No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Atrial Fibrillation Detection in ICU Patients: A Pilot Study on MIMIC III Data *
Abstract
Atrial fibrillation (AF) is the most prevalent arrhythmia, resulting in varying and irregular heartbeats. AF increases risk for numerous cardiovascular diseases including stroke, heart failure and as a result, computer aided efficient monitoring of AF is crucial, especially for intensive care unit (ICU) patients. In this paper, we present an automated and robust algorithm to detect AF from ICU patients using electrocardiogram (ECG) signals. Several statistical parameters including root mean square of successive differences, Shannon entropy, Sample entropy and turning point ratio are calculated from the heart rate. A subset of the Medical Information Mart for Intensive Care (MIMIC) III database containing 36 subjects is used in this study. We compare the AF detection performance of several classifiers for both the training and blinded test data. Using the support vector machine classifier with radial basis kernel, the proposed method achieves 99.95% cross-validation accuracy on the training data and 99.88% sensitivity, 99.65% specificity and 99.75% accuracy on the blinded test data.
... Various methods for detecting AF from ECG signals have been explored, with some focusing on analyzing atrial and ventricular activities and others employing techniques like wavelet transforms and machine learning. Algorithms relying solely on P wave absence often struggle with noise and baseline signal variations, compromising accuracy, particularly in the presence of premature beats [39]. To address these challenges, a novel AF detection algorithm integrating long-term ECG analysis, premature beat detection, and AF identification in critically ill patients is proposed [39]. ...
... Algorithms relying solely on P wave absence often struggle with noise and baseline signal variations, compromising accuracy, particularly in the presence of premature beats [39]. To address these challenges, a novel AF detection algorithm integrating long-term ECG analysis, premature beat detection, and AF identification in critically ill patients is proposed [39]. Procalcitonin (PCT), as a marker of inflammation and tissue injury in bacterial sepsis, aids in diagnosing and prognosticating infections [40]. ...
Atrial fibrillation (AF) is a common arrhythmia in critically ill patients. The objective of this narrative review is to evaluate the characteristics of patients who develop new-onset atrial fibrillation (NOAF) because of sepsis, current management of NOAF in sepsis patients, special consideration in different populations that developed NOAF, health economic and quality of life of patients. We conducted a literature search on PubMed to find research related to NOAF, sepsis and critical illness. Nineteen studies were analyzed for risk factors and outcomes. The incidence rate ranges from 0.53% to 43.9% among these studies. There were numerous risk factors that had been reported from these articles. The most reported risk factors included advanced age, male sex, White race, and cardiovascular comorbidities. The management of septic patients is significantly challenging because of the unfavorable cardiovascular consequences and thromboembolic hazards associated with NOAF. There are comprehensive guidelines available for managing AF, but the effectiveness and safety of therapies in patients with sepsis are still uncertain. Various approaches for managing newly diagnosed AF have been explored. Sinus rhythm can be restored through either pharmacological or non-pharmacological intervention or combination of both. In addition, thromboembolism is a complication that can occur in patients with AF and can have a negative impact on the prognosis of sepsis patients. The use of anticoagulation to prevent stroke after NOAF in sepsis patients is still controversial. Extensive prospective investigations are required to have a deeper understanding of the necessity for anticoagulation following NOAF in sepsis. Beside the treatment of NOAF, early detection of NOAF in sepsis plays a critical role. The prompt initiation of rhythm control medication following a clinical diagnosis of AF can enhance cardiovascular outcomes and reduce mortality in patients with AF and cardiovascular risk factors. Additionally, NOAF in the intensive care unit can prolong hospital stays, increasing hospitalization costs and burdening the hospital. Therefore, preventing and managing NOAF effectively not only benefit the patients but also the hospital in financial aspect. Lastly, to address the existing gaps in knowledge, future research should focus on developing machine learning models that can accurately anticipate risks, establish long-term follow-up protocols, and create complete monitoring systems. The focus is on early intervention and personalized approaches to improve outcomes and quality of life.
... Bashar et al. [120] presented an automated and robust algorithm to detect Atrial Fibrillation using electrocardiogram (ECG) signals from ICU patients. Several statistical parameters were calculated from the heart rate, including root mean square of successive differences, Shannon entropy, Sample entropy, and turning point ratio. ...
... Bashar et al.[120] SVM with the RBF kernel has the best outcome, resulting in 99.88% sensitivity, 99.65% specificity and 99.75% accuracy. ...
The prevalence of cardiovascular diseases is increasing around the world. However, the technology is evolving and can be monitored with low-cost sensors anywhere at any time. This subject is being researched, and different methods can automatically identify these diseases, helping patients and healthcare professionals with the treatments. This paper presents a systematic review of disease identification, classification, and recognition with ECG sensors. The review was focused on studies published between 2017 and 2022 in different scientific databases, including PubMed Central, Springer, Elsevier, Multidisciplinary Digital Publishing Institute (MDPI), IEEE Xplore, and Frontiers. It results in the quantitative and qualitative analysis of 103 scientific papers. The study demonstrated that different datasets are available online with data related to various diseases. Several ML/DP-based models were identified in the research, where Convolutional Neural Network and Support Vector Machine were the most applied algorithms. This review can allow us to identify the techniques that can be used in a system that promotes the patient's autonomy.
... Bashar et al. [120] presented an automated and robust algorithm to detect Atrial Fibrillation using electrocardiogram (ECG) signals from ICU patients. Several statistical parameters were calculated from the heart rate, including root mean square of successive differences, Shannon entropy, Sample entropy, and turning point ratio. ...
... Bashar et al.[120] SVM with the RBF kernel has the best outcome, resulting in 99.88% sensitivity, 99.65% specificity and 99.75% accuracy. ...
