor the user’s heart for 5 minutes when sitting, walking, and running. The system created has been able to send heart signal data to be processed by Arduino Nano and ESP32 as a data transfer tool so that the data generated in the form of electrocardiogram graphs can be displayed on the Ubidots application. TFT LCDs can display electrocardiogram graphs generated by the electrical activity of the human heart and beats per minute values. The AD8232 sensor can measure beat per minute (bpm) values with an average increase in bpm values for walking conditions of 17% and running conditions of 19.17%. The AD8232 Sensor Performance can detect the user’s heart rate with a percentage of data difference between walking and running conditions of 2.88% with an accuracy rate of 97.12%.

Design of a low-cost portable electrocardiograph for telemedicine application

Design of a low-cost portable electrocardiograph for telemedicine application

Preliminary survey on the effects of indoor noise in the hospital located close to Tan Son Nhat Airport

An awareness of the harmful effects of environmental noise on vulnerable populations is crucial for improving current guidelines for the design of welfare facilities and land use planning in developing countries. This paper aims to identify the impact of aircraft noise on patients' health in areas with a high frequency of aircraft noise. To achieve this, a preliminary investigation was conducted in September 2022 at a hospital located approximately 1 km from the southeast runway end of Tan Son Nhat Airport (TSN) in Ho Chi Minh City. The noise levels inside and outside the hospital rooms were measured throughout the day to assess the noise exposure levels. The health data of 28 nursing staff living in the hospital's inpatient ward and four patients were collected through face-to-face interviews and the heart rate variability (HRV) of two of them was measured by using portable electrocardiograph monitors attached to their bodies. As an objective evaluation of the effects of noise, HRV data was analyzed to reveal the impact of noise exposure on health. The noise exposure and health data were then analyzed using multivariate analysis. The results of the analysis provide a clear understanding of the impact of aircraft noise on patients' health in areas with a high frequency of aircraft noise.

Cuffless Blood Pressure Estimation Using Dual Physiological Signal and Its Morphological Features

Prehypertension is difficult to diagnose early because of its hidden nature. Long-term monitoring of blood pressure (BP) can help in the early detection and timely treatment of this condition. This study proposes an innovative and efficient BP detection platform that combines portable electrocardiography (ECG) and photoplethysmogram (PPG) signals simultaneous acquisition equipment and BP detection algorithm to obtain real-time BP values conveniently and accurately for a long time. In this study, nine kinds of feature parameters and classification algorithm are used to build multiple linear regression (MLR) models. It not only adopts the Multi-Parameter Intelligent Monitoring in Intensive Care Units (MIMIC-II) database to train and validate the model but also uses self-developed equipment for acquisition and verification in long-term health monitoring. According to the experimental results, the mean absolute error (MAE) and standard deviation (SD) of systolic BP (SBP) have estimated values of 4.46 and 3.20 mmHg, respectively, and simultaneously, the MAE and SD of diastolic BP (DBP) are 4.20 and 3.28 mmHg, respectively. Moreover, both SBP and DBP experimental results conform to the Advancement of Medical Instrumentation BP standard. The proposed BP acquisition platform is proven to be capable of easily acquiring ECG and PPG signals with the proposed sensor device, and the MLR algorithm can also effectively and accurately monitor BP values for a long time.

Remote Monitoring of the QT Interval During Polychemotherapy

Aim To evaluate the incidence of prolonged corrected QT interval (QTc) by remote single-channel electrocardiogram (ECG) monitoring in primary oncological patients with elective polychemotherapy (PCT).Material and methods This study included 49 oncological patients with elective PCT. A single-channel portable CardioQVARK electrocardiograph was used to record single-channel, one-lead ECG between the first and second courses of PCT.Results Analysis of QTc interval detected a prolonged QTc interval >500 msec in 8.2 % of cases, prolonged QTc >480 msec in 18.3 % f cases, and prolonged QTc interval >60 msec compared to baseline in 12.2 % of cases.Conclusion Remote recording of single-channel ECG using a portable electrocardiograph is an effective method for recording and detecting various forms of heart rhythm disorders.

Monitoring with wearable devices will clarify the association between indoor temperature and blood pressure.

