Project Details
Projekt
Neuromorphic Epileptic Seizure Prediction System
Subject Area
Electronic Semiconductors, Components and Circuits, Integrated Systems, Sensor Technology, Theoretical Electrical Engineering
Term
since 2024
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 537857806
Epilepsy is a family of chronic brain disorders characterized by recurrent seizures, typically involving excessive abnormal electrical activities in the brain, e.g. causing stiffening of muscles, involuntary shaking, complete or partial loss of consciousness. These seizures have a strong impact on the quality of life of patients and can cause accidental injuries and even death. Having a long-term risk forecast and a short-term warning alarm of impending seizures can allow patients time to seek a safer posture or take medications to suppress the seizures. Furthermore, It can help in determining the severity of the disease and the efficacy of various therapies guiding the choice of therapeutic options. An automated and personalized out-of-hospital implant for monitoring epilepsy activity can improve the patients' quality of life as well as reduce pressure on the healthcare system. Currently, there are a large number of devices on the market that only aim to detect seizures, and some offer limited intervention. Most of these devices demonstrate poor power consumption figures owing to the use of off-the-shelf electronics and cloud-based complex AI models which require big data for reliable prediction and are not very adaptive to alternating seizure dynamics. Thus such approaches are not very suitable for always-on personalized medicine research. Through this project, we aspire to bring a change to the life of epileptic patients by utilizing state-of-the-art neuromorphic embedded hardware. The proposed NESPreS system aims to provide long-term forecasting of seizure risk as well as short-term prediction of seizure onsets. Its design philosophy embraces long-term operation capabilities enabled by employing spiking neural networks implemented using neuromorphic technology. The robust, always-on and real-time prediction capabilities of this intelligent wearable device will provide a quantum leap in the domain of personalized health and biomedical instrumentation. The core novelty of the proposal lies in the joint co-design of both hardware and software components of the system that will be deployed and validated with clinical studies. NESPreS offers great potential to reduce the physical, psychological, and social burdens of epilepsy by developing the technology that will provide patients with constant reliable monitoring and support. Additionally, the general framework of the project using an integrated mixed-signal brain-on-a-chip hardware can provide a foundation for the creation of a new generation of always-on edge-computing sensors for a wide domain of biomedical signal processing platforms.
DFG Programme
Research Grants
International Connection
Switzerland
Partner Organisation
Schweizerischer Nationalfonds (SNF)
Cooperation Partners
Dr. Elisa Donati; Professor Giacomo Indiveri; Professor Dr. Johannes Sarnthein
