Electrocardiography (ECG) analysis and a new feature extraction method using wavelet transform with scalogram analysis.
denoising
electrocardiography
feature extraction
pulmonary arterial hypertension
scalogram
Journal
Biomedizinische Technik. Biomedical engineering
ISSN: 1862-278X
Titre abrégé: Biomed Tech (Berl)
Pays: Germany
ID NLM: 1262533
Informations de publication
Date de publication:
25 Oct 2020
25 Oct 2020
Historique:
received:
14
06
2019
accepted:
14
02
2020
pubmed:
23
5
2020
medline:
22
4
2021
entrez:
23
5
2020
Statut:
ppublish
Résumé
Electrocardiography (ECG) signals and the information obtained through the analysis of these signals constitute the main source of diagnosis for many cardiovascular system diseases. Therefore, accurate analyses of ECG signals are very important for correct diagnosis. In this study, an ECG analysis toolbox together with a user-friendly graphical user interface, which contains the all ECG analysis steps between the recording unit and the statistical investigation, is developed. Furthermore, a new feature calculation methodology is proposed for ECG analysis, which carries distinct information than amplitudes and durations of ECG main waves and can be used in artificial intelligence studies. Developed toolbox is tested using both Massachusetts Institute of Technology-Beth Israel Hospital (MIT-BIH) Arrhythmia ECG Database and an experimentally collected dataset for performance evaluation. The results show that ECG analysis toolbox presented in this study increases the accuracy and reliability of the ECG main wave detection analysis, highly fasten the process duration compared to manual ones and the new feature set can be used as a new parameter for decision support systems about ECG based on artificial intelligence.
Identifiants
pubmed: 32441663
doi: 10.1515/bmt-2019-0147
pii: /j/bmte.ahead-of-print/bmt-2019-0147/bmt-2019-0147.xml
doi:
pii:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM