An Automated Machine Learning Method to Efficiently Classify the 12-lead ECG Signal Acquisition Quality

• Vandemberg M. O. Júnior UFC
• Vitor R. Evangelista UFC
• Ramon Baronetti Fundação Adib Jatene
• Vinicius R. Uemoto Fundação Adib Jatene
• Danielo G. Gomes UFC
• Mariana F. N. De Marchi Fundação Adib Jatene
• Renata V. Freitas Fundação Adib Jatene
• João P. V. Madeiro UFC

DOI: https://doi.org/10.5753/sbcas.2023.229594

Resumo

Given their low cost and non-invasive nature, ElectroCardioGram (ECG) signals have been widely used as a useful tool for diagnosing heart diseases. However, acquisition issues such as electrode interchange and oscillation noise may negatively impact expert exam interpretation and even automatic classification tasks. Here we propose an automated machine learning method to efficiently classify the 12-lead ECG signal acquisition quality. It consists of a two-stage classification process. Firstly, the ECG signals are processed and segmented aiming to classify them as noisy or acceptable signals. Then, the second classification stage yields the binary classification correct acquisition or limb electrodes interchange. Concerning the electrode positioning, the Random Forest technique presented interesting results (precision of 97%, recall of 89%, and F1-Score of 93%). Concerning noise detection, Random Forest presented a general accuracy of 85%, a recall of 57%, and a precision of 91%. All the obtained results yield to consider the proposed framework for application within a real telemedicine environment.

Referências

Alberto, A. C., Pedrosa, R. C., Zarzoso, V., and Nadal, J. (2020). Association between circadian holter ECG changes and sudden cardiac death in patients with chagas heart disease. Physiological Measurement, 41(2):025006.

Brüser, C., Antink, C. H., Wartzek, T., Walter, M., and Leonhardt, S. (2015). Ambient and unobtrusive cardiorespiratory monitoring techniques. IEEE Reviews in Biomedical Engineering, 8:30–43.

Caldas, W. L., do Vale Madeiro, J. P., Pedrosa, R. C., Gomes, J. P. P., Du, W., and Marques, J. A. L. (2023). Noise Detection and Classification in Chagasic ECG Signals Based on One-Dimensional Convolutional Neural Networks, pages 117–129. Springer International Publishing, Cham.

Ghaffari, A., Homaeinezhad, M., Akraminia, M., Atarod, M., and Daevaeiha, M. (2009). A robust wavelet-based multi-lead electrocardiogram delineation algorithm. Medical engineering & physics, 31(10):1219–1227.

Hayn, D., Jammerbund, B., and Schreier, G. (2012). Qrs detection based ecg quality assessment. Physiological Measurement, 33:1449–1461.

Jekova, I., Krasteva, V., Christov, I., and Abächerli, R. (2012). Threshold-based system for noise detection in multilead ecg recordings. Physiological Measurement, 33:1463–1477.

Kwon, J., Kim, K., and Jeon, K. e. a. (2020). Artificial intelligence algorithm for predicting cardiac arrest using electrocardiography. Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine, 28(98).

Li, H. and Boulanger, P. (2020). A survey of heart anomaly detection using ambulatory electrocardiogram (ecg). Sensors, 20(5):1461.

Li, J. P., Haq, A. U., Din, S. U., Khan, J., Khan, A., and Saboor, A. (2020). Heart disease identification method using machine learning classification in e-healthcare. IEEE Access, 8:107562–107582.

Li, Q., Rajagopalan, C., and Clifford, G. D. (2014). A machine learning approach to multi-level ecg signal quality classification. Computer Methods and Programs in Biomedicine, 117(3):435–447.

Liu, G., Han, X., Tian, L., Zhou, W., and Liu, H. (2021). Ecg quality assessment based on hand-crafted statistics and deep-learned s-transform spectrogram features. Computer Methods and Programs in Biomedicine, 208.

Liu, Y., Zhang, H., Zhao, K., Liu, H., Long, F., Chen, L., and Yang, Y. (2023). An automatic ecg signal quality assessment method based on resnet and self-attention. Applied Sciences, 13:1313.

Madeiro, J. P., Cortez, P. C., Marques, J. A., Seisdedos, C. R., and Sobrinho, C. R. (2012). An innovative approach of qrs segmentation based on first-derivative, hilbert and wavelet transforms. Medical engineering & physics, 34(9):1236–1246.

Moody, G. and Mark, R. (2001). The impact of the mit-bih arrhythmia database. IEEE Eng. Med. Biol, 20(3):45–50.

Rjoob, K., Bond, R., Finlay, D., McGilligan, V., Leslie, S. J., Rababah, A., Guldenring, D., Iftikhar, A., Knoery, C., McShane, A., et al. (2020). Machine learning techniques for detecting electrode misplacement and interchanges when recording ecgs: a systematic review and meta-analysis. Journal of Electrocardiology, 62:116–123.

Satija, U., Ramkumar, B., and Manikandan, M. S. (2018). A review of signal processing techniques for electrocardiogram signal quality assessment. IEEE Reviews in Biomedical Engineering, 11:36–52.

Satija, U., Ramkumar, B., and Sabarimalai Manikandan, M. (2017). Real-time signal quality-aware ecg telemetry system for iot-based health care monitoring. IEEE Internet of Things Journal, 4(3):815–823.

Silva, I., Moody, G. B., and Celi, L. (2011). Improving the quality of ecgs collected using mobile phones: The physionet/computing in cardiology challenge 2011. In 2011 Computing in Cardiology, pages 273–276. IEEE.

Vonesch, C., Blu, T., and Unser, M. (2007). Generalized daubechies wavelet families. IEEE Transactions on Signal Processing, 55(9):4415–4429.

Zhao, Z. and Zhang, Y. (2018). Sqi quality evaluation mechanism of single-lead ecg signal based on simple heuristic fusion and fuzzy comprehensive evaluation. Frontiers in Physiology, 9.