Multi-center atrial fibrillation electrocardiogram (ECG) classification using Fourier space convolutional neural networks (FD-CNN) and transfer learning.

There has been a proliferation of machine learning (ML) electrocardiogram (ECG) classification algorithms reaching >85% accuracy for various cardiac pathologies. Despite the high accuracy at individual institutions, challenges remain when it comes to multi-center deployment. Transfer learning (TL) is a technique in which a model trained for a specific task is repurposed for another related task, in this case ECG ML model trained at one institution is fine-tuned to be utilized to classify ECGs at another institution. Models trained at one institution, however, might not be generalizable for accurate classification when deployed broadly due to differences in type, time, and sampling rate of traditional ECG acquisition. In this study, we evaluate the performance of time domain (TD) and frequency domain (FD) convolutional neural network (CNN) classification models in an inter-institutional scenario leveraging three different publicly available datasets. The larger PTB-XL ECG dataset was used to initially train TD and FD CNN models for atrial fibrillation (AFIB) classification. The models were then tested on two different data sets, Lobachevsky University Electrocardiography Database (LUDB) and Korea University Medical Center database (KURIAS). The FD model was able to retain most of its performance (>0.81 F1-score), whereas TD was highly affected (<0.53 F1-score) by the dataset variations, even with TL applied. The FD CNN showed superior robustness to cross-institutional variability and has potential for widespread application with no compromise to ECG classification performance.

Fourier space approach for convolutional neural network (CNN) electrocardiogram (ECG) classification: A proof-of-concept study.

There has been a proliferation of machine learning (ML) electrocardiogram (ECG) classification algorithms reaching > 85% accuracy for various cardiac pathologies. Although the accuracy within institutions might be high, models trained at one institution might not be generalizable enough for accurate detection when deployed in other institutions due to differences in type of signal acquisition, sampling frequency, time of acquisition, device noise characteristics and number of leads. In this proof-of-concept study, we leverage the publicly available PTB-XL dataset to investigate the use of time-domain (TD) and frequency-domain (FD) convolutional neural networks (CNN) to detect myocardial infarction (MI), ST/T-wave changes (STTC), atrial fibrillation (AFIB) and sinus arrhythmia (SARRH). To simulate interinstitutional deployment, the TD and FD implementations were also compared on adapted test sets using different sampling frequencies 50 Hz, 100 Hz and 250 Hz, and acquisition times of 5 s and 10s at 100 Hz sampling frequency from the training dataset. When tested on the original sampling frequency and duration, the FD approach showed comparable results to TD for MI (0.92 FD - 0.93 TD AUROC) and STTC (0.94 FD - 0.95 TD AUROC), and better performance for AFIB (0.99 FD - 0.86 TD AUROC) and SARRH (0.91 FD - 0.65 TD AUROC). Although both methods were robust to changes in sampling frequency, changes in acquisition time were detrimental to the TD MI and STTC AUROCs, at 0.72 and 0.58 respectively. Alternatively, the FD approach was able to maintain the same level of performance, and, therefore, showed better potential for interinstitutional deployment.

A Study of R-R Interval Transition Matrix Features for Machine Learning Algorithms in AFib Detection.

Atrial Fibrillation (AFib) is a heart condition that occurs when electrophysiological malformations within heart tissues cause the atria to lose coordination with the ventricles, resulting in "irregularly irregular" heartbeats. Because symptoms are subtle and unpredictable, AFib diagnosis is often difficult or delayed. One possible solution is to build a system which predicts AFib based on the variability of R-R intervals (the distances between two R-peaks). This research aims to incorporate the transition matrix as a novel measure of R-R variability, while combining three segmentation schemes and two feature importance measures to systematically analyze the significance of individual features. The MIT-BIH dataset was first divided into three segmentation schemes, consisting of 5-s, 10-s, and 25-s subsets. In total, 21 various features, including the transition matrix features, were extracted from these subsets and used for the training of 11 machine learning classifiers. Next, permutation importance and tree-based feature importance calculations determined the most predictive features for each model. In summary, with Leave-One-Person-Out Cross Validation, classifiers under the 25-s segmentation scheme produced the best accuracies; specifically, Gradient Boosting (96.08%), Light Gradient Boosting (96.11%), and Extreme Gradient Boosting (96.30%). Among eleven classifiers, the three gradient boosting models and Random Forest exhibited the highest overall performance across all segmentation schemes. Moreover, the permutation and tree-based importance results demonstrated that the transition matrix features were most significant with longer subset lengths.