The association of blood pressure and temperature is well known in seasonal observation, and low temperature in the winter season is often considered a cause of high blood pressure. The current evidence for short-term studies of temperature and blood pressure is based on the daily observation, however continuous monitoring with wearable devices will allow us to evaluate the rapid effect of cold temperature exposure on blood pressure. In a Japanese, prospective intervention study from 2014 to 2019 (the Smart Wellness Housing survey), approximately 90% of Japanese lived in cold houses (indoor temperature less than 18 °C). Importantly, the indoor temperature was associated with the increase of morning systolic blood pressure. We recently addressed the sympathetic nervous activation of individuals in both their houses and a highly insulated and airtight model house in the winter season using portable electrocardiography equipment. A few subjects showed a morning surge in sympathetic activity, which was more intense at their cold houses, which suggests the importance of the indoor environment in the management of early morning hypertension. In near future, real-time monitoring with wearable devices will provide important information for a better life-environment, leading to risk reduction of morning surge and cardiovascular events.

PS-BPR02-10: SYMPATHETIC NERVE TONE AND THE INDOOR ENVIRONMENT IN A HIGHLY INSULATED AND AIRTIGHT MODEL HOUSE DURING THE COLD WINTER SEASON

While evidence for lifestyle modification such as salt reduction, weight loss, exercise, sobriety, and smoking cessation has accumulated for perfect blood pressure control, there is a lack of evidence on blood pressure fluctuations due to housing environments such as thermal protection and heating. In particular, blood pressure fluctuation due to the difference in room temperature between the living room and the dressing room, toilet, and bedroom is known to be a major risk during the winter season. Therefore, this study focused on the autonomic nervous system as one of the factors causing blood pressure fluctuations, and compared the fluctuations when the same subjects spent two days each at home and in a highly insulated, airtight model house. JOURNAL/jhype/04.03/00004872-202301001-00877/figure1/v/2023-10-24T163949Z/r/image-jpeg Subjects were 6 males and 5 females with 20–70 years old. Autonomic activity was measured and calculated using portable electrocardiography (Heartnote: JSR corporation, Tokyo) for the 8 subjects for whom it was possible to measure heart rate, CVRR (Coefficient of variation of R-R interval) and LF(low frequency)/HF(high frequency) ratio. An sympathetic hyperactivation was defined to show more than +2SD for distribution within an individual in LF/HF calculated every 10 seconds. The salivary amylase was measured at several times per day. The indoor environment temperatures were recorded by installing data loggers equipped with temperature sensors in various rooms and taking continuous measurements at 10-min intervals. In the model house, room temperatures were kept constant at 18°C or more. However, in the home, there were large temperature differences between rooms, and in one case, a room temperature in the bedroom was recorded at 6.3°C. In the overall analysis of heart rate and CVRR, there were large individual differences in both absolute values and diurnal variations, and no obvious differences by housing environment. In a detailed analysis focusing on early morning (2 hours after waking), two subjects showed excessive sympathetic activation in the early morning at home, but not in the model house. There was no correlation between this change and salivary amylase. In terms of sleep quality, there were no differences by housing environment. This case of early morning sympathetic activation in a residential setting suggests the importance of the residential setting in the management of early morning hypertension.

Pilot experience using a portable electrocardiography device for atrial fibrillation detection in an outpatient stroke clinic.

Pilot experience using a portable electrocardiography device for atrial fibrillation detection in an outpatient stroke clinic.

Atrial fibrillation can be diagnosed using a combined wrist-worn photoplethysmography and electrocardiography

