A computational framework to support seizure predictions in epileptic patients is presented. It is based on mining and knowledge discovery in Elec-troencephalogram (EEG) signal. A set of features is extracted and classification techniques are then used to eventually derive an alarm signal predicting a coming seizure. The epileptic patient may then take steps in order to prevent accidents and social exposure.
Keywords
data mining; seizure prediction; computational intyelligence; multidimensional scaling
Subject
Data mining; seizure prediction
Related Project
EPILEPSIAE- Evolving Platform for Improving Living Expectation of Patients
Conference
ICDM- International Conference on Data Mining, July 2008
Cited by
Year 2012 : 2 citations
Nearest neighbor estimate of conditional mutual information in feature selection
A Tsimpiris, I Vlachos, D Kugiumtzis - Expert Systems with Applications, 2012 - Elsevier
Alkiviadis Tsimpiris, Ioannis Vlachos, Dimitris Kugiumtzis, "Nearest neighbor estimate of conditional mutual information in feature selection", Expert Systems with Applications, Vol. 39 Issue 16, November 2012.
Year 2010 : 2 citations
A. Tsimpiris, D. Kugiumtzis, "EEG Features as biomarkers for discrimination of pre-ictal states", International Conference on Biomedical Data & Knowledge Mining: Towards Biomarker Discovery, Chania, Greece, July 2010.
A. Tsimpiris, D. Kugiumtzis and P.G. Larsson, "Discrimination of epileptic pre-ictal states using feature based clustering on EEG", XII Mediterranean Conference on Medical and Biological Engineering and Computing (Medicon 2010), Porto Carras, Chalkidiki, Greece, May 2010.