The prevalence of cardiovascular diseases is increasing around the world. However, the technology is evolving and can be monitored with low-cost sensors anywhere at any time. This subject is being researched, and different methods can automatically identify these diseases, helping patients and healthcare professionals with the treatments. This paper presents a systematic review of disease identification, classification, and recognition with ECG sensors. The review was focused on studies published between 2017 and 2022 in different scientific databases, including PubMed Central, Springer, Elsevier, Multidisciplinary Digital Publishing Institute (MDPI), IEEE Xplore, and Frontiers. It results in the quantitative and qualitative analysis of 103 scientific papers. The study demonstrated that different datasets are available online with data related to various diseases. Several ML/DP-based models were identified in the research, where Convolutional Neural Network and Support Vector Machine were the most applied algorithms. This review can allow us to identify the techniques that can be used in a system that promotes the patient's autonomy.
... Despite training on a different type of ECG data, the model was highly sensitive and specific, with performance in clean data being better than that in noisy data. Prior studies on AF detection from ICU telemetry data have been developed using continuous ECG telemetry waveform data from the Medical Information Mart for Intensive Care (MIMIC III) data set, and have primarily focused on engineered features derived from waveform characteristics [16,24,25]. In the study by Walkey et al. telemetry data from three cohorts of 50 patients was analyzed for interpretable signal using automated signal and noise detection. ...
... We have used a larger sample size than previous studies and have the added benefit of deploying the algorithm on raw data without any computationally expensive pre-processing. Our prevalence-dependent performance metrics (PPV) are more realistic given the class imbalance in our data (14.1% AF), which is in keeping with estimated prevalence of atrial fibrillation in ICUs, compared to the 50% AF prevalence in test sets in other studies [16,24]. ...
Background
Atrial fibrillation (AF) is the most common cardiac arrhythmia in the intensive care unit and is associated with increased morbidity and mortality. New-onset atrial fibrillation (NOAF) is often initially paroxysmal and fleeting, making it difficult to diagnose, and therefore difficult to understand the true burden of disease. Automated algorithms to detect AF in the ICU have been advocated as a means to better quantify its true burden.
Results
We used a publicly available 12-lead ECG dataset to train a deep learning model for the classification of AF. We then conducted an external independent validation of the model using continuous telemetry data from 984 critically ill patients collected in our institutional database. Performance metrics were stratified by signal quality, classified as either clean or noisy. The deep learning model was able to classify AF with an overall sensitivity of 84%, specificity of 89%, positive predictive value (PPV) of 55%, and negative predictive value of 97%. Performance was improved in clean data as compared to noisy data, most notably with respect to PPV and specificity.
Conclusions
This model demonstrates that computational detection of AF is currently feasible and effective. This approach stands to improve the efficiency of retrospective and prospective research into AF in the ICU by automating AF detection, and enabling precise quantification of overall AF burden.
... In order to solve these issues, we aim to develop, validate, and evaluate a novel AF detection algorithm that incorporates the critical elements of long-term ECG analysis, premature atrial and ventricular beat detection, and AF identification in a large-scale electronic health database from the critically ill. It is to be noted that a preliminary study using this MIMIC III subset is reported by the authors in [33], although only RR interval-based variability was used since subjects with ectopic beats were excluded and fewer subjects were analyzed. ...
... These include the root mean square of successive differences (RM SSD), Shannon entropy (ShaEn) and sample entropy (SampEn). These features were calculated because they are reported to capture the randomness and variation of the HR during AF rhythm, thus, discriminating AF from NSR [15], [33]. Moreover, these features are computationally simple unlike the recent machine learning methods which are computationally taxing [29], [39]. ...
... Then, the data is used as input for a machine learning or heuristic approach to classify the signal. Methods to classify AF can be categorized in three groups: The first group is based on the quantification of the heart rate variability, among them [1][2][3]. These approaches detect the QRS complexes and then the distances among them. ...
... Among the physiological signals recorded in ICUs, ECG is one of the most important vital signs [35]. By analyzing the ECG time series, researchers can not only reveal the respiratory rate, heart rate and variability, but also reduce the false alarm in ICUs [36][37][38]. Thus, ECG provides a good chance for understanding the patient's physiological status. ...
Modern healthcare practice, especially in intensive care units, produces a vast amount of multivariate time series of health-related data, e.g., multi-lead electrocardiogram (ECG), pulse waveform, blood pressure waveform and so on. As a result, timely and accurate prediction of medical intervention (e.g., intravenous injection) becomes possible, by exploring such semantic-rich time series. Existing works mainly focused on onset prediction at the granularity of hours that was not suitable for medication intervention in emergency medicine. This research proposes a Multi-Variable Hybrid Attentive Model (MVHA) to predict the impending need of medical intervention, by jointly mining multiple time series. Specifically, a two-level attention mechanism is designed to capture the pattern of fluctuations and trends of different time series. This work applied MVHA to the prediction of the impending intravenous injection need of critical patients at the intensive care units. Experiments on the MIMIC Waveform Database demonstrated that the proposed model achieves a prediction accuracy of 0.8475 and an ROC-AUC of 0.8318, which significantly outperforms baseline models.
... It includes normal sinus rhythm (NSR), premature atrial/ventricular contractions (PAC/PVCs), atrial fibrillation (AF) and the other signal patterns. The MIMIC database is widely used for evaluating the performance blood pressure estimation [64], [65], and detection of PAC/PVCs [68]- [70] and AF events [14], [66], [67]. The PPG signals contain various types of extrasystolic beats and other pathological patterns, and various signal corruptions and artifacts. ...
Real-time photoplethysmogram (PPG) denoising and data compression has become most essential requirements for accurately measuring vital parameters and efficient data transmission but that may introduce different kinds of waveform distortions due to the lossy processing techniques. Subjective quality assessment tests are the most reliable way to assess the quality, but they are time expensive and also cannot be incorporated with quality-driven compression mechanism. Thus, finding a best objective distortion measure is highly demanded for automatically evaluating quality of reconstructed PPG signal that must be subjectively meaningful and simple. In this paper, we present four types of objective distortion measures and evaluate their performance in terms of quality prediction accuracy, Pearson correlation coefficient and computational time. The performance evaluation is performed on different kinds of PPG waveform distortions introduced by the predictive coding, compressed sampling, discrete cosine transform and discrete wavelet transform. On the normal and abnormal PPG signals taken from five standard databases, evaluation results showed that different subjective quality evaluation groups (5-point, 3-point and 2-point rating scale) had different best objective distortion measures in terms of prediction accuracy and Pearson correlation coefficient. Moreover, selection of a best objective distortion measure depends upon type of PPG features that need to be preserved in the reconstructed signal.
