Unsupervised Transformer-Based Anomaly Detection in ECG Signals

被引:17
作者
Alamr, Abrar [1 ]
Artoli, Abdelmonim [1 ]
机构
[1] King Saud Univ, Coll Comp & Informat Sci, Comp Sci Dept, Riyadh 11543, Saudi Arabia
关键词
unsupervised transformers; deep learning; anomaly detection; ECG signal; TIME-SERIES;
D O I
10.3390/a16030152
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Anomaly detection is one of the basic issues in data processing that addresses different problems in healthcare sensory data. Technology has made it easier to collect large and highly variant time series data; however, complex predictive analysis models are required to ensure consistency and reliability. With the rise in the size and dimensionality of collected data, deep learning techniques, such as autoencoder (AE), recurrent neural networks (RNN), and long short-term memory (LSTM), have gained more attention and are recognized as state-of-the-art anomaly detection techniques. Recently, developments in transformer-based architecture have been proposed as an improved attention-based knowledge representation scheme. We present an unsupervised transformer-based method to evaluate and detect anomalies in electrocardiogram (ECG) signals. The model architecture comprises two parts: an embedding layer and a standard transformer encoder. We introduce, implement, test, and validate our model in two well-known datasets: ECG5000 and MIT-BIH Arrhythmia. Anomalies are detected based on loss function results between real and predicted ECG time series sequences. We found that the use of a transformer encoder as an alternative model for anomaly detection enables better performance in ECG time series data. The suggested model has a remarkable ability to detect anomalies in ECG signal and outperforms deep learning approaches found in the literature on both datasets. In the ECG5000 dataset, the model can detect anomalies with 99% accuracy, 99% F1-score, 99% AUC score, 98.1% recall, and 100% precision. In the MIT-BIH Arrhythmia dataset, the model achieved an accuracy of 89.5%, F1 score of 92.3%, AUC score of 93%, recall of 98.2%, and precision of 87.1%.
引用
收藏
页数:15
相关论文
共 45 条
  • [1] Abadi Martin, 2016, arXiv
  • [2] [Anonymous], 1984, Wiley Series in Probability and Mathematical Statistics
  • [3] USAD : UnSupervised Anomaly Detection on Multivariate Time Series
    Audibert, Julien
    Michiardi, Pietro
    Guyard, Frederic
    Marti, Sebastien
    Zuluaga, Maria A.
    [J]. KDD '20: PROCEEDINGS OF THE 26TH ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY & DATA MINING, 2020, : 3395 - 3404
  • [4] Ba J, 2016, ARXIV160706450
  • [5] IoT anomaly detection methods and applications: A survey
    Chatterjee, Ayan
    Ahmed, Bestoun S.
    [J]. INTERNET OF THINGS, 2022, 19
  • [6] An improved and provably secure privacy preserving authentication protocol for SIP
    Chaudhry, Shehzad Ashraf
    Naqvi, Husnain
    Sher, Muhammad
    Farash, Mohammad Sabzinejad
    ul Hassan, Mahmood
    [J]. PEER-TO-PEER NETWORKING AND APPLICATIONS, 2017, 10 (01) : 1 - 15
  • [7] A general framework for never-ending learning from time series streams
    Chen, Yanping
    Hao, Yuan
    Rakthanmanon, Thanawin
    Zakaria, Jesin
    Hu, Bing
    Keogh, Eamonn
    [J]. DATA MINING AND KNOWLEDGE DISCOVERY, 2015, 29 (06) : 1622 - 1664
  • [8] Learning Graph Structures With Transformer for Multivariate Time-Series Anomaly Detection in IoT
    Chen, Zekai
    Chen, Dingshuo
    Zhang, Xiao
    Yuan, Zixuan
    Cheng, Xiuzhen
    [J]. IEEE INTERNET OF THINGS JOURNAL, 2022, 9 (12) : 9179 - 9189
  • [9] Chen ZM, 2018, WIREL TELECOMM SYMP
  • [10] Chollet F., KERAS