Prediction of paroxysmal Atrial Fibrillation: A machine learning based approach using combined feature vector and mixture of expert classification on HRV signal

Ethical Dilemmas of Using Artificial Intelligence in Medicine

BACKGROUND

Artificial intelligence (AI) is considered the fourth industrial revolution that will change the evolution of humanity technically and relationally. Although the term has been around since 1956, it has only recently become apparent that AI can revolutionize technologies and has many applications in the medical field.

AREAS OF UNCERTAINTY

The ethical dilemmas posed by the use of AI in medicine revolve around issues related to informed consent, respect for confidentiality, protection of personal data, and last but not least the accuracy of the information it uses.

DATA SOURCES

A literature search was conducted through PubMed, MEDLINE, Plus, Scopus, and Web of Science (2015-2022) using combinations of keywords, including: AI, future in medicine, and machine learning plus ethical dilemma.

ETHICS AND THERAPEUTIC ADVANCES

The ethical analysis of the issues raised by AI used in medicine must mainly address nonmaleficence and beneficence, both in correlation with patient safety risks, ability versus inability to detect correct information from inadequate or even incorrect information. The development of AI tools that can support medical practice can increase people's access to medical information, to obtain a second opinion, for example, but it is also a source of concern among health care professionals and especially bioethicists about how confidentiality is maintained and how to maintain cybersecurity. Another major risk may be related to the dehumanization of the medical act, given that, at least for now, empathy and compassion are accessible only to human beings.

CONCLUSIONS

AI has not yet managed to overcome certain limits, lacking moral subjectivity, empathy, the level of critical thinking is still insufficient, but no matter who will practice preventive or curative medicine in the next period, they will not be able to ignore AI, which under human control can be an important tool in medical practice.

Early warning of atrial fibrillation using deep learning

Atrial fibrillation (AF), the most prevalent cardiac rhythm disorder, significantly increases hospitalization and health risks. Reverting from AF to sinus rhythm (SR) often requires intensive interventions. This study presents a deep-learning model capable of predicting the transition from SR to AF on average 30.8 min before the onset appears, with an accuracy of 83% and an F1 score of 85% on the test data. This performance was obtained from R-to-R interval signals, which can be accessible from wearable technology. Our model, entitled Warning of Atrial Fibrillation (WARN), consists of a deep convolutional neural network trained and validated on 24-h Holter electrocardiogram data from 280 patients, with 70 additional patients used for testing and further evaluation on 33 patients from two external centers. The low computational cost of WARN makes it ideal for integration into wearable technology, allowing for continuous heart monitoring and early AF detection, which can potentially reduce emergency interventions and improve patient outcomes.

Role of Machine Learning and Artificial Intelligence in Arrhythmias and Electrophysiology

Machine learning (ML), a subset of artificial intelligence (AI) centered on machines learning from extensive datasets, stands at the forefront of a technological revolution shaping various facets of society. Cardiovascular medicine has emerged as a key domain for ML applications, with considerable efforts to integrate these innovations into routine clinical practice. Within cardiac electrophysiology, ML applications, especially in the automated interpretation of electrocardiograms, have garnered substantial attention in existing literature. However, less recognized are the diverse applications of ML in cardiac electrophysiology and arrhythmias, spanning basic science research on arrhythmia mechanisms, both experimental and computational, as well as contributions to enhanced techniques for mapping cardiac electrical function and translational research related to arrhythmia management. This comprehensive review delves into various ML applications within the scope of this journal, organized into 3 parts. The first section provides a fundamental understanding of general ML principles and methodologies, serving as a foundational resource for readers interested in exploring ML applications in arrhythmia research. The second part offers an in-depth review of studies in arrhythmia and electrophysiology that leverage ML methodologies, showcasing the broad potential of ML approaches. Each subject is thoroughly outlined, accompanied by a review of notable ML research advancements. Finally, the review delves into the primary challenges and future perspectives surrounding ML-driven cardiac electrophysiology and arrhythmias research.

Early Diagnosis of Cardiovascular Diseases in the Era of Artificial Intelligence: An In-Depth Review

Cardiovascular diseases (CVDs) are significant health issues that result in high death rates globally. Early detection of cardiovascular events may lower the occurrence of acute myocardial infarction and reduce death rates in people with CVDs. Traditional data analysis is inadequate for managing multidimensional data related to the risk prediction of CVDs, heart attacks, medical image interpretations, therapeutic decision-making, and disease prognosis due to the complex pathological mechanisms and multiple factors involved. Artificial intelligence (AI) is a technology that utilizes advanced computer algorithms to extract information from large databases, and it has been integrated into the medical industry. AI methods have shown the ability to speed up the advancement of diagnosing and treating CVDs such as heart failure, atrial fibrillation, valvular heart disease, hypertrophic cardiomyopathy, congenital heart disease, and more. In clinical settings, AI has shown usefulness in diagnosing cardiovascular illness, improving the efficiency of supporting tools, stratifying and categorizing diseases, and predicting outcomes. Advanced AI algorithms have been intricately designed to analyze intricate relationships within extensive healthcare data, enabling them to tackle more intricate jobs compared to conventional approaches.

The autonomic nervous system in atrial fibrillation-pathophysiology and non-invasive assessment

The autonomic nervous system plays a crucial role in atrial fibrillation pathophysiology. Parasympathetic hyperactivity result in a shortening of the action potential duration, a reduction of the conduction wavelength, and as such facilitates reentry in the presence of triggers. Further, autonomic remodeling of atrial myocytes in AF includes progressive sympathetic hyperinnervation by increased atrial sympathetic nerve density and sympathetic atrial nerve sprouting. Knowledge on the pathophysiological process in AF, including the contribution of the autonomic nervous system, may in the near future guide personalized AF management. This review focuses on the role of the autonomic nervous system in atrial fibrillation pathophysiology and non-invasive assessment of the autonomic nervous system.

Non-linear processing and reinforcement learning to predict rTMS treatment response in depression

BACKGROUND

Forecasting the efficacy of repetitive transcranial magnetic stimulation (rTMS) therapy can lead to substantial time and cost savings by preventing futile treatments. To achieve this objective, we've formulated a machine learning approach aimed at categorizing patients with major depressive disorder (MDD) into two groups: individuals who respond (R) positively to rTMS treatment and those who do not respond (NR).

METHODS

Preceding the commencement of treatment, we obtained resting-state EEG data from 106 patients diagnosed with MDD, employing 32 electrodes for data collection. These patients then underwent a 7-week course of rTMS therapy, and 54 of them exhibited positive responses to the treatment. Employing Independent Component Analysis (ICA) on the EEG data, we successfully pinpointed relevant brain sources that could potentially serve as markers of neural activity within the dorsolateral prefrontal cortex (DLPFC). These identified sources were further scrutinized to estimate the sources of activity within the sensor domain. Then, we integrated supplementary physiological data and implemented specific criteria to yield more realistic estimations when compared to conventional EEG analysis. In the end, we selected components corresponding to the DLPFC region within the sensor domain. Features were derived from the time-series data of these relevant independent components. To identify the most significant features, we used Reinforcement Learning (RL). In categorizing patients into two groups - R and NR to rTMS treatment - we utilized three distinct classification algorithms including K-Nearest Neighbor (KNN), Support Vector Machine (SVM), and Multilayer Perceptron (MLP). We assessed the performance of these classifiers through a ten-fold cross-validation method. Additionally, we conducted a statistical test to evaluate the discriminative capacity of these features between responders and non-responders, opening the door for further exploration in this field.