Atrial fibrillation in ß-thalassemia Major Patients: Diagnosis, Management and Therapeutic Options.

The prevalence of atrial fibrillation (AFib) in ß-thalassemia major (ß-TM) patients has increased in the last few years, reaching up to 33.0%. Several factors may drive this value to even more in the next few years. We summarized the main challenges in the management and therapy of AFib in this very specific group of patients.

Neurological Consequences of Cardiac Arrhythmias: Relationship Between Stroke, Cognitive Decline, and Heart Rhythm Disorders.

Cardiac arrhythmias are one of the most common disorders with high morbidity and mortality. The effect of cardiac arrhythmias on the brain is very pronounced due to the high sensitivity of the brain to oxygen and blood supply. This mortality is preventable by early diagnosis and treatment which improves the patient's quality of life. Intervening at the right time, post arrhythmia is significant in preventing deaths and improving patient outcomes. Multiple pathophysiological mechanisms are studied for the brain-axis implications, that have the potential to be targeted by novel therapies. In this review, we describe the pathophysiological mechanisms and recent advances in detail to understand the functional aspects of the brain-heart axis and neurological implications post-stroke, caused by cardiac disorders.  This paper aims to discuss the current literature on the neurological consequences of cardiac arrhythmias and delve into a deeper understanding of the brain-heart axis, imbalances, and decline, with the aim of summarizing everything and all about the neurological consequences of cardiac arrhythmias.

A Retrospective Review of 30-Day Hospital Readmission Risk After Open Heart Surgery in Patients With Atrial Fibrillation.

Introduction Readmission rates after open heart surgery (OHS) remain an important clinical issue. The causes are varied, with identifying risk factors potentially providing valuable information to reduce healthcare costs and the rate of post-operative complications. This study aimed to characterize the reasons for 30-day hospital readmission rates of patients after open heart surgery. Methods All patients over 18 years of age undergoing OHS at a community hospital from January 2020 through December 2020 were identified. Demographic data, medical history, operative reports, post-operative complications, and telehealth interventions were obtained through chart review. Descriptive statistics and readmission rates were calculated, along with a logistic regression model, to understand the effects of medical history on readmission. Results A total of 357 OHS patients met the inclusion criteria for the study. Within the population, 8.68% of patients experienced readmission, 10.08% had an emergency department (ED) visit, and 95.80% had an outpatient office visit. A history of atrial fibrillation (AFib) significantly predicted 30-day hospital readmissions but not ED or outpatient office visits. Telehealth education was delivered to 66.11% of patients. Conclusion The study investigated factors associated with 30-day readmission following OHS. AFib patients were more likely to be readmitted than patients without atrial fibrillation. No other predictors of readmission, ED visits, or outpatient office visits were found. Patients reporting symptoms of tachycardia, pain, dyspnea, or "other" could be at increased risk for readmission.

The influence of atrial flutter in automated detection of atrial arrhythmias - are we ready to go into clinical practice?"