... Among the most common approaches are the use of measurements such as the mean, root mean 978-1-6654-0855-4/21/$31.00 ©2021 IEEE square, turning point ratio, standard deviation and Shannon entropy. These are applied on the vector of R-R differences, and as a result, signals are classified as normal or AF [4,5]. Some authors have used the spectral analysis to quantify the variability of R-R interval, [6,7], and others have developed their own methods combining statistical metrics and machine learning techniques [8,9]. ...
... In [6], authors illustrated and tested a robust automated system to sense atrial fibrillation (AF) from ICU patients using electrocardiogram (ECG) Signals. ...
Revolution of 5G communication system has already jolted an outstanding impact in machine-to-machine communication by strong interconnection and human machine interaction. High-speed data transfer allows sensor-based devices to response quickly and ensures accurate data transfer process. Internet of things (IoT) is making billions of physical tasks easy and real-time operating. Trends in integration of body sensors and sensor-based cyber physical systems have already enabled IoT, interfaced with 5G communication system. One of the most promising sectors for the utilization of IoT technology is in health care and body sensor-based patient monitoring systems. In this research, four vital data focused ICU-Patient Monitoring System has been implemented; using 32-bit SoC and IoT Protocol. Here, multiple body sensors, level detectors and biosensors have been connected for measuring physiological data. These data have been transferred to the central unit. Aiming to the future model, a four vital based prototype has been developed to justify the feasibility. The experimental outcome from the research has shown anticipated response from the prototype. This research will also enhance the efficiency of future medical-vital analysis in low-cost.
... The 2-minute ECG segments without a predominance of noise were then analyzed with a novel R-wave detection method that detects QRS complexes using variable-frequency complex demodulation-based ECG reconstruction [14]. Next, the variability of R-R intervals was evaluated using sample entropy, a measure of randomness that is expected to be higher for patients with AF than those with normal sinus rhythm [15]. Based on the sample entropy calculated from the R-R intervals, an automated "initial screening" for AF was performed, where the "possible AF" status may include premature atrial and ventricular contraction segments as false-positive detections of AF. ...
Background
Atrial fibrillation (AF) is the most common arrhythmia during critical illness, representing a sepsis-defining cardiac dysfunction associated with adverse outcomes. Large burdens of premature beats and noisy signal during sepsis may pose unique challenges to automated AF detection.
Objective
The objective of this study is to develop and validate an automated algorithm to accurately identify AF within electronic health care data among critically ill patients with sepsis.
Methods
This is a retrospective cohort study of patients hospitalized with sepsis identified from Medical Information Mart for Intensive Care (MIMIC III) electronic health data with linked electrocardiographic (ECG) telemetry waveforms. Within 3 separate cohorts of 50 patients, we iteratively developed and validated an automated algorithm that identifies ECG signals, removes noise, and identifies irregular rhythm and premature beats in order to identify AF. We compared the automated algorithm to current methods of AF identification in large databases, including ICD-9 (International Classification of Diseases, 9th edition) codes and hourly nurse annotation of heart rhythm. Methods of AF identification were tested against gold-standard manual ECG review.
Results
AF detection algorithms that did not differentiate AF from premature atrial and ventricular beats performed modestly, with 76% (95% CI 61%-87%) accuracy. Performance improved (P=.02) with the addition of premature beat detection (validation set accuracy: 94% [95% CI 83%-99%]). Median time between automated and manual detection of AF onset was 30 minutes (25th-75th percentile 0-208 minutes). The accuracy of ICD-9 codes (68%; P=.002 vs automated algorithm) and nurse charting (80%; P=.02 vs algorithm) was lower than that of the automated algorithm.
Conclusions
An automated algorithm using telemetry ECG data can feasibly and accurately detect AF among critically ill patients with sepsis, and represents an improvement in AF detection within large databases.
... These algorithms are based on either variability of RR intervals [7, 8,9] or detection of the absence of a p-wave in the ECG [10,11]. However, with advancements in miniaturization of device technologies, wearable health monitoring is gaining attraction nowadays. ...
Detection of atrial fibrillation (AF) from a wrist watch photoplethysmography (PPG) signal is important because the wrist watch form factor enables long term continuous monitoring of arrhythmia in an easy and non-invasive manner. We have developed a novel method not only to detect AF from a smart wrist watch PPG signal, but also to determine whether the recorded PPG signal is corrupted by motion artifacts or not. In order to detect the motion and noise artifacts from PPG signal, we have used both the accelerometer signal and variable frequency complex demodulation based time-frequency analysis of the PPG signal. After the clean PPG signals are determined, we use the root mean square of successive differences and sample entropy, calculated from the beat-to-beat intervals of the PPG signal, to distinguish AF from normal rhythm. Next, we use a premature atrial and ventricular contraction detection algorithm to have more accurate AF identification and to reduce false alarms. Two separate data sets have been used in this study to test the efficacy of our proposed method, which shows a combined sensitivity, specificity and accuracy of 98.18%, 97.43% and 97.54% across the data sets.
Contemporary methods used to interpret the electrocardiogram (ECG) signal for diagnosis or monitoring are based on expert knowledge and rule-centered algorithms. In recent years, with the advancement of artificial intelligence, more and more researchers are using deep learning (ML) and deep learning (DL) with ECG data to detect different types of cardiac issues as well as other health problems such as respiration rate, sleep apnea, and blood pressure, etc. This study presents an extensive literature review based on research performed in the last few years where ML and DL have been applied with ECG data for many diagnoses. However, the review found that, in published work, the results showed promise. However, some significant limitations kept that technique from implementation in reality and being used for medical decisions; examples of such limitations are imbalanced and the absence of standardized dataset for evaluation, lack of interpretability of the model, inconsistency of performance while using a new dataset, security, and privacy of health data and lack of collaboration with physicians, etc. AI using ECG data accompanied by modern wearable biosensor technologies has the potential to allow for health monitoring and early diagnosis within reach of larger populations. However, researchers should focus on resolving the limitations.