Abstract Background Current ESC Clinical Practice Guideline for management of atrial fibrillation requires a minimum of 30-seconds of atrial fibrillation (AF) to be visually confirmed from an electrocardiography (ECG) recording for diagnosis. Single-channel ECG recorded from any measurement lead is sufficient. The current method of 24–48 h Holter ECG may not detect rarely occurring AF episodes, and thus extended duration recording devices are needed. Intermittent portable ECG devices intended for clinical use exist but those lack continuous rhythm monitoring and are thus unable to notify the user in case of asymptomatic episode. Purpose We evaluated the performance of a wrist-worn solution (Fig. 1) intended for clinical use that combines continuous photoplethysmography (PPG) -based rhythm assessment with intermittent ECG measurements in the detection of atrial fibrillation in ambulatory conditions. Methods 30 adult patients with previous diagnosis of AF were recruited for 48-hour measurements. Simultaneous 5-lead Holter monitoring was used as the reference. The patients were taking intermittent ECG measurements at pre-scheduled times (4 times a day), when notified by the device (irregular rhythm detected by PPG), and if feeling symptoms of AF. Results The obtained sensitivity and specificity for PPG-based AF detection were 95.5% and 99.2%, respectively when evaluated in 5-minute segments. Only the segments during which adequate amount, approximately 30-seconds, of good quality PPG data was available, were considered in the analysis. On average this equals 54% of the segments. Patients took in total 456 30-second ECG measurements. Out of these 411 and 392 were assessed as analysable quality by ECG algorithm and visually by two cardiologists, HS and KK, respectively. Automatic ECG analysis algorithm labelled ECG measurements taken during atrial fibrillation correctly as “Possible arrhythmia” with 95.7% sensitivity and 92.0% specificity. 99.5% of the analysable quality ECGs measured during AF were marked with any type of label e.g., tachycardia that highlights them to the clinician. An example ECG measurement with markings of the analysis algorithm an overall classification is shown in Fig. 2. All patients having AF during the 48-hour monitoring (14/30) were correctly identified based on the wrist device data. Conclusion Based on the obtained results, wrist-worn device combining continuous PPG and intermittent ECG measurements can be reliably used in detecting atrial fibrillation in ambulatory conditions. The user-friendliness and unobtrusiveness of wrist-worn technology enables virtually unlimited monitoring time thus facilitating the detection of rarely occurring AF episodes. The information of continuous rhythm monitoring can also be used for the estimation of AF burden. The solution used in the study has received medical device CE certification in early 2022. Funding Acknowledgement Type of funding sources: Private company. Main funding source(s): The study was funded by PulseOn company.

Wearable Fetal ECG Monitoring System from Abdominal Electrocardiography Recording.

Fetal electrocardiography (ECG) monitoring during pregnancy can provide crucial information for assessing the fetus’s health status and making timely decisions. This paper proposes a portable ECG monitoring system to record the abdominal ECG (AECG) of the pregnant woman, comprising both maternal ECG (MECG) and fetal ECG (FECG), which could be applied to fetal heart rate (FHR) monitoring at the home setting. The ECG monitoring system is based on data acquisition circuits, data transmission module, and signal analysis platform, which consists of low input-referred noise, high input impedance, and high resolution. The combination of the adaptive dual threshold (ADT) and the independent component analysis (ICA) algorithm is employed to extract the FECG from the AECG signals. To validate the performance of the proposed system, AECG is recorded and analyzed of pregnant women in three different postures (supine, seated, and standing). The result shows that the proposed system can record the AECG in different postures with good signal quality and high accuracy in fetal ECG and heart rate information. Sensitivity (Se), positive predictive accuracy (PPV), accuracy (ACC), and their harmonic mean (F1) are utilized as the metrics to evaluate the performance of the fetal QRS (fQRS) complexes extraction. The average Se, PPV, ACC, and F1 score are 99.62%, 97.90%, 97.40%, and 98.66% for the fQRS complexes extraction,, respectively. This paper shows the proposed system has a promising application in fetal health monitoring.

Possibilities of using determination of allelic polymorphism of interleukin-6 G174C and tumour necrosis factor-α G308A genes for the prediction of cardiovascular disorders in children with juvenile idiopathic arthritis

Objectives: Juvenile idiopathic arthritis is a chronic disease that affects the synovial membrane of the joints, but can also lead to secondary lesions of the cardiovascular system. The most important mechanism of myocardial damage is associated with the effect of proinflammatory cytokines. The aim of the study was to propose a method of early detection of cardiovascular system changes and lesions in patients with juvenile idiopathic arthritis based on the determination of allelic polymorphism combined with electrocardiography data. Materials and methods: 102 patients with juvenile idiopathic arthritis underwent a general clinical examination. The overall activity of juvenile idiopathic arthritis was assessed using the Juvenile Arthritis Disease Activity Score. In addition, the patients underwent an electrocardiographic evaluation using the software and hardware complex “Cardioplus P,” which is a portable electrocardiograph providing “signal-averaged” electrocardiography performing the recognition and measurement of amplitude-time parameters, and calculation of secondary electrocardiography parameters. The genotypes of patients were additionally determined by alleles of the TNF-α (G308A) and IL6 (G174C) genes by polymerase chain reaction. Results: The overall number of mutations affects the course of the disease, with two or more mutations being associated with a more aggressive course of the disease, a more pronounced degree of inflammation, and a higher frequency of extra-articular lesions. The complex indicator of the functional state of the myocardium according to the electrocardiography data differed significantly (p = 0.00001) in clusters. Conclusion: Patients with juvenile idiopathic arthritis with two or more mutations in different genes of proinflammatory cytokines have a higher activity of the inflammatory process and a higher frequency of cardiovascular changes according to 4th generation electrocardiography. The determination of polymorphism may be useful in evaluating the risk of development of cardiovascular system abnormalities.

