Detection of epilectic seizures from intracraneal eeg data

e-Archivo Repository

Show simple item record

dc.contributor.advisor Leiva Murillo, José Miguel García Alonso, Eva 2016-07-13T09:21:58Z 2016-07-13T09:21:58Z 2015-06 2015-07-07
dc.description.abstract This Bachelor Thesis presents a new and innovative proposal for epileptic patients that have not the expected results only with medication. The system consists on a device that provides an intracranial electroencephalogram that is a continuous, real-­‐time measurement of the brain activity and brain waves. This simple device placed on the surface of the brain is connected with an advisory system. However, the part in which this Bachelor Thesis is based on is the one related with electroencephalogram data processing. In this way, it opens the 8 possibility of the detection and forecast of an incoming seizure and leaves enough time to allow doctors to act to a possible threat for the patient. In addition, another problem that the antiepileptic drugs present is that they are not a specialized treatment for each of the patient, whose epilepsy could be different, and in consequence, the medication will not act exactly in the same manner in all the patients. This innovative system will provide an individual treatment for each patient, acting only when necessary for each of the individuals of the study. Thus, the idea is to study the data of intracranial electroencephalogram (iEEG) that is implanted in 4 dogs and 8 patients in order to get those patterns and get some periodicity between them to detect the attacks. This iEEG uses algorithms to detect the seizures and deliver responsive stimulation to avoid them. In addition, it allows us to investigate the origin of the seizures. [4] The first objective of this Bachelor Thesis is to create an algorithm that allow us to classify a set of data (test data) into ictal regions (epilepsy attacks) and interictal regions (non epilepsy attacks) on the basis on another set of data that has its own classification (training data). In order to do that, some concepts about logistic regression and cross-­‐validation will be essential. Once this problem is solved, the second aim is to separate the first fifteen seconds of the attack (early ictal segments) from the rest of the attack segments (late ictal segments). Detecting early ictal segments will provide the possibility of acting against an incoming seizure before it occurs or at the very beginning of the epilepsy attack. In addition, assessing future health care benefits as well as a study of those benefits will be achieved with this project. The software used for these objectives will be Matlab_R2014a software that is a mathematical tool that allows to apply all these statistical principles previously appointed like logistic regression, characterization of signals or data analysis. 9 In the following chapters, the way to detect and control these seizures will be explained as well as previous knowledge that is necessary to understand how and why the device is implanted and in which mechanisms it is based on.
dc.format.mimetype application/pdf
dc.language.iso eng
dc.rights Atribución-NoComercial-SinDerivadas 3.0 España
dc.subject.other Epilectic seizures
dc.subject.other Intracraneal eeg data
dc.subject.other Neurophysiology
dc.subject.other Epilepsy
dc.subject.other Intracranial electroencephalogram (iEEG)
dc.subject.other Ictal segments
dc.title Detection of epilectic seizures from intracraneal eeg data
dc.type bachelorThesis
dc.subject.eciencia Biología y Biomedicina
dc.rights.accessRights openAccess Ingeniería Biomédica
dc.contributor.departamento Universidad Carlos III de Madrid. Departamento de. Teoría de la Señal y Comunicaciones
 Find Full text

Files in this item

*Click on file's image for preview. (Embargoed files's preview is not supported)

The following license files are associated with this item:

This item appears in the following Collection(s)

Show simple item record