Abnormal ECG detection based on an adversarial autoencoder

Shan, Lishen; Liu, Yu; Jiang, Hua; Zhou, Peng; Niu, Jing; Liu, Ran; Wang, Yuanyuan; Jiao, Peng; Yu, Hao; Sha, Xianyi; Chang, Shijie

doi:10.3389/fphys.2022.961724

Cited by 18 publications

(5 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This network exhibits excellent interpretability 7 . The autoencoder network has garnered significant attention from researchers in the field of ECG classification and detection, with the improved model based on this network being widely utilized for ECG classification tasks 8 .…”

Section: Related Workmentioning

confidence: 99%

ECG autoencoder based on low-rank attention

Zhang,

Fang,

Ren

2024

Sci Rep

View full text Add to dashboard Cite

The prevalence of cardiovascular disease (CVD) has surged in recent years, making it the foremost cause of mortality among humans. The Electrocardiogram (ECG), being one of the pivotal diagnostic tools for cardiovascular diseases, is increasingly gaining prominence in the field of machine learning. However, prevailing neural network models frequently disregard the spatial dimension features inherent in ECG signals. In this paper, we propose an ECG autoencoder network architecture incorporating low-rank attention (LRA-autoencoder). It is designed to capture potential spatial features of ECG signals by interpreting the signals from a spatial perspective and extracting correlations between different signal points. Additionally, the low-rank attention block (LRA-block) obtains spatial features of electrocardiogram signals through singular value decomposition, and then assigns these spatial features as weights to the electrocardiogram signals, thereby enhancing the differentiation of features among different categories. Finally, we utilize the ResNet-18 network classifier to assess the performance of the LRA-autoencoder on both the MIT-BIH Arrhythmia and PhysioNet Challenge 2017 datasets. The experimental results reveal that the proposed method demonstrates superior classification performance. The mean accuracy on the MIT-BIH Arrhythmia dataset is as high as 0.997, and the mean accuracy and $$F_1$$ F 1 -score on the PhysioNet Challenge 2017 dataset are 0.850 and 0.843.

show abstract

Section: Related Workmentioning

confidence: 99%

ECG autoencoder based on low-rank attention

Zhang,

Fang,

Ren

2024

Sci Rep

View full text Add to dashboard Cite

show abstract

“…UMAP has been used extensively in all domains to reduce dimensionality for various purposes including interpretability, primarily for feature selection in context of ECG [28], [19], [15], [35] and heart sound classification [4], [5]. While various anomaly detection techniques have been used, there is limited use of Local Outlier Filtering (LOF) [37], [33], [16]. In our literature search, we found only a single work exploring anomaly detection with UMAP in an unsupervised approach unlike our supervised multiview approach [10] on non-healthcare acoustic scenes.…”

Section: Related Workmentioning

confidence: 99%

Multiview Outlier Filtered Pediatric Heart Sound Classification

Dakshit

2024

ijacsa

View full text Add to dashboard Cite

The advancements in deep learning has generated a large-scale interest in development of black-box models for various use cases in different domains such as healthcare, in both at-home and critical setting for diagnosis and monitoring of various health conditions. The use of audio signals as a view for diagnosis is nascent and the success of deep learning models in ingesting multimedia data provides an opportunity for use as a diagnostic medium. For the widespread use of these decision support systems, it is prudent to develop high performing systems which require large quantities of data for training and low-cost method of data collection making it more accessible for developing regions of the world and general population. Data collected from low-cost collection especially wireless devices are prone to outliers and anomalies. The presence of outliers skews the hypothesis space of the model and leads to model drift on deployment. In this paper, we propose a multiview pipeline through interpretable outlier filtering on the small Mendeley Children Heart Sound dataset collected using wireless low-cost digital stethoscope. Our proposed pipeline explores and provides dimensionally reduced interpertable visualizations for functional understanding of the effect of various outlier filtering methods on deep learning model hypothesis space and fusion strategies for multiple views of heart sound data namely raw time-series signal and Mel Frequency Cepstrum Coefficients achieving 98.19% state-of-the-art testing accuracy.

show abstract

“…The model is tested using the 1-lead ECG5000 dataset, achieving an accuracy of 98.42% with precision, recall and F1 values of 94.52%, 98.22% and 96.33%, respectively. Shan et al [23] proposed a framework to perform the ECG anomaly detection based on an adversarial AE and temporal convolutional network (TCN) which consist of three modules: an AE, the discriminator, and an outlier detector. The whole framework is evaluated using ECG signals from two different datasets: the MIT-BIH arrhythmia dataset (obtaining an accuracy of 96.76%) and CMUH dataset (with an accuracy 93.58%).…”

Section: Background and Related Workmentioning

confidence: 99%

AUTAN-ECG: An AUToencoder bAsed system for anomaly detectioN in ECG signals

Lomoio,

Vizza,

Giancotti

et al. 2023

Preprint

View full text Add to dashboard Cite

<p>Electrocardiographic (ECG) signals that monitor heart activity can help identifying disease-related anomalies. Reliable automatic anomaly detection has been shown to be useful in supporting physicians in reading ECG signals. Decision support systems may be useful in such cases but their reliability can be guaranteed. </p> <p>Autoencoders (AEs) have been extensively used to analyse signals in many fields. Convolutional Autoencoders (CAE) are a particular class of AE showing optimal performances in detecting signal anomalies. Thus, CAEs can be used to support and automatise the task of anomaly detection. We design and use a CAE-based system to detect anomalies in ECG signals to support cardiologists in identifying anomalies related to possible diseases. Our tool outperforms other state-of-the-art ECG anomaly detection approaches tested on a real dataset. In the task of anomaly detection, our CAE obtains a ROC AUC of 97.82% with a simulated test set and a ROC AUC of 99.75% using on a real test set. The tool and the source code are available at <a href="https://github.com/UgoLomoio/ECG_DSS_CAE" target="_blank">https://github.com/UgoLomoio/EG_DSS_CAE</a>.</p>

show abstract

Abnormal ECG detection based on an adversarial autoencoder

Cited by 18 publications

References 41 publications

ECG autoencoder based on low-rank attention

ECG autoencoder based on low-rank attention

Multiview Outlier Filtered Pediatric Heart Sound Classification

AUTAN-ECG: An AUToencoder bAsed system for anomaly detectioN in ECG signals

Contact Info

Product

Resources

About