Automatic Identification of Interictal Epileptiform Discharges with the Use of Complex Networks

Autor(es)

Tomanik, Gustavo H. ; Betting, Luiz E. ; Campanharo, Andriana S. L. O. ; Rojas, I ; Joya, G ; Catala, A

• Tipo de documento: Artigo Científico
• Fonte: ADVANCES IN COMPUTATIONAL INTELLIGENCE, IWANN 2019, PT I; v. 11506, p. 10-pg., 2019-01-01.

Resumo

The identification of Interictal Epileptiform Discharges (IEDs), which are characterized by spikes and waves in electroencephalographic (EEG) data, is highly beneficial to the automated detection and prediction of epileptic seizures. In this paper, a novel single-step approach for IEDs detection based on the complex network theory is proposed. Our main goal is to illustrate how the differences in dynamics in EEG signals from patients diagnosed with idiopathic generalized epilepsy are reflected in the topology of the corresponding networks. Based on various network metrics, namely, the strongly connected component, the shortest path length and the mean jump length, our results show that this method enables the discrimination between IEDs and free IEDs events. A decision about the presence of epileptiform activity in EEG signals was made based on the confusion matrix. An overall detection accuracy of 98.2% was achieved. (AU)

• Processo FAPESP: 17/09216-7 - Previsão de convulsões em dados de EEG com o uso de redes complexas
• Beneficiário: Gustavo Henrique Tomanik
• Modalidade de apoio: Bolsas no Exterior - Estágio de Pesquisa - Iniciação Científica
• Processo FAPESP: 16/17914-3 - Efeito dos fármacos antiepilépticos e das descargas epileptiformes na conectividade funcional de pacientes com epilepsia mioclônica juvenil
• Beneficiário: Luiz Eduardo Gomes Garcia Betting
• Modalidade de apoio: Auxílio à Pesquisa - Regular
• Processo FAPESP: 18/25358-9 - Uso de redes complexas na detecção automática, no diagnóstico e na classificação da Doença de Alzheimer
• Beneficiário: Andriana Susana Lopes de Oliveira Campanharo
• Modalidade de apoio: Auxílio à Pesquisa - Regular
• Processo FAPESP: 18/02014-2 - Uso de redes complexas na previsão de convulsões e na localização da zona epileptogênica de pacientes epilépticos
• Beneficiário: Gustavo Henrique Tomanik
• Modalidade de apoio: Bolsas no Brasil - Mestrado