RESULTS

We identified EEG features that can anticipate the response to rTMS treatment. The most robust discriminators included EEG beta power, the sum of bispectrum diagonal elements in the delta and beta frequency bands. When these features were combined into a single vector, the classification of responders and non-responders achieved impressive performance, with an accuracy of 95.28 %, specificity at 94.23 %, sensitivity reaching 96.29 %, and precision standing at 94.54 %, all achieved using SVM.

CONCLUSIONS

The results of this study suggest that the proposed approach, utilizing power, non-linear, and bispectral features extracted from relevant independent component time-series, has the capability to forecast the treatment outcome of rTMS for MDD patients based solely on a single pre-treatment EEG recording session. The achieved findings demonstrate the superior performance of our method compared to previous techniques.

Clinical applicability of an artificial intelligence prediction algorithm for early prediction of non-persistent atrial fibrillation

Background and aims

It is difficult to document atrial fibrillation (AF) on ECG in patients with non-persistent atrial fibrillation (non-PeAF). There is limited understanding of whether an AI prediction algorithm could predict the occurrence of non-PeAF from the information of normal sinus rhythm (SR) of a 12-lead ECG. This study aimed to derive a precise predictive AI model for screening non-PeAF using SR ECG within 4 weeks.

Methods

This retrospective cohort study included patients aged 18 to 99 with SR ECG on 12-lead standard ECG (10 seconds) in Ewha Womans University Medical Center for 3 years. Data were preprocessed into three window periods (which are defined with the duration from SR to non-PeAF detection) - 1 week, 2 weeks, and 4 weeks from the AF detection prospectively. For experiments, we adopted a Residual Neural Network model based on 1D-CNN proposed in a previous study. We used 7,595 SR ECGs (extracted from 215,875 ECGs) with window periods of 1 week, 2 weeks, and 4 weeks for analysis.

Results

The prediction algorithm showed an AUC of 0.862 and an F1-score of 0.84 in the 1:4 matched group of a 1-week window period. For the 1:4 matched group of a 2-week window period, it showed an AUC of 0.864 and an F1-score of 0.85. Finally, for the 1:4 matched group of a 4-week window period, it showed an AUC of 0.842 and an F1-score of 0.83.

Conclusion

The AI prediction algorithm showed the possibility of risk stratification for early detection of non-PeAF. Moreover, this study showed that a short window period is also sufficient to detect non-PeAF.

Machine learning approaches and non-linear processing of extracted components in frontal region to predict rTMS treatment response in major depressive disorder

Predicting the therapeutic result of repetitive transcranial magnetic stimulation (rTMS) treatment could save time and costs as ineffective treatment can be avoided. To this end, we presented a machine-learning-based strategy for classifying patients with major depression disorder (MDD) into responders (R) and nonresponders (NR) to rTMS treatment. Resting state EEG data were recorded using 32 electrodes from 88 MDD patients before treatment. Then, patients underwent 7 weeks of rTMS, and 46 of them responded to treatment. By applying Independent Component Analysis (ICA) on EEG, we identified the relevant brain sources as possible indicators of neural activity in the dorsolateral prefrontal cortex (DLPFC). This was served through estimating the generators of activity in the sensor domain. Subsequently, we added physiological information and placed certain terms and conditions to offer a far more realistic estimation than the classic EEG. Ultimately, those components mapped in accordance with the region of the DLPFC in the sensor domain were chosen. Features extracted from the relevant ICs time series included permutation entropy (PE), fractal dimension (FD), Lempel-Ziv Complexity (LZC), power spectral density, correlation dimension (CD), features based on bispectrum, frontal and prefrontal cordance, and a combination of them. The most relevant features were selected by a Genetic Algorithm (GA). For classifying two groups of R and NR, K-Nearest Neighbor (KNN), Support Vector Machine (SVM), and Multilayer Perceptron (MLP) were applied to predict rTMS treatment response. To evaluate the performance of classifiers, a 10-fold cross-validation method was employed. A statistical test was used to assess the capability of features in differentiating R and NR for further research. EEG characteristics that can predict rTMS treatment response were discovered. The strongest discriminative indicators were EEG beta power, the sum of bispectrum diagonal elements in delta and beta bands, and CD. The Combined feature vector classified R and NR with a high performance of 94.31% accuracy, 92.85% specificity, 95.65% sensitivity, and 92.85% precision using SVM. This result indicates that our proposed method with power and nonlinear and bispectral features from relevant ICs time-series can predict the treatment outcome of rTMS for MDD patients only by one session pretreatment EEG recording. The obtained results show that the proposed method outperforms previous methods.

Algorithms for automated diagnosis of cardiovascular diseases based on ECG data: A comprehensive systematic review

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.

A robust multiple heartbeats classification with weight-based loss based on convolutional neural network and bidirectional long short-term memory

Background: Analysis of electrocardiogram (ECG) provides a straightforward and non-invasive approach for cardiologists to diagnose and classify the nature and severity of variant cardiac diseases including cardiac arrhythmia. However, the interpretation and analysis of ECG are highly working-load demanding, and the subjective may lead to false diagnoses and heartbeats classification. In recent years, many deep learning works showed an excellent role in accurate heartbeats classification. However, the imbalance of heartbeat classes is universal in most of the available ECG databases since abnormal heartbeats are always relatively rare in real life scenarios. In addition, many existing approaches achieved prominent results by removing noise and extracting features in data preprocessing, which relies heavily on powerful computers. It is a pressing need to develop efficient and automatic light weighted algorithms for accurate heartbeats classification that can be used in portable ECG sensors. Objective: This study aims at developing a robust and efficient deep learning method, which can be embedded into wearable or portable ECG monitors for classifying heartbeats. Methods: We proposed a novel and light weighted deep learning architecture with weight-based loss based on a convolutional neural network (CNN) and bidirectional long short-term memory (Bi-LSTM) that can automatically identify five types of ECG heartbeats according to the AAMI EC57 standard. It was also true that the raw ECG signals were simply segmented without noise removal and other feature extraction processing. Moreover, to tackle the challenge of classification bias due to imbalanced ECG datasets for different types of arrhythmias, we introduced a weight-based loss function to reduce the influence of over-weighted categories in the ECG dataset. For avoiding the influence of the division of validation dataset, k-fold method was adopted to improve the reliability of the model. Results: The proposed algorithm is trained and tested on MIT-BIH Arrhythmia Database, and achieves an average of 99.33% accuracy, 93.67% sensitivity, 99.18% specificity, 89.85% positive prediction, and 91.65% F₁ score.

Atrioventricular Synchronization for Detection of Atrial Fibrillation and Flutter in One to Twelve ECG Leads Using a Dense Neural Network Classifier

This study investigates the use of atrioventricular (AV) synchronization as an important diagnostic criterion for atrial fibrillation and flutter (AF) using one to twelve ECG leads. Heart rate, lead-specific AV conduction time, and P-/f-wave amplitude were evaluated by three representative ECG metrics (mean value, standard deviation), namely RR-interval (RRi-mean, RRi-std), PQ-interval (PQi-mean, PQI-std), and PQ-amplitude (PQa-mean, PQa-std), in 71,545 standard 12-lead ECG records from the six largest PhysioNet CinC Challenge 2021 databases. Two rhythm classes were considered (AF, non-AF), randomly assigning records into training (70%), validation (20%), and test (10%) datasets. In a grid search of 19, 55, and 83 dense neural network (DenseNet) architectures and five independent training runs, we optimized models for one-lead, six-lead (chest or limb), and twelve-lead input features. Lead-set performance and SHapley Additive exPlanations (SHAP) input feature importance were evaluated on the test set. Optimal DenseNet architectures with the number of neurons in sequential [1st, 2nd, 3rd] hidden layers were assessed for sensitivity and specificity: DenseNet [16,16,0] with primary leads (I or II) had 87.9-88.3 and 90.5-91.5%; DenseNet [32,32,32] with six limb leads had 90.7 and 94.2%; DenseNet [32,32,4] with six chest leads had 92.1 and 93.2%; and DenseNet [128,8,8] with all 12 leads had 91.8 and 95.8%, indicating sensitivity and specificity values, respectively. Mean SHAP values on the entire test set highlighted the importance of RRi-mean (100%), RR-std (84%), and atrial synchronization (40-60%) for the PQa-mean (aVR, I), PQi-std (V2, aVF, II), and PQi-mean (aVL, aVR). Our focus on finding the strongest AV synchronization predictors of AF in 12-lead ECGs would lead to a comprehensive understanding of the decision-making process in advanced neural network classifiers. DenseNet self-learned to rely on a few ECG behavioral characteristics: first, characteristics usually associated with AF conduction such as rapid heart rate, enhanced heart rate variability, and large PQ-interval deviation in V2 and inferior leads (aVF, II); second, characteristics related to a typical P-wave pattern in sinus rhythm, which is best distinguished from AF by the earliest negative P-peak deflection of the right atrium in the lead (aVR) and late positive left atrial deflection in lateral leads (I, aVL). Our results on lead-selection and feature-selection practices for AF detection should be considered for one- to twelve-lead ECG signal processing settings, particularly those measuring heart rate, AV conduction times, and P-/f-wave amplitudes. Performances are limited to the AF diagnostic potential of these three metrics. SHAP value importance can be used in combination with a human expert's ECG interpretation to change the focus from a broad observation of 12-lead ECG morphology to focusing on the few AV synchronization findings strongly predictive of AF or non-AF arrhythmias. Our results are representative of AV synchronization findings across a broad taxonomy of cardiac arrhythmias in large 12-lead ECG databases.

Perspectives on Sex- and Gender-Specific Prediction of New-Onset Atrial Fibrillation by Leveraging Big Data

Atrial fibrillation (AF), the most common sustained cardiac arrhythmia, has a large impact on quality of life and is associated with increased risk of hospitalization, morbidity, and mortality. Over the past two decades advances regarding the clinical epidemiology and management of AF have been established. Moreover, sex differences in the prevalence, incidence, prediction, pathophysiology, and prognosis of AF have been identified. Nevertheless, AF remains to be a complex and heterogeneous disorder and a comprehensive sex- and gender-specific approach to predict new-onset AF is lacking. The exponential growth in various sources of big data such as electrocardiograms, electronic health records, and wearable devices, carries the potential to improve AF risk prediction. Leveraging these big data sources by artificial intelligence (AI)-enabled approaches, in particular in a sex- and gender-specific manner, could lead to substantial advancements in AF prediction and ultimately prevention. We highlight the current status, premise, and potential of big data to improve sex- and gender-specific prediction of new-onset AF.

Artificial Intelligence in Cardiology-A Narrative Review of Current Status

Artificial intelligence (AI) is an integral part of clinical decision support systems (CDSS), offering methods to approximate human reasoning and computationally infer decisions. Such methods are generally based on medical knowledge, either directly encoded with rules or automatically extracted from medical data using machine learning (ML). ML techniques, such as Artificial Neural Networks (ANNs) and support vector machines (SVMs), are based on mathematical models with parameters that can be optimally tuned using appropriate algorithms. The ever-increasing computational capacity of today’s computer systems enables more complex ML systems with millions of parameters, bringing AI closer to human intelligence. With this objective, the term deep learning (DL) has been introduced to characterize ML based on deep ANN (DNN) architectures with multiple layers of artificial neurons. Despite all of these promises, the impact of AI in current clinical practice is still limited. However, this could change shortly, as the significantly increased papers in AI, machine learning and deep learning in cardiology show. We highlight the significant achievements of recent years in nearly all areas of cardiology and underscore the mounting evidence suggesting how AI will take a central stage in the field.

Current Advancement in Diagnosing Atrial Fibrillation by Utilizing Wearable Devices and Artificial Intelligence: A Review Study

Atrial fibrillation (AF) is a common arrhythmia affecting 8–10% of the population older than 80 years old. The importance of early diagnosis of atrial fibrillation has been broadly recognized since arrhythmias significantly increase the risk of stroke, heart failure and tachycardia-induced cardiomyopathy with reduced cardiac function. However, the prevalence of atrial fibrillation is often underestimated due to the high frequency of clinically silent atrial fibrillation as well as paroxysmal atrial fibrillation, both of which are hard to catch by routine physical examination or 12-lead electrocardiogram (ECG). The development of wearable devices has provided a reliable way for healthcare providers to uncover undiagnosed atrial fibrillation in the population, especially those most at risk. Furthermore, with the advancement of artificial intelligence and machine learning, the technology is now able to utilize the database in assisting detection of arrhythmias from the data collected by the devices. In this review study, we compare the different wearable devices available on the market and review the current advancement in artificial intelligence in diagnosing atrial fibrillation. We believe that with the aid of the progressive development of technologies, the diagnosis of atrial fibrillation shall be made more effectively and accurately in the near future.

Advances in Cardiac Pacing: Arrhythmia Prediction, Prevention and Control Strategies

Cardiac arrhythmias constitute a tremendous burden on healthcare and are the leading cause of mortality worldwide. An alarming number of people have been reported to manifest sudden cardiac death as the first symptom of cardiac arrhythmias, accounting for about 20% of all deaths annually. Furthermore, patients prone to atrial tachyarrhythmias such as atrial flutter and fibrillation often have associated comorbidities including hypertension, ischemic heart disease, valvular cardiomyopathy and increased risk of stroke. Technological advances in electrical stimulation and sensing modalities have led to the proliferation of medical devices including pacemakers and implantable defibrillators, aiming to restore normal cardiac rhythm. However, given the complex spatiotemporal dynamics and non-linearity of the human heart, predicting the onset of arrhythmias and preventing the transition from steady state to unstable rhythms has been an extremely challenging task. Defibrillatory shocks still remain the primary clinical intervention for lethal ventricular arrhythmias, yet patients with implantable cardioverter defibrillators often suffer from inappropriate shocks due to false positives and reduced quality of life. Here, we aim to present a comprehensive review of the current advances in cardiac arrhythmia prediction, prevention and control strategies. We provide an overview of traditional clinical arrhythmia management methods and describe promising potential pacing techniques for predicting the onset of abnormal rhythms and effectively suppressing cardiac arrhythmias. We also offer a clinical perspective on bridging the gap between basic and clinical science that would aid in the assimilation of promising anti-arrhythmic pacing strategies.