The most prevalent kind of heart disease, Atrial Fibrillation (AF), is said to increase the risk of stroke, heart failure, and other health concerns. Clinical observation of the electrocardiogram (ECG) waveform needs an experienced person to observe and takes long hours. We provide an approach to identify AF in the MIT-BIH Arrhythmia and AF databases in this study. Several parameters, including QRS complex, RR Interval, heart rate, coefficient of variance (CV), normalized root mean square of successive difference (nRMSSD) and peak frequency are calculated from the features of the ECG signal. With holdout validation, we analyse the AF classification performance of various classifiers. With the input parameters mentioned, the greatest outcome in AF classification is obtained by the weighted KNN classifier using the DWT algorithm and modified windowing algorithm in holdout validation, with sensitivity, specificity, and accuracy of 90%, 100%, and 92.31%, accordingly. On this basis, it is recommended that classifying ECG signals using machine learning methods will assist in improving the research’s accuracy. Further research is needed to test the proposed algorithm on a large database for better accuracy. This algorithm will be involved in hardware and implemented in the detection of real-time AF ECG patients.
Atrial fibrillation (AF) is the most common arrhythmia found in the intensive care unit (ICU), and is associated with many adverse outcomes. Effective handling of AF and similar arrhythmias is a vital part of modern critical care, but obtaining knowledge about both disease burden and effective interventions often requires costly clinical trials. A wealth of continuous, high frequency physiological data such as the waveforms derived from electrocardiogram telemetry are promising sources for enriching clinical research. Automated detection using machine learning and in particular deep learning has been explored as a solution for processing these data. However, a lack of labels, increased presence of noise, and inability to assess the quality and trustworthiness of many machine learning model predictions pose challenges to interpretation. In this work, we propose an approach for training deep AF models on limited, noisy data and report uncertainty in their predictions. Using techniques from the fields of weakly supervised learning, we leverage a surrogate model trained on non-ICU data to create imperfect labels for a large ICU telemetry dataset. We combine these weak labels with techniques to estimate model uncertainty without the need for extensive human data annotation. AF detection models trained using this process demonstrated higher classification performance (0.64–0.67 F1 score) and improved calibration (0.05–0.07 expected calibration error).
Atrial fibrillation (AF) is the most sustained arrhythmia in the heart and also the most common complication developed after cardiac surgery. Due to its progressive nature, timely detection of AF is important. Currently, physicians use a surface electrocardiogram (ECG) for AF diagnosis. However, when the patient develops AF, its various development stages are not distinguishable for cardiologists based on visual inspection of the surface ECG signals. Therefore, severity detection of AF could start from differentiating between short-lasting AF and long-lasting AF. Here, de novo post-operative AF (POAF) is a good model for short-lasting AF while long-lasting AF can be represented by persistent AF. Therefore, we address in this paper a binary severity detection of AF for two specific types of AF. We focus on the differentiation of these two types as de novo POAF is the first time that a patient develops AF. Hence, comparing its development to a more severe stage of AF (e.g., persistent AF) could be beneficial in unveiling the electrical changes in the atrium. To the best of our knowledge, this is the first paper that aims to differentiate these different AF stages. We propose a method that consists of three sets of discriminative features based on fundamentally different aspects of the multi-channel ECG data, namely based on the analysis of RR intervals, a greyscale image representation of the vectorcardiogram, and the frequency domain representation of the ECG. Due to the nature of AF, these features are able to capture both morphological and rhythmic changes in the ECGs. Our classification system consists of a random forest classifier, after a feature selection stage using the ReliefF method. The detection efficiency is tested on 151 patients using 5-fold cross-validation. We achieved 89.07% accuracy in the classification of de novo POAF and persistent AF. The results show that the features are discriminative to reveal the severity of AF. Moreover, inspection of the most important features sheds light on the different characteristics of de novo post-operative and persistent AF.
We have developed a novel method to accurately detect QRS complex peaks using the variable frequency complex demodulation (VFCDM) method. The approach’s novelty stems from reconstructing an ECG signal using only the frequency components associated with the QRS waveforms by VFCDM decomposition. After signal reconstruction, both top and bottom sides of the signal are used for peak detection, after which we compare locations of the peaks detected from both sides to ensure false peaks are minimized. Finally, we impose position-dependent adaptive thresholds to remove any remaining false peaks from the prior step. We applied the proposed method to the widely benchmarked MIT-BIH arrhythmia dataset, and obtained among the best results compared to many of the recently published methods. Our approach resulted in 99.94% sensitivity, 99.95% positive predictive value and a 0.11% detection error rate. Three other datasets—the MIMIC III database, University of Massachusetts atrial fibrillation data, and SCUBA diving in salt water ECG data—were used to further test the robustness of our proposed algorithm. For all these three datasets, our method retained consistently higher accuracy when compared to the BioSig Matlab toolbox, which is publicly available and known to be reliable for ECG peak detection.
Linear discriminant analysis (LDA) is a very popular supervised feature extraction method and has been extended to different variants. However, classical LDA has the following problems: 1) The obtained discriminant projection does not have good interpretability for features. 2) LDA is sensitive to noise. 3) LDA is sensitive to the selection of number of projection directions. In this paper, a novel feature extraction method called robust sparse linear discriminant analysis (RSLDA) is proposed to solve the above problems. Specifically, RSLDA adaptively selects the most discriminative features for discriminant analysis by introducing the l2;1 norm. An orthogonal matrix and a sparse matrix are also simultaneously introduced to guarantee that the extracted features can hold the main energy of the original data and enhance the robustness to noise, and thus RSLDA has the potential to perform better than other discriminant methods. Extensive experiments on six databases demonstrate that the proposed method achieves the competitive performance compared with other state-of-the-art feature extraction methods. Moreover, the proposed method is robust to the noisy data.
