Detecção de Anomalias em Frequências Cardíacas Utilizando Autocodificadores para Identificação Precoce de COVID-19

• Thayná Rosa Silvestre UFAM
• Eulanda Miranda dos Santos UFAM
• Rafael Albuquerque Pinto UFAM
• Lucas Gabriel Coimbra Evangelista UFAM

DOI: https://doi.org/10.5753/sbcas.2022.222546

Abstract

The SARS-CoV-2 virus that causes the coronavirus disease has already been transmitted around the world for more than two years, in addition to its many mutations. Among different ways to test whether a person is infected, anomaly detection based on signals such as resting heart rate, which are collected using wearable devices, are investigated as an alternative to identify COVID-19 in its pre-symptomatic phase. There are some studies in the literature that have already demonstrated different techniques to detect these anomalies even with limited data representing this problem. In this work we investigate whether Autocodificadors designed with convolution layers are effective on detecting COVID-19 before the onset of symptoms. Our experiments are conducted using a public database composed by resting heart rate signals from 25 people infected with the virus.

References

Alavi, A., Bogu, G. K., Wang, M., Rangan, E. S., Brooks, A. W., Wang, Q., Higgs, E., Celli, A., Mishra, T., Metwally, A. A., Cha, K., Knowles, P., Alavi, A. A., Bhasin, R., Panchamukhi, S., Celis, D., Aditya, T., Honkala, A., Rolnik, B., Hunting, E., Dagan-Rosenfeld, O., Chauhan, A., Li, J. W., Bejikian, C., Krishnan, V., McGuire, L., Li, X., Bahmani, A., and Snyder, M. P. (2022). Real-time alerting system for covid-19 and other stress events using wearable data. Nature Medicine.

Bogu, G. K. and Snyder, M. P. (2021). Deep learning-based detection of covid-19 using wearables data. medRxiv.

Dunn, J., Kidzinski, L., Runge, R., Witt, D., Hicks, J. L., Rose, S.-F., Miryam, S., Li, X., Bahmani, A., Delp, S. L., Hastie, T., and Snyder, M. P. (2021). Wearable sensors enable personalized predictions of clinical laboratory measurements. Nature Medicine.

Gopali, S., Abri, F., Siami-Namini, S., and Namin, A. S. (2021). A comparative study of detecting anomalies in time series data using LSTM and TCN models. CoRR, abs/2112.09293.

Mishra, T., Wang, M., Metwally, A. A., Bogu, G. K., Brooks, A. W., Bahmani, A., Alavi, A., Celli, A., Higgs, E., Dagan-Rosenfeld, O., Fay, B., Kirkpatrick, S., Kellogg, R., Gibson, M., Wang, T., Hunting, E. M., Mamic, P., Ganz, A. B., Rolnik, B., Li, X., and Snyder, M. P. (2020). Pre-symptomatic detection of covid-19 from smartwatch data. Nature Biomedical Engineering.