Data reconstruction for missing electrocardiogram using linear predictive coding

An electrocardiogram (ECG) reconstruction method based on a linear prediction technique is proposed in this paper. The method can reconstruct a rather long missing parts of ECG signals. Each missing data segment may cover 1 to 8 beats. The data used in the experiments are from the MIT-BIH normal sin...

全面介紹

Saved in:
書目詳細資料
Main Authors: Nipon Theera-Umpon, Panyaphon Phiphatkhunarnon, Sansanee Auephanwiriyakul
格式: Conference Proceeding
出版: 2018
主題:
在線閱讀:https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=64949199551&origin=inward
http://cmuir.cmu.ac.th/jspui/handle/6653943832/60354
標簽: 添加標簽
沒有標簽, 成為第一個標記此記錄!
實物特徵
總結:An electrocardiogram (ECG) reconstruction method based on a linear prediction technique is proposed in this paper. The method can reconstruct a rather long missing parts of ECG signals. Each missing data segment may cover 1 to 8 beats. The data used in the experiments are from the MIT-BIH normal sinus rhythm database. The experimental results show that our method can perform very well. The reconstructed signals are visually very close to the ground truths. The numerical evaluation also shows that the proposed method yields good results on the heart rate variability (HRV) measure derivation. It gives the time-domain HRV measures that are very close to the ground truths. Its performance is also better than the method commonly used by experts in which the abnormal beats are removed before calculating the HRV measures. © 2008 IEEE.