We present a smartphone-only solution for the detection of Atrial Fibrillation (AFib), which utilizes the builtin accelerometer and gyroscope sensors (Inertial Measurement Unit, IMU) in the detection. Depending on the patient's situation, it is possible to use the developed smartphone application either regularly or occasionally for making a measurement of the subject. The smartphone is placed on the chest of the patient who is adviced to lay down and perform a non-invasive recording, while no external sensors are needed. After that, the application determines whether the patient suffers from AFib or not. The presented method has high potential to detect paroxysmal ("silent") AFib from large masses. In this paper, we present the pre-processing, feature extraction, feature analysis and classification results of the envisioned AFib detection system based on clinical data acquired with a standard mobile phone equipped with Google Android OS. Test data was gathered from 16 AFib patients (validated against ECG), as well as a control group of 23 healthy individuals with no diagnosed heart diseases. We obtained an accuracy of 97.4% in AFib vs. healthy classification (a sensitivity of 93.8% and a specificity of 100%). Due to the wide availability of smart devices/sensors with embedded IMU, the proposed methods could potentially also scale to other domains such as embedded body-sensor networks.
Background:
The World Heart Federation has undertaken an initiative to develop a series of Roadmaps to promote development of national policies and health systems approaches, and to identify potential roadblocks on the road to effective prevention, detection, and management of cardiovascular disease in low-and middle-income countries (LMICs) and develop strategies for overcoming these. This Roadmap focuses on atrial fibrillation (AF). AF is the most common, clinically significant arrhythmia and, among other clinical outcomes, is associated with increased risk of stroke.
Methods:
Development of this Roadmap included a review of published guidelines and research papers, and consultation with an expert committee comprising experts in clinical management of AF and health systems research in LMICs. The Roadmap identifies 1) key interventions for detection, diagnosis, and management of AF; 2) gaps in implementation of these interventions (knowledge-practice gaps); 3) health system roadblocks to implementation of AF interventions in LMICs; and 4) potential strategies for overcoming these.
Results:
More research is needed on determinants and primary prevention of AF. Knowledge-practice gaps for detection, diagnosis, and management of AF are present worldwide, but may be more prominent in LMICs. Potential barriers to implementation of AF interventions include long distances to health facilities, shortage of health care professionals with training in AF, including interpretation of ECG, unaffordability of oral anticoagulants for patient households, reluctance on the part of physicians to initiate oral anticoagulant (OAC) therapy, and lack of awareness of the importance of persistent adherence to OAC therapy. Potential solutions include training of nonphysician health workers and pharmacists in pulse-taking, use of telemedicine technologies to transmit electrocardiogram results, engagement of nonphysician health workers in OAC therapy adherence support, and country-specific support and education programs for noncardiologist health care professionals.
Conclusions:
AF affects millions of people worldwide and, left untreated, increases the risk and severity of stroke and heart failure. Although guidelines for the detection, diagnosis, and management of AF exist, there are gaps in implementation of these guidelines globally, and in particular in LMICs. This Roadmap identifies some potential solutions that may improve AF outcomes in LMICs but require further evaluation in these settings.
Aims
Atrial fibrillation (AF) is the most common arrhythmia encountered in clinical practice, and its paroxysmal nature makes its detection challenging. In this trial, we evaluated a novel App for its accuracy to differentiate between patients in AF and patients in sinus rhythm (SR) using the plethysmographic sensor of an iPhone 4S and the integrated LED only.
Methods and results
For signal acquisition, we used an iPhone 4S, positioned with the camera lens and LED light on the index fingertip. A 5 min video file was recorded with the pulse wave extracted from the green light spectrum of the signal. RR intervals were automatically identified. For discrimination between AF and SR, we tested three different statistical methods. Normalized root mean square of successive difference of RR intervals (nRMSSD), Shannon entropy (ShE), and SD1/SD2 index extracted from a Poincaré plot. Eighty patients were included in the study (40 patients in AF and 40 patients in SR at the time of examination). For discrimination between AF and SR, ShE yielded the highest sensitivity and specificity with 85 and 95%, respectively. Applying a tachogram filter resulted in an improved sensitivity of 87.5%, when combining ShE and nRMSSD, while specificity remained stable at 95%. A combination of SD1/SD2 index and nRMSSD led to further improvement and resulted in a sensitivity and specificity of 95%.
Conclusion
The algorithm tested reliably discriminated between SR and AF based on pulse wave signals from a smartphone camera only. Implementation of this algorithm into a smartwatch is the next logical step.
MIMIC-III (‘Medical Information Mart for Intensive Care’) is a large, single-center database comprising information relating to patients admitted to critical care units at a large tertiary care hospital. Data includes vital signs, medications, laboratory measurements, observations and notes charted by care providers, fluid balance, procedure codes, diagnostic codes, imaging reports, hospital length of stay, survival data, and more. The database supports applications including academic and industrial research, quality improvement initiatives, and higher education coursework.
Atrial fibrillation (AF) is the most common cardiac arrhythmia, and a major public health burden associated with significant morbidity and mortality. Automatic detection of AF could substantially help in early diagnosis, management and consequently prevention of the complications associated with chronic AF. In this paper, we propose a novel method for automatic AF detection.
Stationary wavelet transform and support vector machine have been employed to detect AF episodes. The proposed method eliminates the need for P-peak or R-Peak detection (a pre-processing step required by many existing algorithms), and hence its performance (sensitivity, specificity) does not depend on the performance of beat detection. The proposed method has been compared with those of the existing methods in terms of various measures including performance, transition time (detection delay associated with transitioning from a non-AF to AF episode), and computation time (using MIT-BIH Atrial Fibrillation database).
Results of a stratified 2-fold cross-validation reveals that the area under the Receiver Operative Characteristics (ROC) curve of the proposed method is 99.5%. Moreover, the method maintains its high accuracy regardless of the choice of the parameters׳ values and even for data segments as short as 10s. Using the optimal values of the parameters, the method achieves sensitivity and specificity of 97.0% and 97.1%, respectively.
The proposed AF detection method has high sensitivity and specificity, and holds several interesting properties which make it a suitable choice for practical applications.
Copyright © 2015 Elsevier Ltd. All rights reserved.
