Impact of Data Transformation: An ECG Heartbeat Classification Approach
- PMID: 34713072
- PMCID: PMC8521829
- DOI: 10.3389/fdgth.2020.610956
Impact of Data Transformation: An ECG Heartbeat Classification Approach
Abstract
Cardiovascular diseases continue to be a significant global health threat. The electrocardiogram (ECG) signal is a physiological signal that plays a major role in preventing severe and even fatal heart diseases. The purpose of this research is to explore a simple mathematical feature transformation that could be applied to ECG signal segments in order to improve the detection accuracy of heartbeats, which could facilitate automated heart disease diagnosis. Six different mathematical transformation methods were examined and analyzed using 10s-length ECG segments, which showed that a reciprocal transformation results in consistently better classification performance for normal vs. atrial fibrillation beats and normal vs. atrial premature beats, when compared to untransformed features. The second best data transformation in terms of heartbeat detection accuracy was the cubic transformation. Results showed that applying the logarithmic transformation, which is considered the go-to data transformation, was not optimal among the six data transformations. Using the optimal data transformation, the reciprocal, can lead to a 35.6% accuracy improvement. According to the overall comparison tested by different feature engineering methods, classifiers, and different dataset sizes, performance improvement also reached 4.7%. Therefore, adding a simple data transformation step, such as the reciprocal or cubic, to the extracted features can improve current automated heartbeat classification in a timely manner.
Keywords: data Wrangling; feature conversion; feature mapping; feature representation; feature restructuring; feature transformation.
Copyright © 2020 Liang, Hussain, Abbott, Menon, Ward and Elgendi.
Conflict of interest statement
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Figures
Similar articles
-
An Improved Convolutional Neural Network Based Approach for Automated Heartbeat Classification.J Med Syst. 2019 Dec 18;44(2):35. doi: 10.1007/s10916-019-1511-2. J Med Syst. 2019. PMID: 31853698
-
A pyramid-like model for heartbeat classification from ECG recordings.PLoS One. 2018 Nov 14;13(11):e0206593. doi: 10.1371/journal.pone.0206593. eCollection 2018. PLoS One. 2018. PMID: 30427899 Free PMC article.
-
Comparative study of morphological and time-frequency ECG descriptors for heartbeat classification.Med Eng Phys. 2006 Nov;28(9):876-87. doi: 10.1016/j.medengphy.2005.12.010. Epub 2006 Feb 14. Med Eng Phys. 2006. PMID: 16476566
-
A multiview feature fusion model for heartbeat classification.Physiol Meas. 2021 Jun 29;42(6). doi: 10.1088/1361-6579/ac010f. Physiol Meas. 2021. PMID: 33984841
-
A hierarchical method based on weighted extreme gradient boosting in ECG heartbeat classification.Comput Methods Programs Biomed. 2019 Apr;171:1-10. doi: 10.1016/j.cmpb.2019.02.005. Epub 2019 Feb 20. Comput Methods Programs Biomed. 2019. PMID: 30902245
Cited by
-
Impact of Box-Cox Transformation on Machine-Learning Algorithms.Front Artif Intell. 2022 Apr 7;5:877569. doi: 10.3389/frai.2022.877569. eCollection 2022. Front Artif Intell. 2022. PMID: 35527793 Free PMC article.
References
-
- Klabunde RE. Cardiovascular Physiology Concepts, 2nd edn. Lippincott Williams & Wilkins (2011).
-
- Tateno K, Glass L. A method for detection of atrial fibrillation using RR intervals. Comp Cardiol. (2000) 27:391–4. 10.1016/j.amjcard.2011.01.028 - DOI
-
- Elgendi M, Jonkman M, De Boer F. In: 2008 7th IEEE International Conference on Cognitive Informatics. (2008). p. 83–8.
-
- Krishna Prasad GJSS. Classification of ECG arrhythmias using multi-resolution analysis and neural networks. In TENCON 2003. Bangalore (2003). p. 227–31.
LinkOut - more resources
Full Text Sources