PREDICTING CLINICAL RESPONSE TO TRANSCRANIAL MAGNETIC STIMULATION IN MAJOR DEPRESSION USING TIME-FREQUENCY EEG SIGNAL PROCESSING

Repetitive transcranial magnetic stimulation (rTMS) is defined as a noninvasive technique of brain stimulation conducted for both diagnostic and therapeutic purposes. rTMS can effectively excite the brain neurons and increase brain plasticity, which becomes particularly useful in psychiatric and neurological fields. Biomarkers that predict clinical outcomes in depression are essential for increasing the precision of treatments and clinical outcomes. The electroencephalogram (EEG) is a noninvasive neurophysiological test that is promising as a biomarker sensitive to treatment effects. The aim of our study was to investigate a novel nonlinear index of the resting state EEG activity as a predictor of clinical outcome and compare its predictive capacity to traditional frequency-based indices. EEG was recorded from 50 patients with treatment resistant depression (TRD) and 24 healthy comparison (HC) subjects. TRD patients were treated with excitatory rTMS to the dorsolateral prefrontal cortex (DLPFC) for 4–6 weeks. EEG signals were first decomposed using the ICA algorithm and the extracted components were then processed by time-frequency analysis. We then go on to compare the participants’ depression severity before, after, and 2 months after finishing the last treatment session using the proposed rTMS therapy. Absolute powers (APs), band powers (BPs), and theta and beta band entropies (BAs), which were extracted from the EEG, are used as features for the classification of changes in patients and normal cases after applying rTMS. Accordingly, we can go beyond the Beck score and clinically classify the EEG signal into two classes: depression and normal. The results demonstrated 78.37%, 74.32%, and 82.43% accuracy for artificial neural network (ANN), [Formula: see text]-nearest neighbor (KNN), and support vector machine (SVM) classifiers, respectively, indicating the superiority of the proposed method to those mentioned in similar studies. Also, the electrophysiological changes are shown to be evident in patients with major depression. Our data show that the time-frequency index yields superior outcome prediction performance compared to the traditional frequency band indices. Our findings warrant further investigation of EEG-based biomarkers in depression.

Prediction of Atrial Fibrillation Using Machine Learning: A Review

There has been recent immense interest in the use of machine learning techniques in the prediction and screening of atrial fibrillation, a common rhythm disorder present with significant clinical implications primarily related to the risk of ischemic cerebrovascular events and heart failure. Prior to the advent of the application of artificial intelligence in clinical medicine, previous studies have enumerated multiple clinical risk factors that can predict the development of atrial fibrillation. These clinical parameters include previous diagnoses, laboratory data (e.g., cardiac and inflammatory biomarkers, etc.), imaging data (e.g., cardiac computed tomography, cardiac magnetic resonance imaging, echocardiography, etc.), and electrophysiological data. These data are readily available in the electronic health record and can be automatically queried by artificial intelligence algorithms. With the modern computational capabilities afforded by technological advancements in computing and artificial intelligence, we present the current state of machine learning methodologies in the prediction and screening of atrial fibrillation as well as the implications and future direction of this rapidly evolving field.

Paroxysmal atrial fibrillation prediction based on morphological variant P-wave analysis with wideband ECG and deep learning

BACKGROUND AND OBJECTIVE

Atrial fibrillation (AF) is one of the most frequent asymptomatic arrhythmias associated with significant morbidity and mortality. Identifying the susceptibility to AF based on routine or continuous ECG recording is of considerable interest. Despite several P-wave characteristics and skin sympathetic nerve activity (SKNA) linked to AF onset, neither factor has offered accurate predictability. We propose a deep learning enabled method for AF risk prediction.

METHODS

We develop a novel MVPNet to predict the upcoming onset of paroxysmal AF. MVPNet combines wavelet-based feature extraction and a deep learning classifier. MVPNet detect the approaching of AF onset by analyzing the template and frequency in P-wave segments. The morphological variant P-wave (MVP) analysis includes P-wave and SKNA features cross temporal-spectral domain. Subsequently, we designed an optimized lightweight convolutional neural network model to detect the MVP features of pre-AF episodes during sinus rhythm segments. Wideband ECG data obtained through the neuECG protocol from eight PAF patients with 177 times AF occurrence in this study. We compared the accuracy of AF prediction between ordinary ECG and neuECG.

RESULTS

The MVPNet effectively predicted the onset of AF episodes. 89% of ECG recorded at 5 min before the AF onset can be identified using neuECG. The proposed deep learning model, MVPNet, obtained a better precision and inference speed with less computing resources than existing models. The gradient activation map showed that neuECG recording may be a superior AF risk predictor.

CONCLUSIONS

MVP analysis combined SKNA and P-wave parameters to improve predictive accuracy. The proposed MVPNet based on neuECG is superior to existing AF risk assessment with improved reliability and effectiveness. The method can be potentially applied in clinical scenarios for real-time, continuous AF prediction.

Review of Deep Learning-Based Atrial Fibrillation Detection Studies

Atrial fibrillation (AF) is a common arrhythmia that can lead to stroke, heart failure, and premature death. Manual screening of AF on electrocardiography (ECG) is time-consuming and prone to errors. To overcome these limitations, computer-aided diagnosis systems are developed using artificial intelligence techniques for automated detection of AF. Various machine learning and deep learning (DL) techniques have been developed for the automated detection of AF. In this review, we focused on the automated AF detection models developed using DL techniques. Twenty-four relevant articles published in international journals were reviewed. DL models based on deep neural network, convolutional neural network (CNN), recurrent neural network, long short-term memory, and hybrid structures were discussed. Our analysis showed that the majority of the studies used CNN models, which yielded the highest detection performance using ECG and heart rate variability signals. Details of the ECG databases used in the studies, performance metrics of the various models deployed, associated advantages and limitations, as well as proposed future work were summarized and discussed. This review paper serves as a useful resource for the researchers interested in developing innovative computer-assisted ECG-based DL approaches for AF detection.

TP-CNN: A Detection Method for atrial fibrillation based on transposed projection signals with compressed sensed ECG

BACKGROUND AND OBJECTIVE

Atrial fibrillation (AF) is the most prevalent arrhythmia, which increases the mortality of several complications. The use of wearable devices to detect atrial fibrillation is currently attracting a great deal of attention. Patients use wearable devices to continuously collect individual ECG signals and transmit them to the cloud for diagnosis. However, the ECG acquisition and transmission of wearable devices consumes a lot of energy. In order to solve this problem, some scholars have skipped the complex reconstruction process of compressed ECG signals and directly classified the compressed ECG signals, but the AF recognition rate is not high by this method. There is no explanation as to why the compressed ECG signals can be used for AF detection.

METHODS

Firstly, a simple deterministic measurement matrix (SDMM) is used to perform random projection operation on the ECG signals to complete the compression. Then, we use the transpose of the SDMM to perform transpose projection operation on the compressed signals in the cloud to obtain the approximate signals. We verify the similarity between the approximate ECG signal and the original ECG signal to explain why the compressed ECG signals are effective in AF detection. Finally, the Transposed Projection - Convolutional Neural Network (TP-CNN) is used to effectively detect AF on the obtained approximate ECG signals. Our proposed method is validated in the MIT-BIH AFDB.

RESULTS

The experimental results show that when compression ratios (CRs) are from 2 to 10, the average Pearson correlation coefficients between the approximate signals and the original signals are from 0.9867 to 0.8326, the average cosine similarities between the four frequency domain-based HRV features (including mean RR, RMSSD, SDNN and R density) are from 1.00 to 0.9958, from 1.00 to 0.9959, from 0.9978 to 0.8619 and from 0.9982 to 0.8707, respectively. Furthermore, when CR=10 (ECG was compressed to 1/10 of the original signal), the accuracy, specificity, f1 score and matthews correlation coefficient for AF detection of approximate signals were 99.32%, 99.43%, 99.14% and 98.57%, respectively.

CONCLUSION

Our proposed method illustrates the approximate signals have significant characteristics of the original signals and they are valid to classify the approximate signals. Meanwhile, comparing with the state-of-the-art methods, TP-CNN exceeded the results of the method for compressed signals and were also competitive compared with the classification results of the original signals, and is a promising method for AF detection in wearable application scenarios.

Estimation of PQ distance dispersion for atrial fibrillation detection

BACKGROUND AND OBJECTIVE

Atrial fibrillation (AF) is the most common cardiac arrhythmia in the world. It is associated with significantly increased morbidity and mortality. Diagnosis of the disease can be based on the analysis of the electrical atrial activity, on quantification of the heart rate irregularity or on a mixture of the both approaches. Since the amplitude of the atrial waves is small, their analysis can lead to false results. On the other hand, the heart rate based analysis usually leads to many unnecessary warnings. Therefore, our goal is to develop a new method for effective AF detection based on the analysis of the electrical atrial waves.

METHODS

The proposed method employs the fact that there is a lack of repeatable P waves preceding QRS complexes during AF. We apply the operation of spatio-temporal filtering (STF) to magnify and detect the prominent spatio-temporal patterns (STP) within the P waves in multi-channel ECG recordings. Later we measure their distances (PQ) to the succeeding QRS complexes, and we estimate dispersion of the obtained PQ series. For signals with normal sinus rhythm, this dispersion is usually very low, and contrary, for AF it is much raised. This allows for effective discrimination of this cardiologic disorder.

RESULTS

Tested on an ECG database consisting of AF cases, normal rhythm cases and cases with normal rhythm restored by the use of cardioversion, the method proposed allowed for AF detection with the accuracy of 98.75% on the basis of both 8-channel and 2-channel signals of 12 s length. When the signals length was decreased to 6 s, the accuracy varied in the range of 95%-97.5% depending on the number of channels and the dispersion measure applied.

CONCLUSIONS

Our approach allows for high accuracy of atrial fibrillation detection using the analysis of electrical atrial activity. The method can be applied to an early detection of the desease and can advantageously be used to decrease the number of false warnings in systems based on the analysis of the heart rate.

Atrial Fibrillation Prediction from Critically Ill Sepsis Patients

Sepsis is defined by life-threatening organ dysfunction during infection and is the leading cause of death in hospitals. During sepsis, there is a high risk that new onset of atrial fibrillation (AF) can occur, which is associated with significant morbidity and mortality. Consequently, early prediction of AF during sepsis would allow testing of interventions in the intensive care unit (ICU) to prevent AF and its severe complications. In this paper, we present a novel automated AF prediction algorithm for critically ill sepsis patients using electrocardiogram (ECG) signals. From the heart rate signal collected from 5-min ECG, feature extraction is performed using the traditional time, frequency, and nonlinear domain methods. Moreover, variable frequency complex demodulation and tunable Q-factor wavelet-transform-based time-frequency methods are applied to extract novel features from the heart rate signal. Using a selected feature subset, several machine learning classifiers, including support vector machine (SVM) and random forest (RF), were trained using only the 2001 Computers in Cardiology data set. For testing the proposed method, 50 critically ill ICU subjects from the Medical Information Mart for Intensive Care (MIMIC) III database were used in this study. Using distinct and independent testing data from MIMIC III, the SVM achieved 80% sensitivity, 100% specificity, 90% accuracy, 100% positive predictive value, and 83.33% negative predictive value for predicting AF immediately prior to the onset of AF, while the RF achieved 88% AF prediction accuracy. When we analyzed how much in advance we can predict AF events in critically ill sepsis patients, the algorithm achieved 80% accuracy for predicting AF events 10 min early. Our algorithm outperformed a state-of-the-art method for predicting AF in ICU patients, further demonstrating the efficacy of our proposed method. The annotations of patients' AF transition information will be made publicly available for other investigators. Our algorithm to predict AF onset is applicable for any ECG modality including patch electrodes and wearables, including Holter, loop recorder, and implantable devices.

A Classification and Prediction Hybrid Model Construction with the IQPSO-SVM Algorithm for Atrial Fibrillation Arrhythmia

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

How machine learning is impacting research in atrial fibrillation: implications for risk prediction and future management

There has been an exponential growth of artificial intelligence (AI) and machine learning (ML) publications aimed at advancing our understanding of atrial fibrillation (AF), which has been mainly driven by the confluence of two factors: the advances in deep neural networks (DeepNNs) and the availability of large, open access databases. It is observed that most of the attention has centred on applying ML for dvsetecting AF, particularly using electrocardiograms (ECGs) as the main data modality. Nearly a third of them used DeepNNs to minimize or eliminate the need for transforming the ECGs to extract features prior to ML modelling; however, we did not observe a significant advantage in following this approach. We also found a fraction of studies using other data modalities, and others centred in aims, such as risk prediction, AF management, and others. From the clinical perspective, AI/ML can help expand the utility of AF detection and risk prediction, especially for patients with additional comorbidities. The use of AI/ML for detection and risk prediction into applications and smart mobile health (mHealth) technology would enable ‘real time’ dynamic assessments. AI/ML could also adapt to treatment changes over time, as well as incident risk factors. Incorporation of a dynamic AI/ML model into mHealth technology would facilitate ‘real time’ assessment of stroke risk, facilitating mitigation of modifiable risk factors (e.g. blood pressure control). Overall, this would lead to an improvement in clinical care for patients with AF.

Prediction of paroxysmal atrial fibrillation using new heart rate variability features

Paroxysmal atrial fibrillation (PAF) is a cardiac arrhythmia that can eventually lead to heart failure or stroke if left untreated. Early detection of PAF is therefore crucial to prevent any further complications and avoid fatalities. An implantable defibrillator device could be used to both detect and treat the condition though such devices have limited computational capability. With this constraint in mind, this paper presents a novel set of features to accurately predict the presence of PAF. The method is evaluated using ECG signals from the widely used atrial fibrillation prediction database (AFPDB) from PhysioNet. We analysed 106 signals from 53 pairs of ECG recordings. Each pair of signals contains one 5-min ECG segment that ends just before the onset of a PAF event and another 5-min ECG segment at least 45 min distant from the PAF event, to represent a non-PAF event. Seven novel features are extracted through the Poincaré representation of R-R interval signals, and are prioritised through feature ranking schemes. The features are used with four standard classification techniques for PAF prediction and compared to the existing state of the art from the literature. Using only the seven proposed features, classification performance outperforms those of the classical state-of-the-art feature set, registering sensitivity and specificity measurements of over 96%. The results further improve when the features are combined with several of the classical features, with an accuracy increasing to 98% using a linear kernel SVM. The results show that the proposed features provide a useful representation of the PAF condition and achieve good prediction with off-the-shelf classification techniques that would be suitable for ICU deployment.

Artificial Intelligence in Hypertension: Seeing Through a Glass Darkly

Hypertension remains the largest modifiable cause of mortality worldwide despite the availability of effective medications and sustained research efforts over the past 100 years. Hypertension requires transformative solutions that can help reduce the global burden of the disease. Artificial intelligence and machine learning, which have made a substantial impact on our everyday lives over the last decade may be the route to this transformation. However, artificial intelligence in health care is still in its nascent stages and realizing its potential requires numerous challenges to be overcome. In this review, we provide a clinician-centric perspective on artificial intelligence and machine learning as applied to medicine and hypertension. We focus on the main roadblocks impeding implementation of this technology in clinical care and describe efforts driving potential solutions. At the juncture, there is a critical requirement for clinical and scientific expertise to work in tandem with algorithmic innovation followed by rigorous validation and scrutiny to realize the promise of artificial intelligence-enabled health care for hypertension and other chronic diseases.