ECG Graph Monitoring System using Ad8232 with ECG Sensor and Arduino

Patient monitoring is the heart of the fitness care domain in day to day lifestyles both at domestic or at medical institution. This paper affords the layout of a transportable electrocardiograph (ECG) device the use of the AD8232 microchip as the analog front-stop (AFE). Beginning with the producer’s evaluation board of the AFE chip for testing circuit configurations, open-supply hardware and software components had been integrated into a breadboard prototype. In the long run, a custom printed circuit board (PCB) changed into produced. The prototype required to accommodate the microchip on a SMD-to-DIP adapter for checking out with the breadboard-pleasant Arduino microcontroller alongside a statistics logger and a Bluetooth breakout board. The analog ECG sign from the AFE output changed into digitized the use of one channel of the 10-bit analog-to-virtual Converter (ADC) of the ATmega328 microcontroller contained in the Arduino Nano board. The digitized ECG signal may be transmitted not simplest with the aid of serial cable using the Arduino capabilities, but also thru Bluetooth to a computer or to an Android telephone gadget whilst the HC-06 guard is used. The records logging guard presents gigabytes of garage, and the sign is recorded to a micro SD card adapter in conjunction with the date and time stamp information of the pattern seize (actual-time clock provided). further to hardware and software improvement, a simulation was used in the analog circuit design with SPICE Multisim software program and the related macro model library to assess machine stability. besides the analog filters within the AFE degree, virtual filtering through easy distinction equations become investigated. A menu turned into incorporated to pick out from the several modes of operation of the device. The ECG test signals were acquired from a affected person simulator (SimCube) and real sufferers. A portable ECG gadget for tracking applications that complies with electric protection regulations and medical equipment design became found out.

Mobile Single-Lead Electrocardiogram Technology for Atrial Fibrillation Detection in Acute Ischemic Stroke Patients.

(1) Background: AliveCor KardiaMobile (KM) is a portable electrocardiography recorder for detection of atrial fibrillation (AF). The aim of the study was to define the group of acute ischemic stroke (AIS) patients who can use the KM device and assess the diagnostic test accuracy. (2) Methods: the AIS patients were recruited to the study. Thirty-second single-lead electrocardiogram (ECG) usages were recorded on demand for three days using KM portable device. Each KM ECG record was verified by a cardiologist. The feasibility was evaluated using operationalization criteria. (3) Results: the recruitment rate among AIS patients was 26.3%. The withdrawal rate before the start of the intervention was 26%. The withdrawal rate after the start of the intervention was 6%. KM device detected AF in 2.8% of AIS patients and in 2.2% of ECG records. Cardiologist confirmed the AF in 0.3% AIS patients. Sensitivity and specificity of KM for AF was 100% and 98.3%, respectively. (4) Conclusions: the results of this study suggest that it is feasible to use KM device to detect AF in the selected AIS patients (younger and in better neurological condition). KM detected AF in the selected AIS patients with high specificity and sensitivity.

A Portable Electrocardiograph System Design based on STM32 Chip

A Portable Electrocardiograph System Design based on STM32 Chip

SLC-GAN: An automated myocardial infarction detection model based on generative adversarial networks and convolutional neural networks with single-lead electrocardiogram synthesis

Electrocardiography (ECG) is a sophisticated tool for the diagnosis of myocardial infarction (MI). Deep learning approaches can support MI diagnosis based on ECG data. However, due to privacy issues and the sensitive nature of ECG data, the use of deep learning methods to access large amounts of ECG data for training remains challenging. Insufficient data will lead to a decrease in diagnosis performance. In addition, most deep learning approaches are trained with multiple-lead ECG data, thereby limiting the extension to portable single-lead ECG monitoring devices. Investigating automated single-lead ECG interpretation is a continuing concern in the development of mobile and wearable ECG monitoring devices. This paper proposes an automated myocardial infarction detection model — SLC-GAN — that synthesizes single-lead ECG data with high morphological similarity through generative adversarial networks (GANs). It automatically detects MI using convolutional neural networks (CNNs) with real ECG data and synthetic ECGs from the GAN. The empirical results indicate that our new SLC-GAN method performs impressively compared with other MI classification methodologies on single-lead ECG from the PTB Diagnostic ECG Database. The MI classification accuracy of SLC-GAN reaches 99.06% with 5-fold cross-validation.

