EEG-based epileptic seizures prediction using machine learning / Imane Meguenni

Public ISBD Titre : EEG-based epileptic seizures prediction using machine learning Type de document : document multimédia Auteurs : Imane Meguenni, Auteur ; Hiba Reggab, Auteur ; Mohamed El Habib Maicha, Directeur de thèse Editeur : Laghouat : Université Amar Telidji - Département d'informatique Année de publication : 2024 Importance : 93 p. Accompagnement : 1 disque optique numérique (CD-ROM) Note générale : Option : Computer Science Langues : Anglais Mots-clés : Epilepsy  Seizure Detection  Machine learning  Electroencephalogram  Binary Classification  Feature Selection  Random Forest algorithm Résumé : Epilepsy is a neurological disorder characterized by recurrent seizures. Current methods often rely on clinical observation for seizure detection, potentially missing subtle pre-seizure indicators. This thesis explores the potential of machine learning to predict the probability of a seizure occurring within a specific timeframe, offering a more nuanced understanding of pre-seizure brain activity. We leverage a custom dataset combining publicly available electroencephalogram (EEG) recordings with data collected from collaborating healthcare institutions. This approach ensures the relevance and validity of the predictive models developed. Our models move beyond the typical binary classification ("seizure" or "nonseizure") by predicting the percentage chance of a seizure. This probabilistic approach empowers individuals with epilepsy to make informed decisions based on the predicted risk level. It represents a significant departure from prevalent methods, potentially leading to a more informative and impactful seizure prediction system. The thesis tackles several challenges associated with machine learning for seizure prediction, including data heterogeneity, feature selection, model interpretability, and clinical validation. The specific machine learning algorithm we will explore is the Random Forest algorithm. By addressing these challenges and using a custom dataset, we aim to contribute to the development of effective tools for predicting epilepsy onset, ultimately improving patient outcomes and healthcare delivery. note de thèses : Mémoire de master en informatique EEG-based epileptic seizures prediction using machine learning [document multimédia] / Imane Meguenni, Auteur ; Hiba Reggab, Auteur ; Mohamed El Habib Maicha, Directeur de thèse . - Laghouat : Université Amar Telidji - Département d'informatique, 2024 . - 93 p. + 1 disque optique numérique (CD-ROM). Option : Computer Science Langues : Anglais Mots-clés : Epilepsy  Seizure Detection  Machine learning  Electroencephalogram  Binary Classification  Feature Selection  Random Forest algorithm Résumé : Epilepsy is a neurological disorder characterized by recurrent seizures. Current methods often rely on clinical observation for seizure detection, potentially missing subtle pre-seizure indicators. This thesis explores the potential of machine learning to predict the probability of a seizure occurring within a specific timeframe, offering a more nuanced understanding of pre-seizure brain activity. We leverage a custom dataset combining publicly available electroencephalogram (EEG) recordings with data collected from collaborating healthcare institutions. This approach ensures the relevance and validity of the predictive models developed. Our models move beyond the typical binary classification ("seizure" or "nonseizure") by predicting the percentage chance of a seizure. This probabilistic approach empowers individuals with epilepsy to make informed decisions based on the predicted risk level. It represents a significant departure from prevalent methods, potentially leading to a more informative and impactful seizure prediction system. The thesis tackles several challenges associated with machine learning for seizure prediction, including data heterogeneity, feature selection, model interpretability, and clinical validation. The specific machine learning algorithm we will explore is the Random Forest algorithm. By addressing these challenges and using a custom dataset, we aim to contribute to the development of effective tools for predicting epilepsy onset, ultimately improving patient outcomes and healthcare delivery. note de thèses : Mémoire de master en informatique

Réservation

Réserver ce document

Exemplaires

Code-barres	Cote	Support	Localisation	Section	Disponibilité
MF 02-70	MF 02-70	CD	BIBLIOTHEQUE DE FACULTE DES SCIENCES	théses (sci)	Disponible

---

page 1/1