An efficient approach for Paroxysmal Atrial Fibrillation events prediction using Extreme Learning Machine

Paroxysmal Atrial Fibrillation (PAF) is a special class of Atrial Fibrillation. Predicting PAF events from electrocardiogram (ECG) signal streams plays a vital role in generating real-time alerts for cardiac disorders. These alerts are extremely important to cardiologists in taking precautions to prevent their patients from having a stroke. In this study, an effective predictive approach to PAF events using the Extreme Learning Machine classification technique is proposed. Besides, we propose a feature extraction method that integrates new ECG signal features to its time-domain ones. The new features are based on the construction of sparse vectors for peaks in ECG signals that provide high overlap between similar ECGs. The proposed prediction approach with the new ECG features representation were evaluated on a real PAF dataset using the five-fold cross-validation method. Experiments show promising results for predicting PAF in terms of accuracy and execution time compared to other existing studies. The proposed approach achieved classification accuracy of 97% for non-streaming ECG signals mode and 94.4% for streaming mode.

Machine Learning in Arrhythmia and Electrophysiology

Machine learning (ML), a branch of artificial intelligence, where machines learn from big data, is at the crest of a technological wave of change sweeping society. Cardiovascular medicine is at the forefront of many ML applications, and there is a significant effort to bring them into mainstream clinical practice. In the field of cardiac electrophysiology, ML applications have also seen a rapid growth and popularity, particularly the use of ML in the automatic interpretation of ECGs, which has been extensively covered in the literature. Much lesser known are the other aspects of ML application in cardiac electrophysiology and arrhythmias, such as those in basic science research on arrhythmia mechanisms, both experimental and computational; in the development of better techniques for mapping of cardiac electrical function; and in translational research related to arrhythmia management. In the current review, we examine comprehensively such ML applications as they match the scope of this journal. The current review is organized in 3 parts. The first provides an overview of general ML principles and methodologies that will afford readers of the necessary information on the subject, serving as the foundation for inviting further ML applications in arrhythmia research. The basic information we provide can serve as a guide on how one might design and conduct an ML study. The second part is a review of arrhythmia and electrophysiology studies in which ML has been utilized, highlighting the broad potential of ML approaches. For each subject, we outline comprehensively the general topics, while reviewing some of the research advances utilizing ML under the subject. Finally, we discuss the main challenges and the perspectives for ML-driven cardiac electrophysiology and arrhythmia research.

Detecting undiagnosed atrial fibrillation in UK primary care: Validation of a machine learning prediction algorithm in a retrospective cohort study

AIMS

To evaluate the ability of a machine learning algorithm to identify patients at high risk of atrial fibrillation in primary care.

METHODS

A retrospective cohort study was undertaken using the DISCOVER registry to validate an algorithm developed using a Clinical Practice Research Datalink (CPRD) dataset. The validation dataset included primary care patients in London, England aged ≥30 years from 1 January 2006 to 31 December 2013, without a diagnosis of atrial fibrillation in the prior 5 years. Algorithm performance metrics were sensitivity, specificity, positive predictive value, negative predictive value (NPV) and number needed to screen (NNS). Subgroup analysis of patients aged ≥65 years was also performed.

RESULTS

Of 2,542,732 patients in DISCOVER, the algorithm identified 604,135 patients suitable for risk assessment. Of these, 3.0% (17,880 patients) had a diagnosis of atrial fibrillation recorded before study end. The area under the curve of the receiver operating characteristic was 0.87, compared with 0.83 in algorithm development. The NNS was nine patients, matching the CPRD cohort. In patients aged ≥30 years, the algorithm correctly identified 99.1% of patients who did not have atrial fibrillation (NPV) and 75.0% of true atrial fibrillation cases (sensitivity). Among patients aged ≥65 years (n = 117,965), the NPV was 96.7% with 91.8% sensitivity.

CONCLUSIONS

This atrial fibrillation risk prediction algorithm, based on machine learning methods, identified patients at highest risk of atrial fibrillation. It performed comparably in a large, real-world population-based cohort and the developmental registry cohort. If implemented in primary care, the algorithm could be an effective tool for narrowing the population who would benefit from atrial fibrillation screening in the United Kingdom.

Localizing confined epileptic foci in patients with an unclear focus or presumed multifocality using a component-based EEG-fMRI method

Precise localization of epileptic foci is an unavoidable prerequisite in epilepsy surgery. Simultaneous EEG-fMRI recording has recently created new horizons to locate foci in patients with epilepsy and, in comparison with single-modality methods, has yielded more promising results although it is still subject to limitations such as lack of access to information between interictal events. This study assesses its potential added value in the presurgical evaluation of patients with complex source localization. Adult candidates considered ineligible for surgery on account of an unclear focus and/or presumed multifocality on the basis of EEG underwent EEG-fMRI. Adopting a component-based approach, this study attempts to identify the neural behavior of the epileptic generators and detect the components-of-interest which will later be used as input in the GLM model, substituting the classical linear regressor. Twenty-eight sets interictal epileptiform discharges (IED) from nine patients were analyzed. In eight patients, at least one BOLD response was significant, positive and topographically related to the IEDs. These patients were rejected for surgery because of an unclear focus in four, presumed multifocality in three, and a combination of the two conditions in two. Component-based EEG-fMRI improved localization in five out of six patients with unclear foci. In patients with presumed multifocality, component-based EEG-fMRI advocated one of the foci in five patients and confirmed multifocality in one of the patients. In seven patients, component-based EEG-fMRI opened new prospects for surgery and in two of these patients, intracranial EEG supported the EEG-fMRI results. In these complex cases, component-based EEG-fMRI either improved source localization or corroborated a negative decision regarding surgical candidacy. As supported by the statistical findings, the developed EEG-fMRI method leads to a more realistic estimation of localization compared to the conventional EEG-fMRI approach, making it a tool of high value in pre-surgical evaluation of patients with refractory epilepsy. To ensure proper implementation, we have included guidelines for the application of component-based EEG-fMRI in clinical practice.

Proposing a convolutional neural network for stress assessment by means of derived heart rate from functional near infrared spectroscopy

BACKGROUND

Stress is known as one of the major factors threatening human health. A large number of studies have been performed in order to either assess or relieve stress by analyzing the brain and heart-related signals.

METHOD

In this study, a method based on the Convolutional Neural Network (CNN) approach is proposed to assess stress induced by the Montreal Imaging Stress Task. The proposed model is trained on the heart rate signal derived from functional Near-Infrared Spectroscopy (fNIRS), which is referred to as HRF. In this regard, fNIRS signals of 20 healthy volunteers were recorded using a configuration of 23 channels located on the prefrontal cortex. The proposed deep learning system consists of two main parts where in the first part, the one-dimensional convolutional neural network is employed to build informative activation maps, and then in the second part, a stack of deep fully connected layers is used to predict the stress existence probability. Thereafter, the employed CNN method is compared with the Dense Neural Network, Support Vector Machine, and Random Forest regarding various classification metrics.

RESULTS

Results clearly showed the superiority of CNN over all other methods. Additionally, the trained HRF model significantly outperforms the model trained on the filtered fNIRS signals, where the HRF model could achieve 98.69 ± 0.45% accuracy, which is 10.09% greater than the accuracy obtained by the fNIRS model.

CONCLUSIONS

Employment of the proposed deep learning system trained on the HRF measurements leads to higher stress classification accuracy than the accuracy reported in the existing studies where the same experimental procedure has been done. Besides, the proposed method suggests better stability with lower variation in prediction. Furthermore, its low computational cost opens up the possibility to be applied in real-time monitoring of stress assessment.

