Learning ECG Representations for Multi-Label Classification of Cardiac Abnormalities

Suh, Jangwon; Kim, Jimyeong; Lee, Eunjung; Kim, Jaeill; Hwang, Duhun; Lee, Junghoon; Park, Jaeseung; Moon, Seo-Yoon; Kim, Yeonsu; Kang, Min Woo; Kwon, Soonil; Choi, Eue‐Keun; Rhee, Wonjong

doi:10.23919/cinc53138.2021.9662753

Cited by 7 publications

(4 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The challenge winner, ISIBrno-AIMT team, used an ensemble of residual networks approach that achieved 97.1% AUROC and an 83% F1-score when using only the leads I and II [15]. Considering only the AF class (and two lead approaches), the snu adsl team achieved the best AUROC (98.0%) and F1-score (88.3%) by using representational learning and an EfficientNet-B3 model [13]. LiteVGG-11 has a lower AUROC than both models but a higher F1-score.…”

Section: Resultsmentioning

confidence: 99%

“…However, the computational complexity of most classification models, particularly those based on deep learning algorithms, is one of the challenges for using wearable devices in continuous rhythm monitoring. To improve accuracy, these models frequently use deeper and more complex models [13,14] or ensembles of models [15], usually without taking into account the energy consumption or computational limitations of portable devices [16].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Lightweight Unidimensional Deep Learning Model for Atrial Fibrillation Detection

Soares¹,

Monteiro²,

Jatene³

et al. 2022

Computing in Cardiology Conference (CinC)

View full text Add to dashboard Cite

Continuous rhythm monitoring using wearable devices is a potential tool for early identification of atrial fibrillation (AF), the most frequent cardiac arrhythmia (with 0,51% worldwide prevalence, increasing with time), and is also a tool for remote monitoring patients after cardiac surgery. However, AF detection directly through wearable devices is limited by the computational complexity of the classifier model.In this work we propose a lightweight AF classifier model based on the VGG-11 architecture (LiteVGG-11), focusing on reducing the number of parameters and numerical operations. Using a low number of filters, depthwise separable convolution, and global pooling, this model has only 20,454 parameters and needs 6.9 MFLOP to make an inference for an input of 10 seconds of the ECG leads I and II, sampled at 200 Hz.To test its effectiveness for AF detection we used the PhysioNet/CinC Challenge 2021 public dataset, stratifying the classes into sinus rhythm, AF, and other rhythms. After 10 Monte Carlo cross-validation splits, with 24,260 unbalanced samples for training and 1,536 balanced samples for validation and testing, the observed metrics (mean±standard deviation) were: Se 94.1±0.1%; Sp 91.9±0.8%; F1-Score 89.50.7±%; and AUC 96.1±0.6%.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

A Lightweight Unidimensional Deep Learning Model for Atrial Fibrillation Detection

Soares¹,

Monteiro²,

Jatene³

et al. 2022

Computing in Cardiology Conference (CinC)

View full text Add to dashboard Cite

show abstract

“…Using a dataset in an unsupervised manner, the models are first trained, and then, it is tried to fit them correctly in a supervised way. While the main focus of this study is on the application of self-supervised learning for effective ECG learning, we investigate several other aspects in order to increase efficiency [10]. Several techniques, including knowledge-based properties and supervised pre-training, are employed in order to achieve the maximum accuracy and stability of heartbeat classification in the context of a weak supervision [11].…”

Section: Related Literature 21 Deep Learning Tasks For Ecg Signal Cla...mentioning

confidence: 99%

Classification of electrocardiogram signals based on federated learning and a gaussian multivariate aggregation module

Meqdad

Hussein²,

Husain

et al. 2023

IJEECS

View full text Add to dashboard Cite

Categorization of cardiac abnormalities received from several centers is not possible within the quickest time because of privacy and security restrictions. Today, individuals’ security problem is considered as one of the most important research fields in most research sciences. This study provides a novel approach for detection of cardiac abnormalities based on federated learning (FL). This approach addresses the challenge of accessing data from remote centers and presents the possibility of learning without the need for transferring data from the main center. We present a novel aggregation approach in the FL for addressing the challenge of imbalanced data using the averaging stochastic weights (SWA) optimizer and a multivariate Gaussian in order to make a better and more accurate detection possible. The advantage of the present proposed approach is robust and secure aggregation for unbalanced electrocardiogram (ECG) data from heterogeneous clients. We were able to achieve 87.98% accuracy in testing with the robust VGG19 architecture.

show abstract

“…For example, Biomedic2ai segmented the signals into 5-second windows with a stride of one second for a 4-second overlap for adjacent signals (Clark et al 2021). snu_adsl selected a random window with a width of 13.3 seconds and zero-padded ECG signals shorter than 13.3 seconds at the end of the signal (Suh et al 2021). prna set a fixed window width of 15.36 seconds, allowing the signal to be split into divisible segments sizes and zero-padding the ends of the signals as needed (Natarajan et al 2021).…”

Section: Rank Team [Reference]mentioning

confidence: 99%

Issues in the automated classification of multilead ecgs using heterogeneous labels and populations

et al. 2022

View full text Add to dashboard Cite

Objective: The standard twelve-lead electrocardiogram (ECG) is a widely used tool for monitoring cardiac function and diagnosing cardiac disorders. The development of smaller, lower-cost, and easier-to-use ECG devices may improve access to cardiac care in lower-resource environments, but the diagnostic potential of these devices is unclear. This work explores these issues through a public competition: the 2021 PhysioNet Challenge. In addition, we explore the potential for performance boosting through a meta-learning approach. Approach: We sourced 131,149 twelve-lead ECG recordings from ten international sources. We posted 88,253 annotated recordings as public training data and withheld the remaining recordings as hidden validation and test data. We challenged teams to submit containerized, open-source algorithms for diagnosing cardiac abnormalities using various ECG lead combinations, including the code for training their algorithms. We designed and scored algorithms using an evaluation metric that captures the risks of different misdiagnoses for 30 conditions. After the Challenge, we implemented a semi-consensus voting model on all working algorithms. Main results: A total of 68 teams submitted 1,056 algorithms during the Challenge, providing a variety of automated approaches from both academia and industry. The performance differences across the different lead combinations were smaller than the performance differences across the different test databases, showing that generalizability posed a larger challenge to the algorithms than the choice of ECG leads. A voting model improved performance by 3.5%. Significance: The use of different ECG lead combinations allowed us to assess the diagnostic potential of reduced-lead ECG recordings, and the use of different data sources allowed us to assess the generalizability of algorithms to diverse institutions and populations. The submission of working, open-source code for both training and testing and the use of a novel evaluation metric improved the reproducibility, generalizability, and applicability of the research conducted during the Challenge.

show abstract

Learning ECG Representations for Multi-Label Classification of Cardiac Abnormalities

Cited by 7 publications

References 3 publications

A Lightweight Unidimensional Deep Learning Model for Atrial Fibrillation Detection

A Lightweight Unidimensional Deep Learning Model for Atrial Fibrillation Detection

Classification of electrocardiogram signals based on federated learning and a gaussian multivariate aggregation module

Issues in the automated classification of multilead ecgs using heterogeneous labels and populations

Contact Info

Product

Resources

About