Abnormal Electrocardiogram Signal Detection Based on the BiLSTM Network
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
The health of the human heart is commonly measured using ECG (Electrocardiography) signals. To identify any anomaly in the human heart, the time-sequence of ECG signals is examined manually by a cardiologist or cardiac electrophysiologist. Lightweight anomaly detection on ECG signals in an embedded system is expected to be popular in the near future, because of the increasing number of heart disease symptoms. Some previous research uses deep learning networks such as LSTM and BiLSTM to detect anomaly signals without any handcrafted feature. Unfortunately, lightweight LSTMs show low precision and heavy LSTMs require heavy computing powers and volumes of labeled dataset for symptom classification. This paper proposes an ECG anomaly detection system based on two level BiLSTM for acceptable precision with lightweight networks, which is lightweight and usable at home. Also, this paper presents a new threshold technique which considers statistics of the current ECG pattern. This paper’s proposed model with BiLSTM detects ECG signal anomaly in 0.467 ~ 1.0 F1 score, compared to 0.426 ~ 0.978 F1 score of the similar model with LSTM except one highly noisy dataset.
Abstract
The health of the human heart is commonly measured using ECG (Electrocardiography) signals. To identify any anomaly in the human heart, the time-sequence of ECG signals is examined manually by a cardiologist or cardiac electrophysiologist. Lightweight anomaly detection on ECG signals in an embedded system is expected to be popular in the near future, because of the increasing number of heart disease symptoms. Some previous research uses deep learning networks such as LSTM and BiLSTM to detect anomaly signals without any handcrafted feature. Unfortunately, lightweight LSTMs show low precision and heavy LSTMs require heavy computing powers and volumes of labeled dataset for symptom classification. This paper proposes an ECG anomaly detection system based on two level BiLSTM for acceptable precision with lightweight networks, which is lightweight and usable at home. Also, this paper presents a new threshold technique which considers statistics of the current ECG pattern. This paper’s proposed model with BiLSTM detects ECG signal anomaly in 0.467 ~ 1.0 F1 score, compared to 0.426 ~ 0.978 F1 score of the similar model with LSTM except one highly noisy dataset.
참고문헌 (34)
- Lilly, Leonard S, "Pathophysiology of Heart Disease: A Collaborative Project of Medical Students and Faculty" (2016), Lippincott Williams & Wilkins
- Yama Y., "Development of a wireless capacitive sensor for ambulatory ECG monitoring over clothes", Proceedings of the 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBS ’07), pp.5727~5730 (2007)
- M. Basseville, "Detection of abrupt changes: theory and application, vol. 104" (1993), prentice Hall
- S. Hochreiter, "Long Short-Term Memory", Neural Comput, 9(8), pp.1735~1780 (1997)[DOI]
- L. Medsker, "Recurrent neural networks: design and applications" (1999), CRC press
- H. Sanchez, "Anomaly detection in streaming time series based on bounding boxes", presented at the International Conference on Similarity Search and Applications (2014), Springer
- M. Leng, "Emerging Technologies for Information Systems, Computing, and Management", pp.305~311 (2013), Springer
- C. C. Lin, "Heartbeat Classification Using Normalized RR Intervals and Wavelet Features", presented at the 2014 international symposium on computer, consumer and control (2014), IEEE
- K. Li, "Detecting ECG abnormalities via transductive transfer learning", presented at the ACM Conference on Bioinformatics, Computational Biology and Biomedicine (2012)
- Y.-H. Noh, "Implementation of real-time abnormal ECG detection algorithm for wearable healthcare", presented at the 6th International Conference on Computer Sciences and Convergence Information Technology (ICCIT) (2011), IEEE
- P. Grünwald, "The minimum description length principle" (2007), MIT press
- S. Yingchareonthawornchai, "Efficient proper length time series motif discovery", presented at the IEEE 13th International Conference on Data Mining (2013), IEEE
- B. Hu, "Using the minimum description length to discover the intrinsic cardinality and dimensionality of time series", Data Min. Knowl. Discov, 29(2), pp.358~399 (2015)[DOI]
- C. Chase, "Artifactual electrocardiographic change mimicking clinical abnormality on the ECG", Am. J. Emerg. Med, 18(3), pp.312~316 (2000)[DOI]
- E. Keigh, "HOT SAX: efficiently finding the most unusual time series subsequence", presented at the Fifth IEEE International Conference on Data Mining (2005), IEEE
- E. Keogh, "Finding the most unusual time series subsequence:algorithms and applications", Knowl. Inf. Syst, 11(1), pp.1~27 (2007)[DOI]
- G. Li, "Finding time series discord based on bit representation clustering", Knowl Based Syst, 54, pp.243~254 (2013)[DOI]
- K. Buza, "Fast classification of electrocardiograph signals via instance selection", presented at the IEEE First International Conference on Healthcare Informatics, Imaging and Systems Biology (2011)
- M. C. Chuah, "ECG anomaly detection via time series analysis", presented at the International Symposium on Parallel and Distributed Processing and Applications (2007)
- B. Raghavendra, "Cardiac arrhythmia detection using dynamic time warping of ECG beats in e-healthcare systems" (2011)[DOI]
- G. Zhang, "Electrocardiogram data mining based on frame classification by dynamic time warping matching", Comput. Methods Biomech. Biomed. Engin, 12(6), pp.701~707 (2009)[DOI]
- K. Polat, "Detection of ECG arrhythmia using a differential expert system approach based on principal component analysis and least square support vector machine", Appl. Math. Comput, 186(1), pp.898~906 (2007)[DOI]
- H. M. Rai, "ECG signal processing for abnormalities detection using multiresolution wavelet transform and Artificial Neural Network classifier", Measurement, 46(9), pp.3238~3246 (2013)[DOI]
- A. P. Lemos, "ECG anomalies identification using a time series novelty detection technique", presented at the IV Latin American Congress on Biomedical Engineering 2007 (2007)
- S. Chauhan, "Anomaly detection in ECG time signals via deep long short-term memory networks", presented at the IEEE International Conference on Data Science and Advanced Analytics (DSAA) (2015)
- Savelie Cornegruta, "Modelling Radiological Language with Bidirectional Long Short-Term Memory Networks", ~ (2016)
- A. Graves, "Framewise phoneme classification with bidirectional LSTM and other neural network architectures", Neural Netw, 18(5), pp.602~610 (2005)[DOI]
- F. Zhu, "Electrocardiogram generation with a bidirectional LSTMCNN generative adversarial network", Sci. Rep, 9(1), pp.1~11 (2019)[DOI]
- A. Mostayed, "Classification of 12-lead ECG signals with bi-directional LSTM network", ~ (2018)
- R. Azzam, "A stacked LSTM-based approach for reducing semantic pose estimation error", IEEE Trans. Instrum. Meas, (70), pp.1~14 (2020)[DOI]
- "MIT-BIH Arrhythmia Database" (2005)[URL]
- G. Moody, "The impact of the MIT-BIH Arrhythmia Database", IEEE Eng Med Biol, 20(3), pp.45~50[DOI]
- "Keras: Deep Learning for humans"[URL]
- "TensorFlow"[URL]