Identification of pre-ictal states based on an EEG-ECG multi-feature clustering approach

A seizure prediction system can improve lives of patients with epilepsy. Although seizure prediction has been a widely-researched topic, there is still no reliable and robust system that has strong theoretical support to be able to differentiate between inter-ictal and pre-ictal states. This study a...

Full description

Autores:
Flórez Torres, Nicolás
Tipo de recurso:
Fecha de publicación:
2019
Institución:
Universidad de los Andes
Repositorio:
Séneca: repositorio Uniandes
Idioma:
eng
OAI Identifier:
oai:repositorio.uniandes.edu.co:1992/44053
Acceso en línea:
http://hdl.handle.net/1992/44053
Palabra clave:
Procesamiento de señales - Técnicas digitales - Investigaciones
Epilepsia
Electroencefalografía - Investigaciones
Electrocardiografía - Investigaciones
Aprendizaje automático (Inteligencia artificial)
Ingeniería
Rights
openAccess
License
http://creativecommons.org/licenses/by-nc-nd/4.0/
Description
Summary:A seizure prediction system can improve lives of patients with epilepsy. Although seizure prediction has been a widely-researched topic, there is still no reliable and robust system that has strong theoretical support to be able to differentiate between inter-ictal and pre-ictal states. This study aims at identify if there is a natural grouping of data that allows differentiating these pre- seizure states using different strategies of electroencephalography (EEG) and electrocardiographic (ECG) signal characteristics of epileptic patients and allowing the anticipation of seizures using algorithms of unsupervised machine learning, seeking to be a reliable tool for predicting seizures.