Feature Extraction from Electrocardiogram Data Based on Mathematical Modeling
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
This study proposes a novel approach for electrocardiogram (ECG) classification by leveraging the sparse identification of nonlinear dynamical systems (SINDy) algorithm for feature extraction. ECG signals represented as time series data are transformed into coefficients derived from second-order differential equations, enabling a compact and interpretable representation of the dynamics of each heartbeat. Using SINDy-extracted features, three machine learning models - logistic regression, random forest, and XGBoost - were evaluated on the MIT-BIH and PTB-XL datasets. For the MIT-BIH dataset, which involves multiclass classification, experiments were conducted on both the original imbalanced and resampled balanced data. Results indicate that SINDy-based features enabled high classification accuracy, with random forest achieving 97.39% accuracy on the balanced MIT-BIH data. In the binary classification experiments with the PTB-XL dataset, XGBoost achieved the highest accuracy of 92.98%. These findings suggest that differential-equation-based feature extraction offers a promising alternative to traditional ECG classification methods, combining high accuracy with interpretability. Future research will explore refining differential equations to capture complex ECG dynamics more precisely, as well as extending this approach to multicycle ECG patterns to further enhance classification accuracy.
Abstract
This study proposes a novel approach for electrocardiogram (ECG) classification by leveraging the sparse identification of nonlinear dynamical systems (SINDy) algorithm for feature extraction. ECG signals represented as time series data are transformed into coefficients derived from second-order differential equations, enabling a compact and interpretable representation of the dynamics of each heartbeat. Using SINDy-extracted features, three machine learning models - logistic regression, random forest, and XGBoost - were evaluated on the MIT-BIH and PTB-XL datasets. For the MIT-BIH dataset, which involves multiclass classification, experiments were conducted on both the original imbalanced and resampled balanced data. Results indicate that SINDy-based features enabled high classification accuracy, with random forest achieving 97.39% accuracy on the balanced MIT-BIH data. In the binary classification experiments with the PTB-XL dataset, XGBoost achieved the highest accuracy of 92.98%. These findings suggest that differential-equation-based feature extraction offers a promising alternative to traditional ECG classification methods, combining high accuracy with interpretability. Future research will explore refining differential equations to capture complex ECG dynamics more precisely, as well as extending this approach to multicycle ECG patterns to further enhance classification accuracy.
참고문헌 (21)
- "Cardiovascular diseases (CVDs) fact sheet" (2021)[URL]
- Moody GB, "The impact of the MIT-BIH arrhythmia database", IEEE Eng in Med and Biol, 20, pp.45~50 (2001)
- Bousseljot R, "Nutzung der EKG-Sig-naldatenbank CARDIODAT der PTB über das internet", Biomed Te, 1995, pp.40(s1), 317~318
- Ansari Y, "Deep learning for ECG arrhythmia detection and classification : an overview of pro-gress for the period 2017-2023", Front Physiol, 14, pp.1246746~ (2023)
- Ebrahimi Z, "A review on deep learning methods for ECG arrhythmia classification", Comput Biol Med, 113, pp.103802~ (2020)
- Brunton SL, "Discovering governing equations from data by sparse identification of nonlinear dynamical systems", Proc Natl Acad Sci U S A, 113, pp.3932~3937 (2016)
- Narotamo H, "Deep learning for ECG classification : a comparative study of 1D and 2D representations and multimodal fusion approaches", Biomed Signal Process Control, 93, pp.106141~ (2024)
- Śmigiel S, "ECG Signal classification using deep learning techniques based on the PTB-XL dataset", Entropy, 28(23), pp.1121~ (2021)
- Pyakillya B, "Deep learning for ECG classification", J Phys Conf Ser, 913, pp.012004~ (2017)
- Ribeiro AH, "Automatic diagnosis of the 12-lead ECG using a deep neural network", Nat Commun, 11, pp.1760~ (2020)
- Cheng J, "ECG signal classification based on deep CNN and BiLSTM", BMC Med Inform Decis Mak, 21, pp.365~ (2021)
- Aziz S, "ECG-based machine-learning algorithms for heartbeat classification", Sci Rep, 11, pp.18738~ (2021)
- Hassaballah M, "ECG heartbeat classification using machine learning and metaheuristic optimization for smart healthcare systems", Bioengineering, 10, pp.429~ (2023)
- Rudy SH, "Data-driven discovery of partial differential equations", Sci Adv, 3, pp.e1602614~ (2017)
- Chen J, "Machine learning techniques for stock price prediction and graphic signal recognition", Eng Appl Artif Intell, 121, pp.106038~ (2023)
- Chung D, "Forecast of COVID-19 confirmed cases across region in Korea using long short-term memory", QBS, 41, pp.37~45 (2022)
- Chung D, "Tracking the outbreak of Covid-19 variants in Korea using physics informed neural networks", QBS, 42, pp.17~26 (2023)
- Chung D, "From time series to Topology : advancing League of Legends win prediction with Vietoris-Rips Persistents Homoloy", QBS, 43, pp.87~93 (2024)
- Hosmer DW, "Applied Logistic Regression" (2013), John Wiley & Sons
- Breiman L, "Random forests", Mach Learn, 45, pp.5~32 (2001)
- Chen T, "XGBoost: A scalable tree boosting system", Proceedings of the 22nd ACM SIGKDD International Con-ference on Knowledge Discovery and Data Mining, pp.785~794 (2016)