Atrial fibrillation is the most common sustained arrhythmia in clinical practice worldwide. Several algorithms have been developed to detect atrial fibrillation, which either rely on atrial activity analysis or are based on the irregularity of RR intervals. This paper is addressed to study the latter type of algorithms. The main question is whether there is sufficient information in the sequence of RR intervals for reliable detection of atrial fibrillation and whether the atrial fibrillation can be differentiated from other significant ECG arrhythmias.
We have tested various types of algorithms existing in the technical papers utilizing MIT-BIH databases.
Except the atrial fibrillation all other arrhythmias have some regularity, self-similarity and some degree of predictability. Consequently, algorithms utilizing only the values of RR intervals without their order may misclassify other irregular rhythms as atrial fibrillation. The best algorithms use the scatter plot of successive RR differences or Sample Entropy. The error rate was about 5 %.
It is possible to create a robust atrial fibrillation detection algorithm relying only on RR intervals considering their places in the sequence.
Atrial fibrillation (AF) is the most common and debilitating abnormalities of the arrhythmias worldwide, with a major impact on morbidity and mortality. The detection of AF becomes crucial in preventing both acute and chronic cardiac rhythm disorders.
Our objective is to devise a method for real-time, automated detection of AF episodes in electrocardiograms (ECGs). This method utilizes RR intervals, and it involves several basic operations of nonlinear/linear integer filters, symbolic dynamics and the calculation of Shannon entropy. Using novel recursive algorithms, online analytical processing of this method can be achieved.
Four publicly-accessible sets of clinical data (Long-Term AF, MIT-BIH AF, MIT-BIH Arrhythmia, and MIT-BIH Normal Sinus Rhythm Databases) were selected for investigation. The first database is used as a training set; in accordance with the receiver operating characteristic (ROC) curve, the best performance using this method was achieved at the discrimination threshold of 0.353: the sensitivity (Se), specificity (Sp), positive predictive value (PPV) and overall accuracy (ACC) were 96.72%, 95.07%, 96.61% and 96.05%, respectively. The other three databases are used as testing sets. Using the obtained threshold value (i.e., 0.353), for the second set, the obtained parameters were 96.89%, 98.25%, 97.62% and 97.67%, respectively; for the third database, these parameters were 97.33%, 90.78%, 55.29% and 91.46%, respectively; finally, for the fourth set, the Sp was 98.28%. The existing methods were also employed for comparison.
Overall, in contrast to the other available techniques, the test results indicate that the newly developed approach outperforms traditional methods using these databases under assessed various experimental situations, and suggest our technique could be of practical use for clinicians in the future.
The application of non-linear metrics to physiological signals is a valuable tool because “hidden information” related to underlying mechanisms can be obtained. In this respect, approximate entropy (ApEn) is the most popular non-linear regularity index that has been applied to physiological time series. However, ApEn presents some shortcomings, such as bias, relative inconsistency and dependence on the sample length. A modification of ApEn, named sample entropy (SampEn), was introduced to overcome these deficiencies. Recently, in the context of electrocardiography, SampEn has been applied to study non-invasively atrial fibrillation (AF), which is the most common arrhythmia encountered in clinical practice with unknown mechanisms provoking its onset and termination. Useful clinical information, that could help for a better understanding of AF mechanisms, has been obtained through the application of SampEn to electrocardiographic (ECG) recordings. This work reviews its application in the context of non-invasive analysis of AF. During this arrhythmia, atrial and ventricular components can be regarded as unsynchronized activities, whereby, the application of SampEn to the analysis of each component will be described separately. In first place, clinical challenges in which SampEn has been successfully applied to estimate AF organization from the atrial activity pattern are presented. The AF organization study can provide information on the number of active reentries, which can help to improve AF treatment and to take the appropriate decisions on its management. Next, the heart rate variability study via SampEn, to characterize ventricular response and predict AF onset, is described. Through the aforementioned applications it is remarked throughout this review that SampEn can be considered as a very promising and useful tool towards the non-invasive understanding of AF.
Atrial fibrillation (AF) is the most common sustained arrhythmia and is associated with significant morbidity and mortality. Timely diagnosis of the arrhythmia, particularly transient episodes, can be difficult since patients may be asymptomatic. In this study, we describe a robust algorithm for automatic detection of AF based on the randomness, variability and complexity of the heart beat interval (RR) time series. Specifically, we employ a new statistic, the Turning Points Ratio, in combination with the Root Mean Square of Successive RR Differences and Shannon Entropy to characterize this arrhythmia. The detection algorithm was tested on two databases, namely the MIT-BIH Atrial Fibrillation Database and the MIT-BIH Arrhythmia Database. These databases contain several long RR interval series from a multitude of patients with and without AF and some of the data contain various forms of ectopic beats. Using thresholds and data segment lengths determined by Receiver Operating Characteristic (ROC) curves we achieved a high sensitivity and specificity (94.4% and 95.1%, respectively, for the MIT-BIH Atrial Fibrillation Database). The algorithm performed well even when tested against AF mixed with several other potentially confounding arrhythmias in the MIT-BIH Arrhythmia Database (Sensitivity = 90.2%, Specificity = 91.2%).
Nonrheumatic atrial fibrillation is common among the elderly and is associated with an increased risk of stroke. We investigated whether anticoagulation with warfarin would reduce this risk.
We conducted a randomized, double-blind, placebo-controlled study to evaluate low-intensity anticoagulation with warfarin (prothrombin-time ratio, 1.2 to 1.5) in 571 men with chronic nonrheumatic atrial fibrillation; 525 patients had not previously had a cerebral infarction, whereas 46 patients had previously had such an event. The primary end point was cerebral infarction; secondary end points were cerebral hemorrhage and death.
