Constantin Krempp

Constantin Krempp

Faculty Mentor: Sydney S. Cash, M.D., Ph.D.
Department of Neurology, Massachusetts General Hospital
Project Title: Single pulse electrical stimulation for epileptic seizures prediction and sleep neuromodulation
Constantin Krempp

Project Summary: Electrical stimulation is a common tool to interact with neural networks, for both research and clinical purposes. A typical example is deep brain stimulation (DBS) that is used to treat various diseases including Parkinson’s disease or epilepsy. This kind of stimulation involves high currents during extended durations of stimulation, disturbing the neural networks with possible side effects. In this project we investigate the usage of ‘gentler’ methods using single pulses (SPES) for both interrogating and modulating the neural networks.

In a first project we use SPES to predict seizures in a rodent model of epilepsy. Seizure prediction is usually a very tough problem involving complex deep learning methods with suboptimal levels of accuracy. In our methodology, instead of using data based on ‘passive’ EEG recordings, we actively probe the epileptogenic locus of the animals with SPES and base our predictions on the evoked response to stimulation. Fundamentally, the underlying question is to determine if actively interrogating the neural circuitries can provide better insights about the brain states than usual EEG recordings.

In a second project, we use SPES to modulate the neural activity during sleep. We modulate sleep spindles, which are typical oscillations happening during sleep and thought to be related to memory consolidation. Using a closed-loop system developed in the lab, we detect spindles in human patients implanted with intracranial EEG and stimulate as they are occurring. We then quantify our influence.

Fellowships

Fellowship Year