TY - GEN
T1 - Brain and heart physiological networks analysis employing neural networks granger causality
AU - Jaimes-Albarracin, Anggie D.
AU - Orjuela-Canon, Alvaro D.
AU - Jutinico, Andres L.
AU - Bazurto, Maria A.
AU - Duenas, Elida
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021/5/4
Y1 - 2021/5/4
N2 - This paper presents the study of brain-heart interactions in pediatric patients diagnosed with obstructive sleep apnea (OSA). A comparison between pre- and post- treatment stages was treated, searching to analyze the therapy effect. For this purpose, polysomnography signals were characterized, making use of electroencephalography and electrocardiography to compute the heart rate variability. Physiological networks analysis was driven through the computation of Granger causality to find interactions among brain and heart in both directions. Also, a proposal based on artificial neural networks was employed to include a nonlinear Granger causality sense. A preprocessing was implemented, according to five spectral subbands for the brain case, associate to five rhythms, and three for the heart case. Results showed that the treatment allowed recovery of connections mainly in subsystems involving the delta and gamma subbands. Finally, a notorious difference was evidenced between the results obtained with both methods where the nonlinear analysis obtained complementary results.
AB - This paper presents the study of brain-heart interactions in pediatric patients diagnosed with obstructive sleep apnea (OSA). A comparison between pre- and post- treatment stages was treated, searching to analyze the therapy effect. For this purpose, polysomnography signals were characterized, making use of electroencephalography and electrocardiography to compute the heart rate variability. Physiological networks analysis was driven through the computation of Granger causality to find interactions among brain and heart in both directions. Also, a proposal based on artificial neural networks was employed to include a nonlinear Granger causality sense. A preprocessing was implemented, according to five spectral subbands for the brain case, associate to five rhythms, and three for the heart case. Results showed that the treatment allowed recovery of connections mainly in subsystems involving the delta and gamma subbands. Finally, a notorious difference was evidenced between the results obtained with both methods where the nonlinear analysis obtained complementary results.
UR - http://www.scopus.com/inward/record.url?scp=85107476701&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85107476701&partnerID=8YFLogxK
U2 - 10.1109/NER49283.2021.9441375
DO - 10.1109/NER49283.2021.9441375
M3 - Conference contribution
AN - SCOPUS:85107476701
T3 - International IEEE/EMBS Conference on Neural Engineering, NER
SP - 469
EP - 472
BT - 2021 10th International IEEE/EMBS Conference on Neural Engineering, NER 2021
PB - IEEE Computer Society
T2 - 10th International IEEE/EMBS Conference on Neural Engineering, NER 2021
Y2 - 4 May 2021 through 6 May 2021
ER -