Among the patients with no history of stroke, cerebral infarction occurred in 19 of the 265 patients in the placebo group during an average follow-up of 1.7 years (4.3 percent per year) and in 4 of the 260 patients in the warfarin group during an average follow-up of 1.8 years (0.9 percent per year). The reduction in risk with warfarin therapy was 0.79 (95 percent confidence interval, 0.52 to 0.90; P = 0.001). The annual event rate among the 228 patients over 70 years of age was 4.8 percent in the placebo group and 0.9 percent in the warfarin group (risk reduction, 0.79; P = 0.02). The only cerebral hemorrhage occurred in a 73-year-old patient in the warfarin group. Other major hemorrhages, all gastrointestinal, occurred in 10 patients: 4 in the placebo group, for a rate of 0.9 percent per year, and 6 in the warfarin group, for a rate of 1.3 percent per year. There were 37 deaths that were not preceded by a cerebral end point--22 in the placebo group and 15 in the warfarin group (risk reduction, 0.31; P = 0.19). Cerebral infarction was more common among patients with a history of cerebral infarction (9.3 percent per year in the placebo group and 6.1 percent per year in the warfarin group) than among those without such a history.
Low-intensity anticoagulation with warfarin prevented cerebral infarction in patients with nonrheumatic atrial fibrillation without producing an excess risk of major hemorrhage. This benefit extended to patients over 70 years of age.
The newly inaugurated Research Resource for Complex Physiologic Signals, which was created under the auspices of the National Center for Research Resources of the National Institutes of Health, is intended to stimulate current research and new investigations in the study of cardiovascular and other complex biomedical signals. The resource has 3 interdependent components. PhysioBank is a large and growing archive of well-characterized digital recordings of physiological signals and related data for use by the biomedical research community. It currently includes databases of multiparameter cardiopulmonary, neural, and other biomedical signals from healthy subjects and from patients with a variety of conditions with major public health implications, including life-threatening arrhythmias, congestive heart failure, sleep apnea, neurological disorders, and aging. PhysioToolkit is a library of open-source software for physiological signal processing and analysis, the detection of physiologically significant events using both classic techniques and novel methods based on statistical physics and nonlinear dynamics, the interactive display and characterization of signals, the creation of new databases, the simulation of physiological and other signals, the quantitative evaluation and comparison of analysis methods, and the analysis of nonstationary processes. PhysioNet is an on-line forum for the dissemination and exchange of recorded biomedical signals and open-source software for analyzing them. It provides facilities for the cooperative analysis of data and the evaluation of proposed new algorithms. In addition to providing free electronic access to PhysioBank data and PhysioToolkit software via the World Wide Web (http://www.physionet. org), PhysioNet offers services and training via on-line tutorials to assist users with varying levels of expertise.
A method is presented for automatic analysis of the P-wave, based on lead II of a 12-lead standard ECG, in resting conditions during a routine examination for the detection of patients prone to atrial fibrillation (AF), one of the most prevalent arrhythmias. First, the P-wave was delineated, and this was achieved in two steps: the detection of the QRS complexes for ECG segmentation, using a wavelet analysis method, and a hidden Markov model to represent one beat of the signal for P-wave isolation. Then, a set of parameters to detect patients prone to AF was calculated from the P-wave. The detection efficiency was validated on an ECG database of 145 patients, including a control group of 63 people and a study group of 82 patients with documented AF. A discriminant analysis was applied, and the results obtained showed a specificity and a sensitivity between 65% and 70%.
Automatic detection of atrial activity (P waves) in an electrocardiogram (ECG) is a crucial task to diagnose the presence of arrhythmias. The P wave is difficult to detect and most of the approaches in the literature have been evaluated on normal sinus rhythms and rarely considered arrhythmia contexts other than atrial flutter and fibrillation. A novel knowledge-based P wave detector algorithm is presented. It is self-adaptive to the patient and able to deal with certain arrhythmias by tracking the PP rhythm. The detector has been tested on 12 records of the MIT-BIH arrhythmia database containing several ventricular and supra-ventricular arrhythmias. On the overall records, the detector demonstrates Se = 96.60% and Pr = 95.46%; for the normal sinus rhythm, it reaches Se = 97.76% and Pr = 96.80% and, in the case of Mobitz type II, it demonstrates Se = 72.79% and Pr = 99.51%. It also shows good performance for trigeminy and bigeminy, and outperforms some more sophisticated techniques. Although the results emphasize the difficulty of P wave detection in difficult arrhythmias (supra and ventricular tachycardias), it shows that domain knowledge can efficiently support signal processing techniques.
Objective:
Atrial fibrillation (AF) and other types of abnormal heart rhythm are related to multiple fatal cardiovascular diseases that affect the quality of human life. Hence the development of an automated robust method that can reliably detect AF, in addition to other non-sinus and sinus rhythms, would be a valuable addition to medicine. The present study focuses on developing an algorithm for the classification of short, single-lead electrocardiogram (ECG) recordings into normal, AF, other abnormal rhythms and noisy classes.
Approach:
The proposed classification framework presents a two-layer, three-node architecture comprising binary classifiers. PQRST markers are detected on each ECG recording, followed by noise removal using a spectrogram power based novel adaptive thresholding scheme. Next, a feature pool comprising time, frequency, morphological and statistical domain ECG features is extracted for the classification task. At each node of the classification framework, suitable feature subsets, identified through feature ranking and dimension reduction, are selected for use. Adaptive boosting is selected as the classifier for the present case. The training data comprises 8528 ECG recordings provided under the PhysioNet 2017 Challenge. F1 scores averaged across the three non-noisy classes are taken as the performance metric.
Main result:
The final five-fold cross-validation score achieved by the proposed framework on the training data has high accuracy with low variance (0.8254 [Formula: see text] 0.0043).
Significance:
Further, the proposed algorithm has achieved joint first place in the PhysioNet/Computing in Cardiology Challenge 2017 with a score of 0.83 computed on a hidden test dataset.
The global burden of atrial fibrillation (AF) is unknown.
