|
|
Detection of High-Frequency EEG Activity in Epileptic Patients
Cimbálník, Jan ; Kremláček, Jan (oponent) ; Jiruška,, Přemysl (oponent) ; Jurák, Pavel (vedoucí práce)
This work deals with automated detection of high-frequency oscillations as a novel electrophysiologic biomarker of epileptogenic tissue in intracranial EEG. Visual detection of these oscillations is a time-consuming process and is prone to reviewer bias. Epilepsy is one of the most common neurological diseases affecting 1 % of population. Even though two thirds of cases are successfully treated with anti-epileptic drugs, the rest of the patients are dependent mainly on surgical procedure, which requires precise localization of pathologic focus. High-frequency oscillations have been studied over the last decade for their potential to localize the focus of pathological tissue. Initial part of this work is a summary of the current state of high-frequency oscillations research and a detailed list of detectors used in research. Within the scope of this work three high-frequency oscillation detectors were developed or enhanced. The description of the algorithms is followed by detector evaluation with regard to the concordance with expert reviewed events, feature estimation and the ability to correctly localize pathological tissue. The final part of the work provides an overview of developed visualization methods and a short summary of achieved scientific results.
|
|
|
Unsupervised Deep Learning Approach for Seizure Onset Zone localization in Epilepsy
Přidalová, Tereza ; Cimbálník, Jan (oponent) ; Mehnen, Lars (vedoucí práce)
Epilepsy affects about 50 million people worldwide, with one-third of patients being drugresistant and therefore candidates for an invasive brain resection surgery. Brain resection surgery candidates undergo invasive intracranial encephalography (iEEG) monitoring to determine the seizure onset zone (SOZ). Recorded data can span over weeks and need to be manually reviewed by a physician to assess SOZ. This process can be time-consuming and burdensome due to the vast amount of collected data. This work investigates utilisation of an deep autoencoder for unsupervised data exploration and specifically its ability to discriminate between SOZ and non-SOZ (NSOZ) iEEG channels. The data used in this thesis consists of iEEG collected from 33 patients in two institutes (Mayo Clinic, Rochester, Minnesota, USA and St. Anne´s University Hospital, Brno, Czech Republic - FNUSA) who underwent invasive presurgical monitoring. The autoencoder’s capability to discriminate between SOZ and NSOZ was evaluated using a self-learned embedded feature space representation of the autoencoder network. Autoencoder features were compared to previously established biomarkers for SOZ determination. Discrimination capability was evaluated for both autoencoder features and biomarkers using a Naive Bayes classifier and leave-one-out cross-validation. The achieved area under receiver operating characteristic curve (AUROC) was 0.68 for the FNUSA and 0.56 for the Mayo dataset. Performance in discriminating between SOZ and NSOZ electrodes was not significantly different between the investigated autoencoder features and previously established biomarkers. Selecting the better performing classifier for each patient increased the AUROC to 0.75 and 0.64 for the FNUSA and Mayo dataset, respectively. The results suggest that future approaches combining biomarkers and self-learning methods have a potential to improve the SOZ vs NSOZ discrimination capability of unsupervised iEEG exploration systems, and thus to enhance the surgical management of epilepsy.
|
|
|
Implementace nové metody do modelu strojového učení na lokalizaci epileptického ložiska u pacientů s farmakorezistentní epilepsií
Pivnička, Martin ; Mívalt, Filip (oponent) ; Filipenská, Marina (vedoucí práce)
Bakalářská práce rozebírá problematiku lokalizace epileptického ložiska u pacientů s farmakorezistentní epilepsií. Teoretická část ve své první části pojednává o podstatě epilepsie a její léčbě. Popisuje princip elektroencefalografického měření a jeho přínos v epileptologii. Taktéž nastiňuje různé varianty lokalizace epileptické zóny v mozku. Druhá polovina teoretického úvodu je zaměřena na principy strojového učení a jejich využití pro léčbu epilepsie. V praktické části je popsána tvorba a funkce gamma metody, stejně jako její statistické ohodnocení. Výsledky zahrnují jak samostatnou funkčnost metody, tak i výkon v rámci existujícího modelu strojového učení. Bylo prokázáno, že gamma metoda představuje cenný specifický parametr pro lokalizaci epileptického ložiska. Její přidání do modelu strojového učení nevedlo k zásadnímu zlepšení práce modelu.
|
|
|
Unsupervised Deep Learning Approach for Seizure Onset Zone localization in Epilepsy
Přidalová, Tereza ; Cimbálník, Jan (oponent) ; Mehnen, Lars (vedoucí práce)
Epilepsy affects about 50 million people worldwide, with one-third of patients being drugresistant and therefore candidates for an invasive brain resection surgery. Brain resection surgery candidates undergo invasive intracranial encephalography (iEEG) monitoring to determine the seizure onset zone (SOZ). Recorded data can span over weeks and need to be manually reviewed by a physician to assess SOZ. This process can be time-consuming and burdensome due to the vast amount of collected data. This work investigates utilisation of an deep autoencoder for unsupervised data exploration and specifically its ability to discriminate between SOZ and non-SOZ (NSOZ) iEEG channels. The data used in this thesis consists of iEEG collected from 33 patients in two institutes (Mayo Clinic, Rochester, Minnesota, USA and St. Anne´s University Hospital, Brno, Czech Republic - FNUSA) who underwent invasive presurgical monitoring. The autoencoder’s capability to discriminate between SOZ and NSOZ was evaluated using a self-learned embedded feature space representation of the autoencoder network. Autoencoder features were compared to previously established biomarkers for SOZ determination. Discrimination capability was evaluated for both autoencoder features and biomarkers using a Naive Bayes classifier and leave-one-out cross-validation. The achieved area under receiver operating characteristic curve (AUROC) was 0.68 for the FNUSA and 0.56 for the Mayo dataset. Performance in discriminating between SOZ and NSOZ electrodes was not significantly different between the investigated autoencoder features and previously established biomarkers. Selecting the better performing classifier for each patient increased the AUROC to 0.75 and 0.64 for the FNUSA and Mayo dataset, respectively. The results suggest that future approaches combining biomarkers and self-learning methods have a potential to improve the SOZ vs NSOZ discrimination capability of unsupervised iEEG exploration systems, and thus to enhance the surgical management of epilepsy.
|
|
|
Detection of High-Frequency EEG Activity in Epileptic Patients
Cimbálník, Jan ; Kremláček, Jan (oponent) ; Jiruška,, Přemysl (oponent) ; Jurák, Pavel (vedoucí práce)
This work deals with automated detection of high-frequency oscillations as a novel electrophysiologic biomarker of epileptogenic tissue in intracranial EEG. Visual detection of these oscillations is a time-consuming process and is prone to reviewer bias. Epilepsy is one of the most common neurological diseases affecting 1 % of population. Even though two thirds of cases are successfully treated with anti-epileptic drugs, the rest of the patients are dependent mainly on surgical procedure, which requires precise localization of pathologic focus. High-frequency oscillations have been studied over the last decade for their potential to localize the focus of pathological tissue. Initial part of this work is a summary of the current state of high-frequency oscillations research and a detailed list of detectors used in research. Within the scope of this work three high-frequency oscillation detectors were developed or enhanced. The description of the algorithms is followed by detector evaluation with regard to the concordance with expert reviewed events, feature estimation and the ability to correctly localize pathological tissue. The final part of the work provides an overview of developed visualization methods and a short summary of achieved scientific results.
|
host ::
RSS.