Classificação Automática de Ritmos Cardíacos com Vision Transformers e Integração de Derivações
Resumo
A interpretação precisa do eletrocardiograma (ECG) é vital para o diagnóstico de doenças cardiovasculares, mas sua análise manual é complexa e sujeita a erros. Propomos um modelo de IA baseado em Vision Transformers para análise automática do sinal cardíaco. Integrando dados de duas derivações, o sistema classifica cinco ritmos (um normal e quatro arritmias). O modelo atingiu 96,17% de acurácia por paciente, demonstrando seu potencial como suporte à decisão clínica na triagem de alterações cardíacas.Referências
Alghieth, M. (2025). Deepecg-net: A hybrid transformer-based deep learning model for real-time ecg anomaly detection. Scientific Reports, 15(1):20714.
Ansari, Y., Mourad, O., Qaraqe, K., and Serpedin, E. (2023). Deep learning for ecg arrhythmia detection and classification: an overview of progress for period 2017–2023. Frontiers in Physiology, 14:1246746.
Breen, C., Kelly, G., and Kernohan, W. (2022). Ecg interpretation skill acquisition: A review of learning, teaching and assessment. Journal of electrocardiology, 73:125–128.
Hannun, A. Y., Rajpurkar, P., Haghpanahi, M., Tison, G. H., Bourn, C., Turakhia, M. P., and Ng, A. Y. (2019). Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network. Nature medicine, 25(1):65–69.
Jaya Prakash, A., Nasreddine Belkacem, A., Elfadel, I. M., Jelinek, H. F., and Atef, M. (2025). Advances in machine and deep learning for ecg beat classification: a systematic review. Frontiers in Digital Health, 7:1649923.
Ji, C., Wang, L., Qin, J., Liu, L., Han, Y., and Wang, Z. (2024). Msgformer: A multi-scale grid transformer network for 12-lead ecg arrhythmia detection. Biomedical Signal Processing and Control, 87:105499.
Kitchenham, B. et al. (2004). Procedures for performing systematic reviews. Keele, UK, Keele University, 33(2004):1–26.
Mahim, S., Hossen, M. E., Al Hasan, S., Islam, M. K., Iqbal, Z., Alibakhshikenari, M., Collotta, M., and Miah, M. S. (2024). Transmixer-af: advanced real-time detection of atrial fibrillation utilizing single-lead electrocardiogram signals. IEEE Access, 12:143149–143162.
Mohan, A., Elbers, D., Zilbershot, O., Afghah, F., and Vorchheimer, D. (2024). Deciphering heartbeat signatures: a vision transformer approach to explainable atrial fibrillation detection from ecg signals. In 2024 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pages 1–6. IEEE.
Sociedade Brasileira de Cardiologia (2026). Cardiômetro: Mortes por Doenças Cardiovasculares no Brasil. [link]. Acessado em: 23 de fevereiro de 2026.
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., and Polosukhin, I. (2017). Attention is all you need. Advances in neural information processing systems, 30.
Wen, Q., Zhou, T., Zhang, C., Chen, W., Ma, Z., Yan, J., and Sun, L. (2022). Transformers in time series: A survey. arXiv preprint arXiv:2202.07125.
World Health Organization (2025). Cardiovascular diseases (CVDs). [link]. Acessado em: 23 de fevereiro de 2026.
Xiao, Q., Lee, K., Mokhtar, S. A., Ismail, I., Pauzi, A. L. b. M., Zhang, Q., and Lim, P. Y. (2023). Deep learning-based ecg arrhythmia classification: A systematic review. Applied Sciences, 13(8):4964.
Yildirim, O., Talo, M., Ciaccio, E. J., San Tan, R., and Acharya, U. R. (2020). Accurate deep neural network model to detect cardiac arrhythmia on more than 10,000 individual subject ecg records. Computer methods and programs in biomedicine, 197:105740.
Zhao, Y., Ren, J., Zhang, B., Wu, J., and Lyu, Y. (2023). An explainable attention-based tcn heartbeats classification model for arrhythmia detection. Biomedical Signal Processing and Control, 80:104337.
Zheng, J., Zhang, J., Danioko, S., Yao, H., Guo, H., and Rakovski, C. (2020). A 12-lead electrocardiogram database for arrhythmia research covering more than 10,000 patients. Scientific Data, 7(1):48.
Zírpolo, A. S., Mesquita, E. T., and Ramos, T. C. (2025). Um modelo explicável para classificação de arritmias cardíacas utilizando a rede lstm. In Simpósio Brasileiro de Computação Aplicada à Saúde (SBCAS), pages 55–60. SBC.
