In der Pathogenese der Temporallappenepilepsie wird kindlichen hippokampalen Schädigungen eine wesentliche Rolle zugeschrieben. Epileptische Krämpfe und perinatale Asphyxie sind zwei häufige Ursachen dieser Schädigungen. Anhaltende epileptiforme Aktivität im Niedrig-Mg2+-Modell als einer experimentellen Form epileptischer Krämpfe führt in organotypischen hippokampalen Schnittkulturen (OHSK) der Ratte, die als Ersatzsystem des kindlichen Hippokampus verwendet werden, zu Zellschäden. Während dieser Untersuchungen ergab sich der Verdacht auf eine zusätzlich schädigende Wirkung erhöhter Sauerstoffspannungen. In meiner ersten Versuchsreihe konnte ich nachweisen, dass erhöhte Sauerstoffspannungen (60 %, 95 %) verglichen mit 20%-Sauerstoffspannung zu reversiblen und irreversiblen Zellschäden in OHSK führen. Die Zellschäden wurden über Veränderungen reizinduzierter Feldpotentiale, d.h. Abnahme der Amplitude, Zunahme der Latenz und Zunahme des Doppelpulsindex, sowie über die Propidium Jodid (PJ)-Fluoreszenzintensität bestimmt. In der zweiten Versuchsreihe konnte gezeigt werden, dass erhöhte Sauerstoffspannungen auch nach einer Hypoxie im Sinne einer hyperoxischen Reoxygenierung verglichen mit normoxischer Reoxygenierung vermehrt Zellschäden in OHSK zur Folge haben. In der dritten Versuchsreihe konnte ich ausschließen, dass erhöhte Sauerstoffspannungen eine notwendige Bedingung für Zellschäden infolge anhaltender epileptiformer Aktivität sind. Um die zellschädigende Rolle von Spreading Depressions (SDs), die während epileptiformer Aktivität auftreten, zu bestimmen, wurde in der vierten Versuchsreihe eine Methode etabliert, SD-ähnliche Ereignisse isoliert und zuverlässig in normoxischen OHSK auszulösen. Auf diese Weise wiederholt ausgelöste SD-ähnliche Ereignisse führten zu Zellschäden, bestimmt über die Veränderung elektrophysiologischer Eigenschaften von SD-ähnlichen Ereignissen, Abnahme der Feldpotentialamplitude und PJ-Fluoreszenzintensität. / Hippocampal damage during infancy is thought to play an important role in the pathogenesis of temporal lobe epilepsy. Epileptic seizures and perinatal asphyxia are two frequent causes of these damages. Sustained epileptiform activity induced in the low Mg2+-model of epileptic seizures leads to cell damage in organotypic hippocampal slice cultures (OHSC) of the rat, which are used as a surrogate for the infantile hippocampus. During a previous study utilising this model the suspicion arose that increased oxygen tension could have an additional damaging effect. My first series of experiments proved that increased oxygen tension (60 %, 95 %) lead to reversible and irreversible cell damage in OHSC compared to 20%-oxygen tension. Cell damage was determined by alterations of evoked field potentials, i.e. decrement of amplitude, increment of latency and paired pulse index, as well as by propidium iodide fluorescence. The second series of experiments showed that increased oxygen tension applied after an hypoxic period (hyperoxic reoxygenation) result in augmented cell damage compared to normoxic reoxygenation. With the third series of experiments it could be excluded that increased oxygen tension is an essential condition for the occurrence of cell damage due to sustained epileptiform activity. In order to elucidate the damaging role of spreading depressions (SD), which emerge during epileptiform activity, a method was established in the fourth series of experiments that allowed the reliable induction of SD-like events in normoxic OHSC. Repetitive SD-like events induced by this method led to cell damage, assessed by alterations of electrophysiological characteristics of SD-like events, decrement of evoked field potential amplitude and propidium iodide fluorescence.

Epilepsy affects more than 50 million people and is one of the most prevalent neurological disorders and has a high impact on the quality of life of those suffering from it. However, 70% of epilepsy patients can live seizure free with proper diagnosis and treatment. Patients are evaluated using scalp EEG recordings which is cheap and non-invasive. Diagnostic yield is however low and qualified personnel need to process large amounts of data in order to accurately assess patients. MindReader is an unsupervised classifier which detects spectral anomalies and generates a hypothesis of the underlying patient state over time. The aim is to highlight abnormal, potentially epileptiform states, which could expedite analysis of patients and let qualified personnel attest the results. It was used to evaluate 95 scalp EEG recordings from healthy adults and adult patients with epilepsy. Interictal Epileptiform discharges (IED) occurring in the samples had been retroactively annotated, along with the patient state and maneuvers performed by personnel, to enable characterization of the classifier’s detection performance. The performance was slightly worse than previous benchmarks on pediatric scalp EEG recordings, with a 7% and 33% drop in specificity and sensitivity, respectively. Electrode positioning and partial spatial extent of events saw notable impact on performance. However, no correlation between annotated disturbances and reduction in performance could be found. Additional explorative analysis was performed on serialized intermediate data to evaluate the analysis design. Hyperparameters and electrode montage options were exposed to optimize for the average Mathew’s correlation coefficient (MCC) per electrode per patient, on a subset of the patients with epilepsy. An increased window length and lowered amount of training along with an common average montage proved most successful. The Euclidean distance of cumulative spectra (ECS), a metric suitable for spectral analysis, and homologous L2 and L1 loss function were implemented, of which the ECS further improved the average performance for all samples. Four additional analyses, featuring new time-frequency transforms and multichannel convolutional autoencoders were evaluated and an analysis using the continuous wavelet transform (CWT) and a convolutional autoencoder (CNN) performed the best, with an average MCC score of 0.19 and 56.9% sensitivity with approximately 13.9 false positives per minute.