Deep Learning Approach for Highly Specific Atrial Fibrillation and Flutter Detection based on RR Intervals
Нема приказа
Аутори
Ivanović, Marija D.Atanasoski, Vladimir
Shvilkin, Alexei
Hadžievski, Ljupčo
Maluckov, Aleksandra
Конференцијски прилог (Објављена верзија)
,
© 2019 IEEE
Метаподаци
Приказ свих података о документуАпстракт
Atrial fibrillation (AF) and atrial flutter (AFL) represent atrial arrhythmias closely related to increasing risk for embolic stroke, and therefore being in the focus of cardiologists. While the reported methods for AF detection exhibit high performances, little attention has been given to distinguishing these two arrhythmias. In this study, we propose a deep neural network architecture, which combines convolutional and recurrent neural networks, for extracting features from sequence of RR intervals. The learned features were used to classify a long term ECG signals as AF, AFL or sinus rhythm (SR). A 10-fold cross-validation strategy was used for choosing an architecture design and tuning model hyperparameters. Accuracy of 88.28 %, with the sensitivities of 93.83%, 83.60% and 83.83% for SR, AF and AFL, respectively, was achieved. After choosing optimal network structure, the model was trained on the entire training set and finally evaluated on the blindfold test set which resulted in 8...9.67% accuracy, and 97.20%, 94.20%, and 77.78% sensitivity for SR, AF and AFL, respectively. Promising performances of the proposed model encourage continuing development of highly specific AF and AFL detection procedure based on deep learning. Distinction between these two arrhythmias can make therapy more efficient and decrease the recovery time to normal heart rhythm. © 2019 IEEE.
Извор:
Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) (41; 2019; Berlin), 2019, 1780-1783Издавач:
- IEEE
Напомена:
- Conference of 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2019 ; Conference Date: 23 July 2019 Through 27 July 2019; Conference Code:152547
DOI: 10.1109/EMBC.2019.8856806
ISBN: 978-1-5386-1311-5
ISSN: 1557-170X
PubMed: 31946242
Scopus: 2-s2.0-85077878364
Колекције
Институција/група
VinčaTY - CONF AU - Ivanović, Marija D. AU - Atanasoski, Vladimir AU - Shvilkin, Alexei AU - Hadžievski, Ljupčo AU - Maluckov, Aleksandra PY - 2019 UR - https://vinar.vin.bg.ac.rs/handle/123456789/8805 AB - Atrial fibrillation (AF) and atrial flutter (AFL) represent atrial arrhythmias closely related to increasing risk for embolic stroke, and therefore being in the focus of cardiologists. While the reported methods for AF detection exhibit high performances, little attention has been given to distinguishing these two arrhythmias. In this study, we propose a deep neural network architecture, which combines convolutional and recurrent neural networks, for extracting features from sequence of RR intervals. The learned features were used to classify a long term ECG signals as AF, AFL or sinus rhythm (SR). A 10-fold cross-validation strategy was used for choosing an architecture design and tuning model hyperparameters. Accuracy of 88.28 %, with the sensitivities of 93.83%, 83.60% and 83.83% for SR, AF and AFL, respectively, was achieved. After choosing optimal network structure, the model was trained on the entire training set and finally evaluated on the blindfold test set which resulted in 89.67% accuracy, and 97.20%, 94.20%, and 77.78% sensitivity for SR, AF and AFL, respectively. Promising performances of the proposed model encourage continuing development of highly specific AF and AFL detection procedure based on deep learning. Distinction between these two arrhythmias can make therapy more efficient and decrease the recovery time to normal heart rhythm. © 2019 IEEE. PB - IEEE C3 - Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) (41; 2019; Berlin) T1 - Deep Learning Approach for Highly Specific Atrial Fibrillation and Flutter Detection based on RR Intervals SP - 1780 EP - 1783 DO - 10.1109/EMBC.2019.8856806 ER -
@conference{ author = "Ivanović, Marija D. and Atanasoski, Vladimir and Shvilkin, Alexei and Hadžievski, Ljupčo and Maluckov, Aleksandra", year = "2019", abstract = "Atrial fibrillation (AF) and atrial flutter (AFL) represent atrial arrhythmias closely related to increasing risk for embolic stroke, and therefore being in the focus of cardiologists. While the reported methods for AF detection exhibit high performances, little attention has been given to distinguishing these two arrhythmias. In this study, we propose a deep neural network architecture, which combines convolutional and recurrent neural networks, for extracting features from sequence of RR intervals. The learned features were used to classify a long term ECG signals as AF, AFL or sinus rhythm (SR). A 10-fold cross-validation strategy was used for choosing an architecture design and tuning model hyperparameters. Accuracy of 88.28 %, with the sensitivities of 93.83%, 83.60% and 83.83% for SR, AF and AFL, respectively, was achieved. After choosing optimal network structure, the model was trained on the entire training set and finally evaluated on the blindfold test set which resulted in 89.67% accuracy, and 97.20%, 94.20%, and 77.78% sensitivity for SR, AF and AFL, respectively. Promising performances of the proposed model encourage continuing development of highly specific AF and AFL detection procedure based on deep learning. Distinction between these two arrhythmias can make therapy more efficient and decrease the recovery time to normal heart rhythm. © 2019 IEEE.", publisher = "IEEE", journal = "Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) (41; 2019; Berlin)", title = "Deep Learning Approach for Highly Specific Atrial Fibrillation and Flutter Detection based on RR Intervals", pages = "1780-1783", doi = "10.1109/EMBC.2019.8856806" }
Ivanović, M. D., Atanasoski, V., Shvilkin, A., Hadžievski, L.,& Maluckov, A.. (2019). Deep Learning Approach for Highly Specific Atrial Fibrillation and Flutter Detection based on RR Intervals. in Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) (41; 2019; Berlin) IEEE., 1780-1783. https://doi.org/10.1109/EMBC.2019.8856806
Ivanović MD, Atanasoski V, Shvilkin A, Hadžievski L, Maluckov A. Deep Learning Approach for Highly Specific Atrial Fibrillation and Flutter Detection based on RR Intervals. in Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) (41; 2019; Berlin). 2019;:1780-1783. doi:10.1109/EMBC.2019.8856806 .
Ivanović, Marija D., Atanasoski, Vladimir, Shvilkin, Alexei, Hadžievski, Ljupčo, Maluckov, Aleksandra, "Deep Learning Approach for Highly Specific Atrial Fibrillation and Flutter Detection based on RR Intervals" in Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC) (41; 2019; Berlin) (2019):1780-1783, https://doi.org/10.1109/EMBC.2019.8856806 . .