Early Detection of Sudden Cardiac Death by Using Ensemble Empirical Mode Decomposition-Based Entropy and Classical Linear Features From Heart Rate Variability Signals

Sudden cardiac death (SCD), which can deprive a person of life within minutes, is a destructive heart abnormality. Thus, providing early warning information for patients at risk of SCD, especially those outside hospitals, is essential. In this study, we investigated the performances of ensemble empirical mode decomposition (EEMD)-based entropy features on SCD identification. EEMD-based entropy features were obtained by using the following technology: (1) EEMD was performed on HRV beats to decompose them into intrinsic mode functions (IMFs), (2) five entropy parameters, namely Rényi entropy (RenEn), fuzzy entropy (FuEn), dispersion Entropy (DisEn), improved multiscale permutation entropy (IMPE), and Renyi distribution entropy(RdisEn), were computed from the first four IMFs obtained, which were named EEMD-based entropy features. Additionally, an automated scheme combining EEMD-based entropy and classical linear (time and frequency domains) features was proposed with the intention of detecting SCD early by analyzing 14 min (at seven successive intervals of 2 min) heart rate variability (HRV) in signals from a normal population and subjects at risk of SCD. Firstly, EEMD-based entropy and classical linear measurements were extracted from HRV beats, and then the integrated measurements were ranked by various methodologies, i.e., t-test, entropy, receiver-operating characteristics (ROC), Wilcoxon, and Bhattacharyya. Finally, these ranked features were fed into a k-Nearest Neighbor algorithm for classification. Compared with several state-of-the-art methods, the proposed scheme firstly predicted subjects at risk of SCD up to 14 min earlier with an accuracy of 96.1%, a sensitivity of 97.5%, and a specificity of 94.4% 14 min before SCD onset. The simulation results exhibited that EEMD-based entropy estimators showed significant difference between SCD patients and normal individuals and outperformed the classical linear estimators in SCD detection, the EEMD-based FuEn and IMPE indexes were particularly useful assessments for identification of patients at risk of SCD and can be used as novel indices to reveal the disorders of rhythm variations of the autonomic nervous system when affected by SCD.

High Precision Digitization of Paper-Based ECG Records: A Step Toward Machine Learning

Introduction: The electrocardiogram (ECG) plays an important role in the diagnosis of heart diseases. However, most patterns of diseases are based on old datasets and stepwise algorithms that provide limited accuracy. Improving diagnostic accuracy of the ECG can be done by applying machine learning algorithms. This requires taking existing scanned or printed ECGs of old cohorts and transforming the ECG signal to the raw digital (time (milliseconds), voltage (millivolts)) form. Objectives: We present a MATLAB-based tool and algorithm that converts a printed or scanned format of the ECG into a digitized ECG signal. Methods: 30 ECG scanned curves are utilized in our study. An image processing method is first implemented for detecting the ECG regions of interest and extracting the ECG signals. It is followed by serial steps that digitize and validate the results. Results: The validation demonstrates very high correlation values of several standard ECG parameters: PR interval 0.984 +/−0.021 (p-value < 0.001), QRS interval 1+/− SD (p-value < 0.001), QT interval 0.981 +/− 0.023 p-value < 0.001, and RR interval 1 +/− 0.001 p-value < 0.001. Conclusion: Digitized ECG signals from existing paper or scanned ECGs can be obtained with more than 95% of precision. This makes it possible to utilize historic ECG signals in machine learning algorithms to identify patterns of heart diseases and aid in the diagnostic and prognostic evaluation of patients with cardiovascular disease.

Simulation and in vivo investigation of light-emitting diode, near infrared Gaussian beam profiles

Near infrared spectroscopy is an optical imaging technique which offers a non-invasive, portable, and low-cost method for continuously measuring the oxygenation of tissues. In particular, it can provide the brain activation through measuring the blood oxygenation and blood volume in the cortex. Understanding and then improving the spatial and depth sensitivity of near infrared spectroscopy measurements to brain tissue are essential for designing experiments as well as interpreting research findings. In this study, we investigate the effect of applying two common light beam profiles including Uniform and Gaussian on the penetration depth of an LED-based near infrared spectroscopy. In this regard, two Gaussian profiles were produced by adjusting plano-convex and bi-convex lenses and the Uniform profile was provided by applying a flat lens. Two experiments were conducted in this study. First, a simulation experiment was carried out based on scanning the intra space of a liquid phantom by using static and pulsating absorbers to compare the penetration depth of the configurations applied on the LED-based near infrared spectroscopy with that of a laser-based near infrared spectroscopy. Second, to show the feasibility of the best proposed configuration applied, an in vivo experiment of stress assessment has been performed and its results have been compared with that results obtained by laser one. The results showed that the LED-based near infrared spectroscopy equipped with bi-convex lens provides a penetration depth and hence quality measurements of near infrared spectroscopy and its extracted heart rate variability signals as well as laser-based near infrared spectroscopy especially in the application of stress assessment.

Applications of Artificial Intelligence in Cardiology. The Future is Already Here

There is currently no other hot topic like the ability of current technology to develop capabilities similar to those of human beings, even in medicine. This ability to simulate the processes of human intelligence with computer systems is known as artificial intelligence (AI). This article aims to clarify the various terms that still sound foreign to us, such as AI, machine learning (ML), deep learning (DL), and big data. It also provides an in-depth description of the concept of AI and its types; the learning techniques and technology used by ML; cardiac imaging analysis with DL; and the contribution of this technological revolution to classical statistics, as well as its current limitations, legal aspects, and initial applications in cardiology. To do this, we conducted a detailed PubMed search on the evolution of original contributions on AI to the various areas of application in cardiology in the last 5 years and identified 673 research articles. We provide 19 detailed examples from distinct areas of cardiology that, by using AI, have shown diagnostic and therapeutic improvements, and which will aid understanding of ML and DL methodology.

Evidence-based Clinical Decision Support Systems for the prediction and detection of three disease states in critical care: A systematic literature review

Background: Clinical decision support (CDS) systems have emerged as tools providing intelligent decision making to address challenges of critical care. CDS systems can be based on existing guidelines or best practices; and can also utilize machine learning to provide a diagnosis, recommendation, or therapy course. Methods: This research aimed to identify evidence-based study designs and outcome measures to determine the clinical effectiveness of clinical decision support systems in the detection and prediction of hemodynamic instability, respiratory distress, and infection within critical care settings. PubMed, ClinicalTrials.gov and Cochrane Database of Systematic Reviews were systematically searched to identify primary research published in English between 2013 and 2018. Studies conducted in the USA, Canada, UK, Germany and France with more than 10 participants per arm were included. Results: In studies on hemodynamic instability, the prediction and management of septic shock were the most researched topics followed by the early prediction of heart failure. For respiratory distress, the most popular topics were pneumonia detection and prediction followed by pulmonary embolisms. Given the importance of imaging and clinical notes, this area combined Machine Learning with image analysis and natural language processing. In studies on infection, the most researched areas were the detection, prediction, and management of sepsis, surgical site infections, as well as acute kidney injury. Overall, a variety of Machine Learning algorithms were utilized frequently, particularly support vector machines, boosting techniques, random forest classifiers and neural networks. Sensitivity, specificity, and ROC AUC were the most frequently reported performance measures. Conclusion: This review showed an increasing use of Machine Learning for CDS in all three areas. Large datasets are required for training these algorithms; making it imperative to appropriately address, challenges such as class imbalance, correct labelling of data and missing data. Recommendations are formulated for the development and successful adoption of CDS systems.

Evidence-based Clinical Decision Support Systems for the prediction and detection of three disease states in critical care: A systematic literature review

Background: Clinical decision support (CDS) systems have emerged as tools providing intelligent decision making to address challenges of critical care. CDS systems can be based on existing guidelines or best practices; and can also utilize machine learning to provide a diagnosis, recommendation, or therapy course. Methods: This research aimed to identify evidence-based study designs and outcome measures to determine the clinical effectiveness of clinical decision support systems in the detection and prediction of hemodynamic instability, respiratory distress, and infection within critical care settings. PubMed, ClinicalTrials.gov and Cochrane Database of Systematic Reviews were systematically searched to identify primary research published in English between 2013 and 2018. Studies conducted in the USA, Canada, UK, Germany and France with more than 10 participants per arm were included. Results: In studies on hemodynamic instability, the prediction and management of septic shock were the most researched topics followed by the early prediction of heart failure. For respiratory distress, the most popular topics were pneumonia detection and prediction followed by pulmonary embolisms. Given the importance of imaging and clinical notes, this area combined Machine Learning with image analysis and natural language processing. In studies on infection, the most researched areas were the detection, prediction, and management of sepsis, surgical site infections, as well as acute kidney injury. Overall, a variety of Machine Learning algorithms were utilized frequently, particularly support vector machines, boosting techniques, random forest classifiers and neural networks. Sensitivity, specificity, and ROC AUC were the most frequently reported performance measures. Conclusion: This review showed an increasing use of Machine Learning for CDS in all three areas. Large datasets are required for training these algorithms; making it imperative to appropriately address, challenges such as class imbalance, correct labelling of data and missing data. Recommendations are formulated for the development and successful adoption of CDS systems.

Toward a closed-loop deep brain stimulation in Parkinson's disease using local field potential in parkinsonian rat model

Deep brain stimulation (DBS) is an invasive method used for treating Parkinson's disease in its advanced stages. Nowadays, the initial adjustment of DBS parameters and their automatic matching proportion to the progression of the disease is viewed as one of the research areas discussed by the researchers, which is called closed-loop DBS. Various studies were conducted regarding finding the signal(s) which reflects different symptoms of the disease. Local Field Potential (LFP) is one of the signals that is suitable for using as feedback, because it can be recorded by the same implemented electrodes for stimulation. The present study aimed to identify the distinguishing features of patients from healthy individuals using LFP signals.

METHODS

In the present study, LFP was recorded from the rats in sham and parkinsonian model groups. After evaluating the signals in the frequency domain, sixty-six features were extracted from power spectral density of LFPs. The features were classified by Support Vector Machine (SVM) to determine the ability of features for separating parkinsonian rats from healthy ones. Finally, the most effective features were selected for distinguishing between the sham and parkinsonian model groups using a genetic algorithm.

RESULTS

The results indicated that the frequency domain features of LFP signals from rats have capacity of using them as a feedback for closed-loop DBS. The accuracy of the Support Vector Machine classification using all 66 features was 80.42% which increased to 84.41% using 38 features selected by genetic algorithm. The proposed method not only increase the accuracy, but it also reduce computation by decreasing the number of the effective features. The results indicate the significant capacity of the proposed method for identifying the effective high-frequency features to control the closed-loop DBS.

CONCLUSIONS

The ability of using LFP signals as feedback in closed-loop DBS was shown by extracting useful information in frequency bands below and above 100 Hz regarding LFP signals of parkinsonian rats and sham ones. Based on the results, features at frequencies above 100 Hz were more powerful and robust than below 100 Hz. The genetic algorithm was used for optimizing the classification problem.

Component-related BOLD response to localize epileptic focus using simultaneous EEG-fMRI recordings at 3T

BACKGROUND

Simultaneous EEG-fMRI experiments record spatiotemporal dynamics of epileptic activity. A shortcoming of spike-based EEG-fMRI studies is their inability to provide information about behavior of epileptic generators when no spikes are visible.

NEW METHOD

We extract time series of epileptic components identified on EEG and fit them with Generalized Linear Model (GLM) model. This allows a precise and reliable localization of epileptic foci in addition to predicting generator's behavior. The proposed method works in the source domain and delineates generators considering spatial correlation between spike template and candidate components in addition to patient's medical records.

RESULTS

The proposed method was applied on 20 patients with refractory epilepsy and 20 age- and gender-matched healthy controls. The identified components were examined statistically and threshold of localization accuracy was determined as 86% based on Receiver Operating Characteristic (ROC) curve analysis. Accuracy, sensitivity, and specificity were found to be 88%, 85%, and 95%, respectively. Contribution of EEG-fMRI and concordance between EEG and fMRI were also evaluated. Concordance was found in 19 patients and contribution in 17.

COMPARISON WITH EXISTING METHODS

We compared the proposed method with conventional methods. Our comparisons showed superiority of the proposed method. In particular, when epileptogenic zone was located deep in the brain, the method outperformed existing methods.

CONCLUSIONS

This study contributes substantially to increasing the yield of EEG-fMRI and presents a realistic estimate of the neural behavior of epileptic generators, to the best of our knowledge, for the first time in the literature.

TOWARDS AN AUTOMATIC DIAGNOSIS SYSTEM FOR LUMBAR DISC HERNIATION: THE SIGNIFICANCE OF LOCAL SUBSET FEATURE SELECTION

Herniation in the lumbar area is one of the most common diseases which results in lower back pain (LBP) causing discomfort and inconvenience in the patients’ daily lives. A computer aided diagnosis (CAD) system can be of immense benefit as it generates diagnostic results within a short time while increasing precision of diagnosis and eliminating human errors. We have proposed a new method for automatic diagnosis of lumbar disc herniation based on clinical MRI data. We use T2-W sagittal and myelograph images. The presented method has been applied on 30 clinical cases, each containing 7 discs (210 lumbar discs) for the herniation diagnosis. We employ Otsu thresholding method to extract the spinal cord from MR images of lumbar disc. A third order polynomial is then aligned on the extracted spinal cords, and by the end of preprocessing stage, all the T2-W sagittal images will have been prepared for specifying disc boundary and labeling. Having extracted an ROI for each disc, we proceed to use intensity and shape features for classification. The extracted features have been selected by Local Subset Feature Selection. The results demonstrated 91.90%, 92.38% and 95.23% accuracy for artificial neural network, K-nearest neighbor and support vector machine (SVM) classifiers respectively, indicating the superiority of the proposed method to those mentioned in similar studies.

A novel approach to emotion recognition using local subset feature selection and modified Dempster-Shafer theory

Background

Emotion recognition is an increasingly important field of research in brain computer interactions.

Introduction

With the advance of technology, automatic emotion recognition systems no longer seem far-fetched. Be that as it may, detecting neural correlates of emotion has remained a substantial bottleneck. Settling this issue will be a breakthrough of significance in the literature.

Methods

The current study aims to identify the correlations between different emotions and brain regions with the help of suitable electrodes. Initially, independent component analysis algorithm is employed to remove artifacts and extract the independent components. The informative channels are then selected based on the thresholded average activity value for obtained components. Afterwards, effective features are extracted from selected channels common between all emotion classes. Features are reduced using the local subset feature selection method and then fed to a new classification model using modified Dempster-Shafer theory of evidence.

Results

The presented method is employed to DEAP dataset and the results are compared to those of previous studies, which highlights the significant ability of this method to recognize emotions through electroencephalography, by the accuracy of about 91%. Finally, the obtained results are discussed and new aspects are introduced.

Conclusions

The present study addresses the long-standing challenge of finding neural correlates between human emotions and the activated brain regions. Also, we managed to solve uncertainty problem in emotion classification which is one of the most challenging issues in this field. The proposed method could be employed in other practical applications in future.