OBJECTIVE: To investigate the impact of atrial flutter (Afl) in the atrial arrhythmias classification task. We additionally advocate the use of a subject-based split for future studies in the field in order to avoid within-subject correlation which may lead to over-optimistic inferences. Finally, we demonstrate the effectiveness of the classifiers outside of the initially studied circumstances, by performing an inter-dataset model evaluation of the classifiers in data from different sources. METHODS: ECG signals of two private and three public (two MIT-BIH and Chapman ecgdb) databases were preprocessed and divided into 10s segments which were then subject to feature extraction. The created datasets were divided into a training and test set in two ways, based on a random split and a patient split. Classification was performed using the XGBoost classifier, as well as two benchmark classification models using both data splits. The trained models were then used to make predictions on the test data of the remaining datasets. RESULTS: The XGBoost model yielded the best performance across all datasets compared to the remaining benchmark models, however variability in model performance was seen across datasets, with accuracy ranging from 70.6% to 89.4%, sensitivity ranging from 61.4% to 76.8%, and specificity ranging from 87.3% to 95.5%. When comparing the results between the patient and the random split, no significant difference was seen in the two private datasets and the Chapman dataset, where the number of samples per patient is low. Nonetheless, in the MIT-BIH dataset, where the average number of samples per patient is approximately 1300, a noticeable disparity was identified. The accuracy, sensitivity, and specificity of the random split in this dataset of 93.6%, 86.4%, and 95.9% respectively, were decreased to 88%, 61.4%, and 89.8% in the patient split, with the largest drop being in Afl sensitivity, from 71% to 5.4%. The inter-dataset scores were also significantly lower than their intra-dataset counterparts across all datasets. CONCLUSIONS: CAD systems have great potential in the assistance of physicians in reliable, precise and efficient detection of arrhythmias. However, although compelling research has been done in the field, yielding models with excellent performances on their datasets, we show that these results may be over-optimistic. In our study, we give insight into the difficulty of detection of Afl on several datasets and show the need for a higher representation of Afl in public datasets. Furthermore, we show the necessity of a more structured evaluation of model performance through the use of a patient-based split and inter-dataset testing scheme to avoid the problem of within-subject correlation which may lead to misleadingly high scores. Finally, we stress the need for the creation and use of datasets with a higher number of patients and a more balanced representation of classes if we are to progress in this mission.

3D transesophageal echocardiography assists in evaluating the morphology, function, and presence of thrombi of left atrial appendage in patients with atrial fibrillation.

BACKGROUND: Left atrial appendage (LAA) is significantly more likely to form thrombi in patients with atrial fibrillation (AFib). Two-dimensional transesophageal echocardiography (2D TEE) is considered the gold standard for assessing and studying LAA morphology and anatomy. However, 2D TEE can only visualize one plane at any given time. Real-time three-dimensional echocardiography (RT-3D TEE) imaging can preserve spatial and temporal resolution, which is a safe, accurate, and reproducible imaging modality. There are few reports of the usage of RT-3D TEE to study LAA in AFib patients. In our research, RT-3D TEE helps to provide detailed LAA information and identifying the presence or absence of thrombi from pectinate muscles in paroxysmal and long-standing AFib patients. METHODS: LAA morphology was analyzed in detail by 2D TEE and RT-3D TEE in 320 patients with paroxysmal or long-standing AFib. The LAA flow pattern, as maximal LAA emptying flow velocity (LAAeV), was retrieved from 2D and 3D TEE imaging. LAA morphological parameters, spontaneous echo contrast (SEC), and thrombi were also detected by 2D and 3D TEE in all patients. In addition, LAA lobes and types were classified according to the morphology by 3D TEE, and the relationship between LAA types and the incidence of thrombi was evaluated. RESULTS: Long-standing AFib had greater enlargement of LAAs (orifice diameters and area), significantly more severe SEC, and a higher thrombi clot incidence rate by 3D-TEE compared with paroxysmal AFib patients (P<0.05). In addition, cauliflower morphology in long-standing AFib patients was associated with a higher LAA thrombus (OR 2.1, 95% CI: 1.1-8.5, P=0.031) and increased prevalence of SEC. Moreover, the uncertainty of thrombi detection was significantly decreased by 3D TEE compared with 2D TEE (P<0.001), and the certainty of thrombi detection by 3D TEE also decreased slightly (P=0.06). CONCLUSIONS: RT-3D TEE is a safe and real-time option for the evaluation of LAA morphology and function. Long-standing AFib has greater LAA and SEC, as well as a higher incidence of thrombi than the paroxysmal group. Cauliflower LAA type was associated with a higher prevalence of SEC and thrombi.

Evaluating the cost-effectiveness of catheter ablation of atrial fibrillation.

