Article
Medicine, General & Internal
P. Krisai, P. Hammerle, S. Blum, P. Meyre, S. Aeschbacher, P. Melchiorre-Mayer, O. Baretella, N. Rodondi, D. Conen, S. Osswald, M. Kuhne, C. S. Zuern
Summary: The presence of atrial fibrillation (AF) on a single surface ECG is significantly associated with increased risk of mortality and hospitalizations for congestive heart failure in AF patients, suggesting that these patients are a high-risk group that may benefit from intensified treatment.
JOURNAL OF INTERNAL MEDICINE
(2021)
Article
Cardiac & Cardiovascular Systems
Anatoly Langer, Jeff S. Healey, F. Russell Quinn, George Honos, Isabelle Nault, Mary Tan, Diane Camara, David M. Newman, Richard Godin
Summary: This study identified a number of undiagnosed AF cases in an asymptomatic but at-risk population. The predictive factors for undiagnosed AF included history of perioperative AF, age over 85, and absence of cardiovascular disease. Further exploration of these variables may be beneficial in identifying potential AF patients for screening.
INTERNATIONAL JOURNAL OF CARDIOLOGY
(2021)
Article
Multidisciplinary Sciences
Hyo-Chang Seo, Seok Oh, Hyunbin Kim, Segyeong Joo
Summary: This study examined the data dependency of a deep learning-based AF detection algorithm using Resnet models trained on different databases. Results showed a decrease in data dependency as the amount of training data increased, with accuracies ranging from 98-99% within the same dataset to 53-92% on external datasets from different sources.
SCIENTIFIC REPORTS
(2021)
Article
Engineering, Electrical & Electronic
Mohamed Abdelazez, Sreeraman Rajan, Adrian D. C. Chan
Summary: Atrial fibrillation (AF) is a serious cardiovascular condition with potential complications such as stroke, heart attack, and death. Compressive sensing techniques can help reduce the requirements of continuous monitoring. The study proposed an AF detector using a deterministic compressively sensed ECG and achieved good performance in detecting AF.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Multidisciplinary Sciences
Cedric Gilon, Jean-Marie Gregoire, Marianne Mathieu, Stephane Carlier, Hugues Bersini
Summary: This article presents a new Holter monitoring database consisting of 167 records from 152 patients with paroxysmal AF, collected from an outpatient cardiology clinic in Belgium. The dataset provides 24 million seconds of annotated Holter monitoring data, sampled at 200 Hz, and offers a valuable resource for researchers in the field of cardiac arrhythmia diagnosis using machine learning and deep learning.
Article
Medicine, General & Internal
Yong Wei, Genqing Zhou, Xiaoyu Wu, Xiaofeng Lu, Xingjie Wang, Bin Wang, Caihong Wang, Yahong Shen, Shi Peng, Yu Ding, Juan Xu, Lidong Cai, Songwen Chen, Wenyi Yang, Shaowen Liu
Summary: This study investigated the incidence and predictors of atrial fibrillation (AF) in individuals aged over 60 years in China. The results showed an incidence rate of 5.2/1000 person-years, and age, gender, hypertension history, cardiac disease history, and various ECG abnormalities were independently associated with AF incidence.
CHINESE MEDICAL JOURNAL
(2023)
Article
Engineering, Electrical & Electronic
Nabasmita Phukan, M. Sabarimalai Manikandan, Ram Bilas Pachori
Summary: This paper explores a lightweight convolutional neural network (CNN) based AFibri event detector considering limited resource-constraints of medical devices and advanced deep learning networks. The study presents extensive evaluation results of different CNN-AFibri models with various model parameters and ECG segment durations. Realtime implementation of the best CNN based method is demonstrated using the Raspberry Pi computing platform, showing promising performance compared to existing methods on validation databases.
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-REGULAR PAPERS
(2023)
Article
Biology
Min-Uk Yang, Dae-In Lee, Seung Park
Summary: Atrial fibrillation (AF) is a common arrhythmia that poses a significant burden on healthcare systems globally. Early detection is crucial for timely treatment and prevention of complications. This study presents a new diagnostic approach using a component-aware transformer to analyze electrocardiograms.
