14 A novel representation of ECG beat-to-beat variation. (26th April 2019)

• Record Type: Journal Article
• Authors: Liu, Nan; Guo, Dagang; Koh, Zhi Xiong; Wah Ho, Andrew Fu; Hock Ong, Marcus Eng
• Is Part Of: BMJ open. Volume 9:Supplement 2(2019)
• Page Start: A6
• Page End: A6
• Publication Date: 2019-04-26
• Subjects: Medicine -- Research -- Periodicals; 610.72
• DOI: 10.1136/bmjopen-2019-EMS.14
• Languages: English
• ISSNs: 2044-6055

Abstract

Abstract : Background: Heart rate variability (HRV) is believed to strongly associate with autonomic nervous system. So far, majority of efforts in HRV research are deriving sophisticated parameters with linear and nonlinear techniques. Furthermore, researchers have been focusing on developing advanced signal processing tools for efficient noise removal and accurate QRS detection, prior to HRV parameter calculation. Method: We propose a novel representation of beat-to-beat variation in ECG, called heart rate n-variability (HRnV), as an alternative to conventional HRV. The derivation of HRnV parameters are based on n RR intervals with or without overlaps. We can create many sets of HRnV parameters which are promising at generating extra information from limited data source. We conducted a simulation study by using the ECG record of subject #16 265 from MIT-BIH Normal Sinus Rhythm Database. Results: Among the time domain parameters, we observed that the values were generally incremental with the increase of n. We observed the same trend of value change in frequency domain parameters. In nonlinear analysis, the differences between HRV and HRnV on Poincare plot measures were obvious, while those on entropy and detrended fluctuation analysis (DFA) metrics were not. Conclusion: HRnV measures enable us to augment the conventional HRV with many more parameters. We believe that HRnV is an important addition to HRV and will have great potential in analyzing prehospital ECGs. Conflict of interest: None. Funding: SingHealth Foundation Research Grant SHF/FG652P/2017.