BACKGROUND: The pursuit of a clearer understanding of the pathogenesis of atrial fibrillation (AFib) and the development of new technology has resulted in a surge of interest in the surgical ablation for AFib. Here, we report our 8-year experience in the surgical treatment and management of AFib alongside, evaluating the cost-effectiveness in southern Mainland China over a 1-year follow-up. METHODS: Data of 3,068 patients from March 2011 through June 2019 was retrospectively extracted from The Provincial National Cardiac Database of Xiangya Second Hospital. The activities considered (and costs calculated) were outpatient consultations, hospital admissions, and drug treatment. Quality of life (QoL) questionnaires were also carried out to assess whether concomitant AFib correction procedures increase risk in patients, or improve patient's QoL. RESULTS: A total of 3,068 patients completed the questionnaires at a minimum of one time-point during the follow-up. The total cost was combined to obtain incremental costs per quality-adjusted life-years (QALYs). The total costs of the AFib catheter ablation group were remarkably higher compared to surgery as usual group. The incremental cost-effectiveness ratio was $76,513,227 (¥542,287,667) per QALY, with an acceptability line graph for cost at 43%. CONCLUSIONS: AFib is an extraordinarily costly and worrisome public health problem. Precision medicine is vital as it provides a platform for the clinical translation of targeted interventions that are designed to help treat and prevent AFib. Thus, to improve the QoL expectancy outcome(s), both therapeutic and surgical interventions should be aimed at addressing the underlying heart disease rather than restoring sinus rhythm.

Comparing rates of atrioesophageal fistula with contact force-sensing and non-contact force-sensing catheters: analysis of post-market safety surveillance data.

PURPOSE: There is limited data on the specific incidence of serious adverse events, such as atrioesophageal fistula (AEF), associated with either contact force (CF) or non-CF ablation catheters. Since the actual number of procedures performed with each type of catheter is unknown, making direct comparisons is difficult. The purpose of this study was to assess the incidence of AEF associated with the use of CF and non-CF catheters. Additionally, we aimed to understand the workflow present in confirmed AEF cases voluntarily provided by physicians. METHODS: The number of AEFs for 2014-2017 associated with each type of catheter was extracted from an ablation device manufacturer's complaint database. Proprietary device sales data, a proxy for the total number of procedures, were used as the denominator to calculate the incidence rates. Additional survey and workflow data were systematically reviewed. RESULTS: Both CF and non-CF ablation catheters have comparably low incidence of AEF (0.006 ± 0.003% and 0.005 ± 0.003%, respectively, p = 0.69). CF catheters are the catheter of choice for left atrium (LA) procedures which pose the greatest risk for AEF injury. Retrospective analysis of seven AEF cases demonstrated that high power and force and long RF duration were delivered on the posterior wall of the left atrium in all cases. CONCLUSIONS: CF and non-CF ablation catheters were found to have similar AEF incidence, despite CF catheters being the catheter of choice for LA procedures. More investigation is needed to understand the range of parameters which may create risk for AEF.

The burden of cardiac arrhythmias in sarcoidosis: a population-based inpatient analysis.

BACKGROUND: Cardiac involvement in the sarcoidosis is known to ensue with diverse clinical forms and its investigation is challenging at times. This article features the under-perceived burden, patterns, and outcomes of different arrhythmias, which may have a prognostic significance in patients with sarcoidosis. METHODS: We queried the National Inpatient Sample (NIS) for 2010-2014 to recognize sarcoidosis, arrhythmia, and comorbidities affecting hospitalizations. The nationwide estimates were attained using discharge records. We assessed incidence and trends in sarcoidosis-related arrhythmia and consequential inpatient mortality, hospital length of stay (LOS), hospitalization charges and predictors of mortality with multivariate analysis. RESULTS: We identified 369,285 sarcoidosis-related hospitalizations. Of these, nearly one-fifth suffered from arrhythmias (n=73,424). The sarcoidosis patients developing arrhythmias were older (61.9 vs. 56.0 years) compared to those without. Males had the higher incidence of arrhythmias compared to females. Atrial fibrillation (Afib) (10.97%) was the most common subtype, followed by ventricular tachycardia (1.97%). There was a rising trend in arrhythmia-related hospital admissions and mortality among sarcoidosis, with Afib incidence displaying the highest increase. Traditional cardiac comorbidities were higher in the sarcoid-arrhythmia group. The arrhythmia group had significantly higher mortality (3.7% vs. 1.5%), mean hospital LOS (6.4 vs. 5.2 days) and hospital charges ($64,118 vs. $41,565) compared to non-arrhythmia group (P<0.001). Incident arrhythmia significantly increased the mortality odds in sarcoidosis (adjusted odds ratio, 2.06). CONCLUSIONS: The growing trend, deteriorating outcomes and higher mortality associated with sarcoid-related arrhythmias highlight the importance of timely diagnosis and aggressive management in this population.