The possibility of determining the systolic function of the left ventricle using a single-channel electrocardiogram monitor with photoplethysmography

Abstract Introduction Assessment of the systolic function of the left ventricle is the basis for the timely diagnosis and treatment of various diseases of the cardiovascular system. Regular monitoring of the systolic function is very often necessary. Remote assessment of the left ventricular systolic function can provide more thorough and convenient monitoring of the patient's condition. Purpose The determination of the left ventricular systolic function using a single-channel electrocardiogram (ECG) monitor with photoplethysmography function. Methods The study prospectively included 400 patients aged 18 and over with various pathologies of the cardiovascular system. All study participants underwent echocardiography and the indicators of the left ventricular (LV) systolic function were determined: LV ejection fraction (LV EF), LV outflow tract velocity time integral (LVOT VTI). Recording of the electrocardiogram (I standard lead) and the photoplethysmogram (PPG) using a single-channel ECG monitor with photoplethysmography were performed in all patients. The single-channel electrocardiogram monitor is a portable electrocardiograph with photoplethysmography in the form of a smartphone case. The ECG and PPG parameters were calculated and compared with echocardiography. Models for assessing the LV systolic function were built based on the obtained ECG and PPG parameters using the random forest algorithm. Results A total of 400 ECG records and 400 PPG records were obtained. Then the parameters ECG and PPG were selected for constructing the models. We obtained three models for the ejection fraction (less than 55%, less than 40%, less than 30%) and two models for the LVOT VTI. The quality of each model was assessed by sensitivity, specificity, area under the ROC curve (AUC). For the model with the ejection fraction less than 55%, the AUC was 91.3% (0.844–0.982), sensitivity – 86.4%, specificity – 91.4%. AUC was 95.5% (0.903–1.000), sensitivity 100%, specificity 87.2% for the model with the ejection fraction less than 40%. For the model with the ejection fraction less than 30% AUC: 96.2% (0.900–1.000), sensitivity 100%, specificity 92.9%. Models for the LVOT VTI: less than 13cm (AUC: 77.6% (0.561–0.992), sensitivity 80%, specificity 78.1%) and less than 16cm (AUC: 78.2% (0.657–0.907), sensitivity 76.5%, specificity 77.4%). Conclusions The models obtained based on ECG and PPG parameters had sufficiently high accuracy. Thus, using the ECG and PPG, it is possible to assess the LV systolic function. Therefore, it will be possible to perform the remote monitoring of the systolic function of the left ventricle using portable the single-channel electrocardiogram monitor with photoplethysmography. Funding Acknowledgement Type of funding sources: None.

Analysis and postprocessing of ECG or heart rate data from wearable devices beyond the proprietary cloud and app infrastructure of the vendors

BackgroundThe impact of medical-grade wearable electrocardiographic (ECG) recording technology is increasing rapidly. A wide range of different portable smartphone-connected ECG and heart rate trackers is available on the market. Smart ECG devices are especially valuable to monitor either supraventricular arrhythmias or prolonged QT intervals to avoid drug-induced life-threatening arrhythmias. However, frequent false alarms or false-positive arrhythmia results from wearable devices are unwanted. Therefore, for clinical evaluation, it should be possible to measure and evaluate the biosignals of the wearables independent of the manufacturer.ObjectiveUnlike radiological devices that do support the universal digital imaging and communications in medicine standard, these medical-grade devices do not yet support a secure standardized exchange pathway between sensors, smartphones/smartwatches, and end services such as cloud storage or universal Web-based application programming interface (API) access. Consequently, postprocessing of recorded ECGs or heart rate interval data requires a whole toolbox of customized software technologies.Methods/ResultsVarious methods for measuring and analyzing nonstandardized ECG and heart rate data are proposed, including online measurement of ECG waveforms within a PDF, access to data using manufacturer-specific software development kits, and access to biosignals using modern Web APIs.ConclusionWith the appropriate workaround, modern software technologies such as JavaScript and PHP allow health care providers and researchers to easily and instantly access necessary and important signal measurements on demand.

