Project

Back to overview

Predict and Monitor Epilepsy After a First Seizure: The Swiss-First Study

Applicant Wiest Roland
Number 180365
Funding scheme Sinergia
Research institution Universitätsinstitut für Diagnostische, Interventionelle & Pädiatrische Radiologie Inselspital
Institution of higher education University of Berne - BE
Main discipline Interdisciplinary
Start/End 01.02.2019 - 31.01.2023
Approved amount 2'666'192.00
Show all

All Disciplines (3)

Discipline
Interdisciplinary
Neurology, Psychiatry
Biomedical Engineering

Keywords (5)

non-epileptic seizure; epileptic seizure; Epilepsy; Advanced Neuroimaging; EEG

Lay Summary (German)

Lead
Das multizentrische Projekt untersucht, welchen zusätzlichen diagnostischen Nutzen computergestützte Auswertealgorithmen nach einem ersten, möglicherweise epileptischen Krampfanfall bieten können.
Lay summary
Hintergrund
Etwa 10 bis 15% der Bevölkerung erleidet einmal in ihrem Leben einen epileptischen Anfall. Ca. 1% der Bevölkerung leidet hingegen an Epilepsie, d.h. an der Prädisposition zu wiederholt und unprovoziert auftretenden Anfällen. Nach einem Erstanfall müssen Patient/inn/en mittels spezifischer Diagnostik abgeklärt werden. Neben der neurologischen Untersuchung stehen dafür in erster Linie ein Routine-Elektroenzephalogramm (EEG) und eine Magnetresonanztomographie (MRT) mit spezialisiertem Epilepsie-Protokoll zur Verfügung. Routineauswertungen von EEG und MRT sind jedoch häufig negativ oder unspezifisch.
Ziel
Es soll untersucht werden, ob und wie computergestützte Auswertealgorithmen für EEG und MRT dazu geeignet sind, die medizinische Diagnostik zu unterstützen. Dabei soll überprüft werden, ob die Erstanfälle damit korrekt als epileptisch bzw. nicht-epileptisch klassifiziert werden können. Eine weitere Fragestellung der Studie ist, mit welcher Genauigkeit vorhergesagt werden kann, ob die Patient/inn/en in einem Beobachtungszeitraum von zwei Jahren weitere Anfälle erleiden werden.
Bedeutung
Im Rahmen des Projektes sollen schweizweit zwischen 500 und 600 Patient/inn/en nach einem Erstanfall in die Studie eingeschlossen werden. Die erhoffte Weiterentwicklung der Diagnostik soll es ermöglichen, gezielter und früher eine anfallsunterdrückende Therapie einleiten zu können. Gesellschaftlich relevant ist neben den Kosten für das Gesundheitswesen v.a. auch eine frühzeitige und richtige Einschätzung der Fahreignung.
Direct link to Lay Summary Last update: 17.06.2020

Responsible applicant and co-applicants

Employees

Project partner

Associated projects

Number Title Start Funding scheme
124089 Imaging large scale neuronal networks in epilepsy 01.05.2009 SPUM
169607 Longitudinal Brain Tumor Segmentation with Uncertainty Estimation using Fully-connected Conditional Random Field and Perturb-and-Maximum-Posterior-Marginal Estimation 01.04.2017 Project funding (Div. I-III)
170060 Stroke treatment goes personalized: Gaining added diagnostic yield by computer-assisted treatment selection (the STRAY-CATS project) 01.10.2017 Project funding (Div. I-III)
140332 Imaging large scale neuronal networks in epilepsy 01.05.2012 SPUM

Abstract

Diagnosis and management of patients who experience a first event with transient neurological deficit or loss of consciousness remains a challenging task for the consulting neurologist in the emergency department (ED). Between 10-15% of the population have a seizure or seizure-like event in their lifetime, which may remain isolated or marks the beginning of epilepsy. If the appropriate diagnosis is missed, outcome may be fatal (e.g. recurrent seizure with trauma, cardiogenic syncope, stroke) or costly (e.g. continuous antiepileptic treatment for psychogenic seizures). Thus, it would help tremendously to have access to hands-on tools, based on easily accessible techniques like EEG and magnetic resonance imaging (MRI), to determine the correct diagnosis and organize appropriate patient management.There is a considerable lack of techniques that support appropriate treatment decisions in the majority of patients admitted to ED or for a first questionable. EEG and MRI findings are frequently negative, or - in case of MRI - if lesions are present, they are not necessarily critical (i.e. incidental). In this proposal, we aim to prospectively collect EEG and neuroimaging data from different centers in Switzerland to investigate the prevalence of EEG- and structural MRI-abnormalities in a population-wide cohort. Since epilepsy is considered a structural and functional network disorder, it appears straightforward to investigate if aberrant structural and functional network architecture allows identifying patients with unprovoked epileptic seizures and if such alterations predispose to the development of epilepsy during follow-up. We make use of recent advances in EEG and MR signal analysis to investigate pre-disposing abnormalities that reflect preexisting epileptogenesis and/or an increased risk for seizure recurrence. We use the EEG microstates values to determine the presence of pathological brain states as potential marker of epilepsy, and combine these measures with morphometric MR and functional connectivity analysis to construct covariance networks that are the starting point for machine learning algorithms. We will also investigate the diagnostic accuracy of widespread available advanced neuroimaging techniques (DWI, perfusion imaging, SWI) to segregate epileptic and non-epileptic patients during a first admission to the ED. We further aim to introduce a new MRI technology based on measurements of local perturbations of the MR signal induced by weak electromagnetic fields that accompany neuronal signaling (phase-cycled stimulus-induced rotary saturation; pc-SIRS).Preliminary studies suggest very good performance of the outlined methods in differentiating epileptogenic from non-epileptogenic activities in patients with chronic epilepsy. With the present grant, we aim to extent these methods to patients admitted at the ED and determine their yield in differentiating early-onset epilepsy versus clinically similar events of other origins. We hypothesize that in case an event is a symptom of an underlying epilepsy, EEG microstates, EEG and MRI functional connectivity, DWI, SWI and pc-SIRS are significantly different from those parameters identified in normal subjects, patients with non-epileptic events and patients with acute symptomatic seizures (e.g. withdrawal seizures). We hope to identify methods or sets of methods with the help of deep learning methods that allow rapid and correct diagnosis of unclear events, and in a further step, analyze the presence of structural and functional abnormalities in neuroimaging data that identify the risk of recurrent seizures.By this, we aim to continue our effort on the investigation of abnormal organization of the epileptic brain that was initiated by the SPUM consortium “imaging large scale networks in epilepsy” in 2009 by the teams from Geneva and Bern, extending now to early onset epilepsy. We now extend our consortium to the Bern Biomedical Engineering (BBME) and the group of the Medical Image and Signal Processing group from Ghent, Belgium. Each of the applicants has a field of renowned expertise in the field of epilepsy, complementary with respect to the goals of the project which aims at improving the care of patients with unclear loss of consciousness.
-