Ansari, Y., Mourad, O., Qaraqe, K., and Serpedin, E. (2023). Deep learning for ecg arrhythmia detection and classification: an overview of progress for period 2017–2023. Frontiers in Physiology, 14:1246746.
Breen, C., Kelly, G., and Kernohan, W. (2022). Ecg interpretation skill acquisition: A review of learning, teaching and assessment. Journal of electrocardiology, 73:125–128.
Hannun, A. Y., Rajpurkar, P., Haghpanahi, M., Tison, G. H., Bourn, C., Turakhia, M. P., and Ng, A. Y. (2019). Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network. Nature medicine, 25(1):65–69.
Jaya Prakash, A., Nasreddine Belkacem, A., Elfadel, I. M., Jelinek, H. F., and Atef, M. (2025). Advances in machine and deep learning for ecg beat classification: a systematic review. Frontiers in Digital Health, 7:1649923.
Ji, C., Wang, L., Qin, J., Liu, L., Han, Y., and Wang, Z. (2024). Msgformer: A multi-scale grid transformer network for 12-lead ecg arrhythmia detection. Biomedical Signal Processing and Control, 87:105499.
Kitchenham, B. et al. (2004). Procedures for performing systematic reviews. Keele, UK, Keele University, 33(2004):1–26.
Mahim, S., Hossen, M. E., Al Hasan, S., Islam, M. K., Iqbal, Z., Alibakhshikenari, M., Collotta, M., and Miah, M. S. (2024). Transmixer-af: advanced real-time detection of atrial fibrillation utilizing single-lead electrocardiogram signals. IEEE Access, 12:143149–143162.
Mohan, A., Elbers, D., Zilbershot, O., Afghah, F., and Vorchheimer, D. (2024). Deciphering heartbeat signatures: a vision transformer approach to explainable atrial fibrillation detection from ecg signals. In 2024 46th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), pages 1–6. IEEE.
Sociedade Brasileira de Cardiologia (2026). Cardiômetro: Mortes por Doenças Cardiovasculares no Brasil. [link]. Acessado em: 23 de fevereiro de 2026.
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., and Polosukhin, I. (2017). Attention is all you need. Advances in neural information processing systems, 30.
Wen, Q., Zhou, T., Zhang, C., Chen, W., Ma, Z., Yan, J., and Sun, L. (2022). Transformers in time series: A survey. arXiv preprint arXiv:2202.07125.
World Health Organization (2025). Cardiovascular diseases (CVDs). [link]. Acessado em: 23 de fevereiro de 2026.
Xiao, Q., Lee, K., Mokhtar, S. A., Ismail, I., Pauzi, A. L. b. M., Zhang, Q., and Lim, P. Y. (2023). Deep learning-based ecg arrhythmia classification: A systematic review. Applied Sciences, 13(8):4964.
Yildirim, O., Talo, M., Ciaccio, E. J., San Tan, R., and Acharya, U. R. (2020). Accurate deep neural network model to detect cardiac arrhythmia on more than 10,000 individual subject ecg records. Computer methods and programs in biomedicine, 197:105740.
Zhao, Y., Ren, J., Zhang, B., Wu, J., and Lyu, Y. (2023). An explainable attention-based tcn heartbeats classification model for arrhythmia detection. Biomedical Signal Processing and Control, 80:104337.
Zheng, J., Zhang, J., Danioko, S., Yao, H., Guo, H., and Rakovski, C. (2020). A 12-lead electrocardiogram database for arrhythmia research covering more than 10,000 patients. Scientific Data, 7(1):48.
Zírpolo, A. S., Mesquita, E. T., and Ramos, T. C. (2025). Um modelo explicável para classificação de arritmias cardíacas utilizando a rede lstm. In Simpósio Brasileiro de Computação Aplicada à Saúde (SBCAS), pages 55–60. SBC.
Publicado
01/06/2026
Como Citar
PEREIRA, Douglas Blanc; RAMOS, Taiane Coelho.
Classificação Automática de Ritmos Cardíacos com Vision Transformers e Integração de Derivações. In: SIMPÓSIO BRASILEIRO DE COMPUTAÇÃO APLICADA À SAÚDE (SBCAS), 26. , 2026, Ouro Preto/MG.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 1146-1157.
ISSN 2763-8952.
DOI: https://doi.org/10.5753/sbcas.2026.21657.