Body Composition and Heart Rhythm Variability in Elite Wrestlers

The aim of the research was to study the body composition and heart rhythm variability in elite wrestlers. In the study we used: Omron BF511 body composition monitor, we obtained and analyzed data on the body composition of 8 highly qualified athletes, members of the Ukrainian national team in Greco-Roman wrestling in 5 weight cate- gories. 67 kg - 1 person, 72 kg - 1 person, 77 kg - 3 people, 87 kg - 2 people and 97 kg - 1 person. Vegetative regula- tion was assessed according to the indicators of statistical and spectral analysis of heart rhythm variability. For this purpose, we used a portable electrocardiograph with finger electrodes for an integral assessment of the functional state of the cardiovascular system "Fazagraf". The parameters of autonomic regulation and the results of spectral analysis in athletes were recorded. As a result of the study, it was found that the indicators of the body composition of the studied athletes have differences, but most of them correspond to the norm according to the Omron Health- care rating scales. It was revealed that the hemodynamic supply of the body in highly qualified athletes occurs at a higher tension of the mechanisms of regulation of cardiac activity. Studies have shown that a high percentage of muscle mass is consistent with the activation of parasympathetic influences on the sinus node of the heart at rest state, which reflects the result of adaptive restructuring of the body during the training process.

Study of changes in certain cardiorespiratory parameters in response to breath holding in swimmers

Background: Swimming subjects the human body to frequent short intervals of breath hold (BH) and resistive forces of water, increased respiratory effort and thermal stress due to immersion in water. This activity demands higher energy expenditure for movement in a viscous medium. To cope with these, certain adaptations occur in cardiovascular and respiratory systems. Aims and Objectives: Our study compared some cardiovascular and respiratory parameters in male swimmers and sedentary controls at rest and at breakpoint of BH. This helped to know whether respiratory system and cardiovascular system are susceptible to adaptation induced by swimming. Materials and Methods: Thirty regular swimmers and 30 sedentary healthy controls were considered for investigations. A portable electrocardiograph (ECG), finger-tip pulse oximeter, spirometer, and sphygmomanometer were used to record ECG (PR interval, QRS interval, QTc interval, and QRS axis), SpO2, heart rate (HR), tidal volume (TV), and blood pressure. Results: In controls, at breakpoint of BH, oxygen desaturation was extremely significant (“P” < 0.0001), TV was extremely significantly increased (“P” = 0.0006), and only within controls group, HR showed significant increase (“P” = 0.04). No statistically significant change in blood pressure was observed between the groups at breakpoint. A significant increase in TV at breakpoint of BH over that of resting value was observed in controls compared to swimmers (“P” = 0.0134). ECG changes in swimmers and controls were not significantly different. Conclusion: Regular swimming causes oxygen conserving effect, an efficient rise in TV in response to BH.

Remote Atrial Fibrillation Burden Estimation Using Deep Recurrent Neural Network

The atrial fibrillation burden (AFB) is defined as the percentage of time spent in atrial fibrillation (AF) over a long enough monitoring period. Recent research has suggested the added prognostic value of using the AFB compared to a binary diagnosis. We evaluate, for the first time, the ability to estimate the AFB over long-term continuous recordings, using a deep recurrent neutral network (DRNN) approach. The models were developed and evaluated on a large database of p = 2,891 patients, totaling t = 68,800 hours of continuous electrocardiography (ECG) recordings from the University of Virginia. Specifically, 24h beat-to-beat time series were obtained from a single portable ECG channel. The network, denoted ArNet, was benchmarked against a gradient boosting (XGB) model, trained on 21 features including the coefficient of sample entropy (CosEn) and AFEvidence that is derived from the number of irregular points revealed by the Lorenz plot. The generalizations of ArNet and XGB were also evaluated on the independent PhysioNet LTAF test database. the absolute AF burden estimation error [Formula: see text], median and interquartile, on the test set, was 1.2 (0.1-6.7) for ArNet and 2.8 (0.9-11.7) for XGB for AF individuals. Generalization results on LTAF were consistent with [Formula: see text] of 2.7 (1.1-14.7) for ArNet and 3.6 (1.0-16.7) for XGB. This research demonstrates the feasibility of AFB estimation from 24h beat-to-beat interval time series utilizing DRNNs. The novel data-driven approach enables robust remote diagnosis and phenotyping of AF.