.png){kind=icon}
.png){kind=icon}
Return
Localization and classification of heartbeats using robust adaptive algorithm
.png){kind=icon}
PRE
AI
DOI：10.1016/j.bspc.2018.11.003.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Automated analysis of heart sound recordings (phonocardiograms or PCG) for classification of heart beats in noisy environments is a challenging problem. The major challenges are introduced due to inter- and intra-beat variation, spectral overlap of heart and lung sounds and noise picked from environment. This paper presents a robust algorithm for heart beat localization and classification using variable length window and peak amplitude threshold variation. The technique presented in this paper first attempts to find peaks which correspond to the fundamental heart sound cycle (FHSC) i.e. S1 ,S2, S1, ... or fundamental heart sound period (FHSP) i.e. S1, S1, .... If FHSC/FHSP is found, an attempt is made to extend it to the whole signal. Once extended, the signal is directly labeled using systole and diastole durations. During classification, history of labels is used to counter the heart rate variations. The proposed technique is evaluated on publicly available Pascal Heart Sound Challenge dataset and reports smaller average error for localization and has high sensitivity and accuracy for heart beat classification. (C) 2018 Elsevier Ltd. All rights reserved.
Keywords:
PCG
Heart beats
Systole and diastole
Classification
Localization
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
4.9
Papers:
9.8K
Citations:
2.4W