ECG-ResNeXt: Age Prediction in Pediatric Electrocardiograms and Its Correlations with Comorbidities

• Pedro Robles Dutenhefner UFMG
• Gabriel Lemos UFMG
• Turi Rezende UFMG
• Jose Geraldo Fernandes UFMG
• Diogo Tuler UFMG
• Gisele Lobo Pappa UFMG
• Gabriela Miana Paixão UFMG
• Antônio Luiz Pinho Ribeiro UFMG
• Wagner Meira Jr. UFMG

DOI: https://doi.org/10.5753/eniac.2024.245277

Abstract

Cardiovascular diseases are the leading cause of global mortality, with the electrocardiogram (ECG) being essential for cardiac health assessment. Beyond automatic diagnosis, the use of artificial intelligence models for age prediction from ECG data has shown great potential, enabling the identification of adverse clinical conditions by discrepancies in predicted age. In this pioneering study, we explore age prediction in pediatric data by proposing the ECG-ResNeXt model, which incorporates advancements such as inverted bottlenecks and Global Response Normalization, outperforming results obtained by previous models. Additionally, we analyze correlations between prediction errors and comorbidities, highlighting the clinical potential of this approach.

Keywords: Age Prediction, Neural Networks, Pediatric Electrocardiogram

References

Chan, T. C., Sharieff, G. Q., and Brady, W. J. (2008). Electrocardiographic manifestations: pediatric ecg. The Journal of emergency medicine, 35(4):421–430.

Chang, C.-H., Lin, C.-S., Luo, Y.-S., Lee, Y.-T., and Lin, C. (2022). Electrocardiogram-based heart age estimation by a deep learning model provides more information on the incidence of cardiovascular disorders. Frontiers in Cardiovascular Medicine, 9:754909.

Che, C., Zhang, P., Zhu, M., Qu, Y., and Jin, B. (2021). Constrained transformer network for ecg signal processing and arrhythmia classification. BMC Medical Informatics and Decision Making, 21(1):184.

Chen, J., Huang, S., Zhang, Y., Chang, Q., Zhang, Y., Li, D., Qiu, J., Hu, L., Peng, X., Du, Y., et al. (2024). Congenital heart disease detection by pediatric electrocardiogram based deep learning integrated with human concepts. Nature Communications, 15(1):976.

Ebrahimi, Z., Loni, M., Daneshtalab, M., and Gharehbaghi, A. (2020). A review on deep learning methods for ecg arrhythmia classification. Expert Systems with Applications: X, 7:100033.

García-Vicente, C., Gutiérrez-Tobal, G. C., Jiménez-García, J., Martín-Montero, A., Gozal, D., and Hornero, R. (2023). Ecg-based convolutional neural network in pediatric obstructive sleep apnea diagnosis. Computers in Biology and Medicine, 167:107628.

Hirota, N., Suzuki, S., Motogi, J., Nakai, H., Matsuzawa, W., Takayanagi, T., Umemoto, T., Hyodo, A., Satoh, K., Arita, T., et al. (2023). Cardiovascular events and artificial intelligence-predicted age using 12-lead electrocardiograms. IJC Heart & Vasculature, 44:101172.

Hu, R., Chen, J., and Zhou, L. (2022). A transformer-based deep neural network for arrhythmia detection using continuous ecg signals. Computers in Biology and Medicine, 144:105325.

Lima, E. M., Ribeiro, A. H., Paixão, G. M., Ribeiro, M. H., Pinto-Filho, M. M., Gomes, P. R., Oliveira, D. M., Sabino, E. C., Duncan, B. B., Giatti, L., et al. (2021). Deep neural network-estimated electrocardiographic age as a mortality predictor. Nature communications, 12(1):5117.

Liu, X., Wang, H., Li, Z., and Qin, L. (2021). Deep learning in ecg diagnosis: A review. Knowledge-Based Systems, 227:107187.

Liu, Z., Mao, H., Wu, C.-Y., Feichtenhofer, C., Darrell, T., and Xie, S. (2022). A convnet for the 2020s. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 11976–11986.

Loshchilov, I. and Hutter, F. (2016). Sgdr: Stochastic gradient descent with warm restarts. arXiv preprint arXiv:1608.03983.

Loshchilov, I. and Hutter, F. (2017). Decoupled weight decay regularization. arXiv pre-print arXiv:1711.05101.

Mayourian, J., Geggel, R., La Cava, W. G., Ghelani, S. J., and Triedman, J. K. (2024). Pediatric electrocardiogram-based deep learning to predict secundum atrial septal defects. Pediatric Cardiology, pages 1–6.

Mori, H., Inai, K., Sugiyama, H., and Muragaki, Y. (2021). Diagnosing atrial septal defect from electrocardiogram with deep learning. Pediatric cardiology, 42(6):1379–1387.

Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., and Chen, L.-C. (2018). Mobilenetv2: Inverted residuals and linear bottlenecks. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 4510–4520.

Woo, S., Debnath, S., Hu, R., Chen, X., Liu, Z., Kweon, I. S., and Xie, S. (2023). Conv-next v2: Co-designing and scaling convnets with masked autoencoders. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 16133–16142.

World Health Organization (2024). Cardiovascular diseases. Accessed: 2024-08-16.

Zeng, Z., Kaur, R., Siddagangappa, S., Rahimi, S., Balch, T., and Veloso, M. (2023). Financial time series forecasting using cnn and transformer. arXiv preprint arXiv:2304.04912.