Citation:

FNU Sudhakar Abhijeet, 2026. "Automated Ecg Classification Using Transfer Learning From Vision Models To Medical Signal Domains", ESP Journal of Engineering & Technology Advancements  6(2): 96-105.

Abstract:

Electrocardiogram (ECG) classification is important for initial detection and diagnosis of cardiovascular diseases, but it remains challenging without human interpretation, as the signals vary and labeled datasets are limited. The latest developments in deep learning have shown excellent performance, but established models still tend to require large amounts of domain-specific training data. The review examines the new paradigm of leveraging transfer learning from vision-based models, such as convolutional neural networks and transformer models, for ECG classification tasks. By converting one-dimensional ECGs into two-dimensional images such as spectrograms and scalograms, vision models that have been trained can be successfully customized to leverage the powerful attributes of medical signals. The paper critically examines the strategies of representation, transfer learning methods, and transfer learning benchmark databases, and highlights performance gains and generalization analyses. Also, major issues such as domain mismatch, interpretability, and constraints are addressed. The review concludes with research directions for the future, including self-supervised learning, multimodal integration, and real-time clinical applications, as well as the possibility of cross-domain knowledge transfer in the development of intelligent healthcare systems.

References:

[1] LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. nature, 521(7553), 436-444.

[2] Heaton, J. (2018). Ian Goodfellow, Yoshua Bengio, and Aaron Courville: Deep learning: The met Press, 2016, 800 pp., sib: 0262035618. Genetic programming and evolvable machines, 19(1), 305-307.

[3] Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). Imagenet classification with deep convolutional neural networks. Advances in Neural Information Processing Systems, 25.

[4] Moody, G. B., & Mark, R. G. (2001). The impact of the MIT-BIH arrhythmia database. IEEE engineering in medicine and biology magazine, 20(3), 45-50.

[5] Rajpurkar, P., Hannun, A. Y., Haghpanahi, M., Bourn, C., & Ng, A. Y. (2017). Cardiologist-level arrhythmia detection with convolutional neural networks. arXiv preprint arXiv:1707.01836.

[6] Shen, D., Wu, G., & Suk, H. I. (2017). Deep learning in medical image analysis. Annual review of biomedical engineering, 19, 221-248.

[7] Pan, S. J., & Yang, Q. (2009). A survey on transfer learning. IEEE Transactions on knowledge and data engineering, 22(10), 1345-1359.

[8] Chen, Y. (2025). Deep Semi-Supervised Point Cloud-Based Models with Uncertainty Awareness for Efficient Abnormality Detection in 3D Medical Imaging (Doctoral dissertation, Queensland University of Technology).

[9] Tolstikhin, I. O., Houlsby, N., Kolesnikov, A., Beyer, L., Zhai, X., Unterthiner, T., ... & Dosovitskiy, A. (2021). Mlp-mixer: An all-mlp architecture for vision. Advances in Neural Information Processing Systems, 34, 24261-24272.

[10] Raghu, M., Zhang, C., Kleinberg, J., & Bengio, S. (2019). Transfusion: Understanding transfer learning for medical imaging. Advances in Neural Information Processing Systems, 32.

[11] Webster, J. G. (Ed.). (2009). Medical instrumentation: application and design. John Wiley & Sons.

[12] L. Sörnmo and P. Laguna, Bioelectrical Signal Processing in Cardiac and Neurological Applications, Elsevier, 2005.

[13] Rangayyan, R. M. (2015). Analysis of Concurrent, Coupled, and Correlated Processes.

[14] Cover, T., & Hart, P. (1967). Nearest neighbor pattern classification. IEEE Transactions on Information Theory, 13(1), 21-27.

[15] Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017). Dermatologist-level classification of skin cancer with deep neural networks. nature, 542(7639), 115-118.

[16] Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., ... & Fei-Fei, L. (2015). Imagenet large-scale visual recognition challenge. International journal of computer vision, 115(3), 211-252.

[17] He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 770-778).

[18] Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., & Wojna, Z. (2016). Rethinking the inception architecture for computer vision. In Proceedings of the IEEE conference on computer vision and pattern recognition (pp. 2818-2826).

[19] Deng, J., Dong, W., Socher, R., Li, L. J., Li, K., & Fei-Fei, L. (2009, June). Imagenet: A large-scale hierarchical image database. In the 2009 IEEE Conference on Computer Vision and Pattern Recognition (pp. 248-255). Ieee.

[20] Wang, Z., & Oates, T. (2015, January). Encoding time series as images for visual inspection and classification using tiled convolutional neural networks. In Workshops at the twenty-ninth AAAI conference on artificial intelligence (Vol. 1, No. 1, pp. 20-954).

Keywords:

ECG Classification, Transfer Learning, Vision Transformers, Convolutional Neural Networks, Biomedical Signal Processing, Domain Adaptation