Task-specific oscillatory synchronization of prefrontal cortex, nucleus reuniens, and hippocampus during working memory.

Working memory requires maintenance of and executive control over task-relevant information on a timescale of seconds. Spatial working memory depends on interactions between hippocampus, for the representation of space, and prefrontal cortex, for executive control. A monosynaptic hippocampal projection to the prefrontal cortex has been proposed to serve this interaction. However, connectivity and inactivation experiments indicate a critical role of the nucleus reuniens in hippocampal-prefrontal communication. We have investigated the dynamics of oscillatory coherence throughout the prefrontal-hippocampal-reuniens network in a touchscreen-based working memory task. We found that coherence at distinct frequencies evolved depending on phase and difficulty of the task. During choice, the reuniens did not participate in enhanced prefrontal-hippocampal theta but in gamma coherence. Strikingly, the reuniens was strongly embedded in performance-related increases in beta coherence, suggesting the execution of top-down control. In addition, we show that during working memory maintenance the prefrontal-hippocampal-reuniens network displays performance-related delay activity.

Intrathecal application of ethosuximide is highly efficient in suppressing seizures in a genetic model of absence epilepsy.

Systemic drug application is the main approach in epilepsy treatment. However, the central nervous system (CNS) is a challenging target for drug delivery as the blood-brain barrier (BBB) restricts the transfer of drugs into the brain. Accordingly, there is a general interest in developing new therapeutic strategies to improve CNS drug accessibility. Intrathecal administration of antiseizure drugs (ASDs) e.g. via pumps or advanced materials could be a possible approach to bypass the BBB and increase the availability of neuroactive compounds in the CNS. The aim of this study was the evaluation of intracerebroventricular (i.c.v.) compared to systemic drug application in generalized epilepsy. The i.c.v. administration of the established ASD ethosuximide (ETX) in Genetic Absence Epilepsy Rats from Strasbourg (GAERS) caused a robust and dose-dependent reduction of spike-wave discharges (SWDs) without causing obvious behavioral abnormalities. Additionally, we could show that i.c.v. treatment with ETX is significantly more effective in seizure suppression than systemic treatment with the same dose. The localized application resulted in reduced systemic drug exposure compared to standard systemic ETX therapy. The tracing of dye distribution throughout the CNS supported the view that i.c.v. applied drugs cross into brain tissue surrounding the ventricles but largely remain restricted to the site of injection. Our data suggest that intrathecal application represents a possible route for the treatment in generalized epilepsy through direct drug penetration from CSF into brain tissue.

Epileptic Seizure Detection on an Ultra-Low-Power Embedded RISC-V Processor Using a Convolutional Neural Network.

The treatment of refractory epilepsy via closed-loop implantable devices that act on seizures either by drug release or electrostimulation is a highly attractive option. For such implantable medical devices, efficient and low energy consumption, small size, and efficient processing architectures are essential. To meet these requirements, epileptic seizure detection by analysis and classification of brain signals with a convolutional neural network (CNN) is an attractive approach. This work presents a CNN for epileptic seizure detection capable of running on an ultra-low-power microprocessor. The CNN is implemented and optimized in MATLAB. In addition, the CNN is also implemented on a GAP8 microprocessor with RISC-V architecture. The training, optimization, and evaluation of the proposed CNN are based on the CHB-MIT dataset. The CNN reaches a median sensitivity of 90% and a very high specificity over 99% corresponding to a median false positive rate of 6.8 s per hour. After implementation of the CNN on the microcontroller, a sensitivity of 85% is reached. The classification of 1 s of EEG data takes t=35 ms and consumes an average power of P≈140 µW. The proposed detector outperforms related approaches in terms of power consumption by a factor of 6. The universal applicability of the proposed CNN based detector is verified with recording of epileptic rats. This results enable the design of future medical devices for epilepsy treatment.