COMPUTERS IN BIOLOGY AND MEDICINE
(2022)
Article
Medicine, General & Internal
Philipp Krisai, Laurent Roten, Ivan Zeljkovic, Nikola Pavlovic, Peter Ammann, Tobias Reichlin, Eric Auf Der Maur, Olivia Streicher, Sven Knecht, Michael Kuhne, Stefan Osswald, Jan Novak, Christian Sticherling
Summary: The study found a high proportion of post-ablation AF in cAFL patients during standardized follow-up, indicating the need for consideration when deciding on long-term oral anticoagulation.
JOURNAL OF CLINICAL MEDICINE
(2021)
Article
Clinical Neurology
Achim Leo Burger, Cornelia Roesler, Johanna Ebner, Peter Sommer, Sebastian Mutzenbach, Walther-Benedikt Winkler, Franz Weidinger, Robin Ristl, Thomas Pezawas, Stefan Greisenegger
Summary: This study found that gapless electrocardiogram (ECG) monitoring significantly increased the detection rate of atrial fibrillation (AF) in patients with ischemic stroke. Patients received continuous ECG monitoring during stroke-unit admission until implantation of an insertable cardiac monitor (ICM). The results showed that 17 patients (15.5%) were newly diagnosed with AF through ICM monitoring, compared to only one patient (0.9%) diagnosed through 24-72-hour Holter ECG monitoring during follow-up.
EUROPEAN JOURNAL OF NEUROLOGY
(2023)
Review
Computer Science, Information Systems
Igor Matias, Nuno Garcia, Sandeep Pirbhulal, Virginie Felizardo, Nuno Pombo, Henriques Zacarias, Miguel Sousa, Eftim Zdravevski
Summary: This study reviewed articles published in the last ten years on predicting atrial fibrillation using artificial intelligence. It found that deep learning techniques can improve accuracy, but are not as widely used as expected. The research also revealed that the field of AI for prediction of AF is still in its early stages, with high potential for further study.
COMPUTER SCIENCE REVIEW
(2021)
Review
Physiology
Andrew S. Tseng, Peter A. Noseworthy
Summary: Machine learning techniques have garnered significant interest in predicting and screening atrial fibrillation due to their ability to utilize clinical data for accurate predictions and screenings, showcasing the potential of artificial intelligence in cardiovascular medicine.
FRONTIERS IN PHYSIOLOGY
(2021)
Article
Health Care Sciences & Services
Yating Hu, Tengfei Feng, Miao Wang, Chengyu Liu, Hong Tang
Summary: A deep learning model was used to accurately detect atrial fibrillation, including both AF and AFL. The model had the ability to discriminate AF from normal rhythm and to detect its onset and offset. Tests on four public datasets validated the effectiveness of the proposed method, with the best performance achieving an accuracy of 98.67%, a sensitivity of 87.69%, and a specificity of 98.56%.
JOURNAL OF PERSONALIZED MEDICINE
(2023)
Article
Chemistry, Analytical
Soonil Kwon, So-Ryoung Lee, Eue-Keun Choi, Hyo-Jeong Ahn, Hee-Seok Song, Young-Shin Lee, Seil Oh
Summary: This study compared 24-hour ECG monitoring between adhesive single-lead and Holter devices in patients with general arrhythmia, showing a high degree of agreement in diagnostics and monitoring parameters between the two devices, suggesting that the single-lead device could be an acceptable alternative for ambulatory ECG monitoring in patients with general arrhythmia.
Article
Biochemical Research Methods
Cai Wu, Maxwell Hwang, Tian-Hsiang Huang, Yen-Ming J. Chen, Yiu-Jen Chang, Tsung-Han Ho, Jian Huang, Kao-Shing Hwang, Wen-Hsien Ho
Summary: This study combined P-wave morphology parameters and heart rate variability parameters for model training, highlighting the value of this parameter combination in improving prediction accuracy. By using a hybrid Taguchi-genetic algorithm to calculate P-wave morphology parameters and utilizing the Stacking ensemble learning method for model training, the accuracy of the prediction model was significantly increased, leading to better early prediction of atrial fibrillation.
BMC BIOINFORMATICS
(2021)