We systematically reviewed population-based studies of AF published 1980-2010, from the 21 Global Burden of Disease (GBD) regions to estimate global/regional prevalence, incidence, as well as morbidity and mortality related to AF (DisModMR software). Of 377 potential studies identified, 184 met pre-specified eligibility criteria. The estimated number of individuals with AF globally in 2010 was 33•5 million [(20•9 million males (UI, 19•5-22•2 million) and 12•6 million females (UI, 12•0-13•7 million)]. Burden associated with AF, measured as disability adjusted life-years (DALYs), increased by 18•8% (UI, 15•8-19•3) in males and 18•9% (UI, 15•8-23•5) in females, from 1990 to 2010. In 1990, the estimated age-adjusted prevalence rates of AF (per 100,000 population) were 569•5 in males [95% uncertainty interval (UI), 532•8-612•7] and 359•9 in females (UI, 334•7-392•6); the estimated age-adjusted incidence rates were 60•7/100,000 person-years in males (UI, 49•2-78•5) and 43•8 in females (UI, 35•9-55•0). In 2010 the prevalence rate increased to 596•2 (UI, 558•4-636•7) in males and 373•1 (UI, 347•9-402•2) in females; incidence rate increased to 77•5 (UI, 65•2-95•4) in males and 59•5 (UI, 49•9-74•9) in females. Mortality associated with AF was higher in females, and increased by 2-fold (UI, 2•0-2•2) and 1•9-fold (UI, 1•8-2•0) in males and females, respectively, from 1990 to 2010.
These findings provide evidence of progressive increases in overall burden, incidence, prevalence and AF-associated mortality between 1990-2010. Systematic, global surveillance of AF is required to better direct prevention and treatment strategies.
Automatic detection of atrial fibrillation (AF) in electrocardiograms (ECGs) is beneficial for AF diagnosis, therapy, and management. In this article, a novel method of AF detection is introduced. Most current methods only utilize the RR interval as a critical parameter to detect AF; thus, these methods commonly confuse AF with other arrhythmias.
We used the average number of f waves in a TQ interval as a characteristic parameter in our robust, real-time AF detection method. Three types of clinical ECG data, including ECGs from normal, AF, and non-AF arrhythmia subjects, were downloaded from multiple open access databases to validate the proposed method.
The experimental results suggested that the method could distinguish between AF and normal ECGs with accuracy, sensitivity, and positive predictive values (PPVs) of 93.67%, 94.13%, and 98.69%, respectively. These values are comparable to those of related methods. The method was also able to distinguish between AF and non-AF arrhythmias and had performance indexes (accuracy 94.62%, sensitivity 94.13%, and PPVs 97.67%) that were considerably better than those of other methods.
Our proposed method has prospects as a practical tool enabling clinical diagnosis, treatment, and monitoring of AF.
In this paper, we propose a novel independent components (ICs) arrangement strategy to cooperate with the independent component analysis (ICA) method used for ECG beat classification. The ICs calculated with a regular ICA algorithm are re-arranged according to the L2 norms of the rows of the de-mixing matrix. The validity of this ICs arrangement strategy is discussed mathematically and testified experimentally. Only when the ICs are arranged in an appropriate order, we are able to select the first couple of components for the calculation of the most significant subset of bases to discriminate different types of ECG beats. The effectiveness and efficiency of the proposed method and three other ICs arrangement strategies are studied. Two kinds of classifiers, including probabilistic neural network and support vector machines, are used to evaluate the proposed method. ECG samples attributing to eight different ECG beat types were extracted from the MIT-BIH arrhythmia database for the study. The experiment results demonstrate that the proposed ICs arrangement strategy outperforms the other strategies in reducing the number of features required for the classifiers. Among the many experimental setups, the scheme with the SVM classifier in conjugate with the log(cosh(·)) contrast function and the proposed ICs arrangement strategy requires only 17 ICs to achieve more than 98.7% classification accuracy and is determined to be most efficient. When comparing to the other methods in the literature, the proposed scheme outperforms the other methods in terms of effectiveness and efficiency. The impressive result validates the strategy in the selection of significant ICs subset and demonstrates the proposed scheme an effective and efficient approach in computer-aided diagnosis of heart diseases based on ECG signals.
In this study, a new supervised noise-artifact-robust heart arrhythmia fusion classification solution, is introduced. Proposed method consists of structurally diverse classifiers with a new QRS complex geometrical feature extraction technique.Toward this objective, first, the events of the electrocardiogram (ECG) signal are detected and delineated using a robust wavelet-based algorithm. Then, each QRS region and also its corresponding discrete wavelet transform (DWT) are supposed as virtual images and each of them is divided into eight polar sectors. Next, the curve length of each excerpted segment is calculated and is used as the element of the feature space. Discrimination power of proposed classifier in isolation of different Gold standard beats was assessed with accuracy 98.20%. Also, proposed learning machine was applied to 7 arrhythmias belonging to 15 different records and accuracy 98.06% was achieved. Comparisons with peer-reviewed studies prove a marginal progress in computerized heart arrhythmia recognition technologies.
Permanent and paroxysmal AF is a risk factor for the occurrence and the recurrence of stroke, which can occur as its first manifestation. However, its automatic identification is still unsatisfactory. In this study, a new mathematical approach was evaluated to automate AF identification. A derivation set of 30 24-hour Holter recordings, 15 with chronic AF (CAF) and 15 with sinus rhythm (SR), allowed the authors to establish specific RR variability characteristics using wavelet and fractal analysis. Then, a validation set of 50 subjects was studied using these criteria, 19 with CAF, 16 with SR, and 15 with paroxysmal AF (PAF); and each QRS was classified as true or false sinus or AF beat. In the SR group, specificity reached 99.9%; in the CAF group, sensitivity reached 99.2%; in the PAF group, sensitivity reached 96.1%, and specificity 92.6%. However, classification on a patient basis provided a sensitivity of 100%. This new approach showed a high sensitivity and a high specificity for automatic AF detection, and could be used in screening for AF in large populations at risk.
This work describes a method for automatic detection of atrial
fibrillation (AF) based on RR intervals. We define ΔRR to be the
difference between successive RR intervals. The standard density
histograms of RR and ΔRR intervals are determined from data in the
MIT-BIH atrial fibrillation/flutter database. The present method
estimates the similarity between the standard density histograms and a
best density histogram by the Kolmogorov-Smirnov (KS) test. The
algorithm returns significance (p) of difference between given
histograms. If the p value is smaller than a value (Pc), the
test density histogram is significantly different from the standard
density histogram. If the test density histogram is not significantly
different from the standard density histogram, we say the data is AF:
Using the standard density histogram of ΔRR with
Pc=0.01, the average sensitivity is 93.2% and the average
specificity is 96.7%