Neuronal networks in Burst Suppression EEG as revealed by source analysis

Dynamic complexity measures and entropy paths for modelling and comparison of evolution of patients with drug resistant epileptic encephalopathy syndromes (DREES)

Epileptic encephalopathies (EE) is a term coined by the International League Against Epilepsy (ILAE) to refer to a group of epilepsies in which the ictal and interictal abnormalities may contribute to progressive cerebral dysfunction. Among them, two affect mainly children and are very difficult to deal with, Doose and Lennox-Gastaut syndromes, (DS and LGS, respectively). So far (Zavala-Yoe et al., J Integr Neurosci 15(2):205-223, 2015a and works of ours there), quantitative analysis of single case studies of EE have been performed. All of them are manifestations of drug resistant epileptic encephalopathies (DREES) and as known, such disorders require a lot of EEG studies through all patient's life. As a consequence, dozens of EEG records are stored by parents and neurologists as time goes by. However, taking into account all this massive information, our research questions (keeping colloquial wording by parents) arise: a) Which zone of the brain has been the most affected so far? b) On which year was the child better? c) How bad is our child with respect to others? We must reflect that despite clinical assessment of the EEG has undergone standardization by establishment of guidelines such as the recently published guidelines of the American Clinical Neurophysiology Society (Tsuchida et al., J Clin Neurophysiol 4(33):301-302, 2016), qualitative EEG will never be as objective as quantitative EEG, since it depends largely on the education and experience of the conducting neurophysiologist (Grant et al., Epilepsy Behav 2014(32):102-107, 2014, Rating, Z Epileptologie, Springer Med 27(2):139-142, 2014). We already answered quantitatively the above mentioned questions in the references of ours given above where we provided entropy curves and an entropy index which encompasses the complexity of bunches of EEG making possible to deal with massive data and to make objective comparisons among some patients simultaneously. However, we have refined that index here and we also offer another two measures which are spatial and dynamic. Moreover, from those indices we also provide what we call a temporal dynamic complexity path which shows in a standard 10-20 system head diagram the evolution of the lowest complexity per brain zone with respect to the EEG period. These results make it possible to compare quantitatively/graphically the progress of several patients at the same time, answering the questions posed above. The results obtained showed that we can associate low spatio-temporal entropy indices to multiple seizures events in several patients at the same time as well as tracking seizure progress in space and time with our entropy path, coinciding with neurophysiologists observations.