On mapping epilepsy : magneto- and electroencephalographic characterizations of epileptic activities

Epilepsy is one of the most common neurological disorder, affecting up to 10 individuals per 1000 persons. The disorder have been known for several thousand years, with the first clinical descriptions dating back to ancient times. Nonetheless, characterization of the dynamics underlying epilepsy remains largely unknown. Understanding these patophysiological processes requires unifying both a neurobiological perspective, as well as a technically advanced neuroimaging perspective. The incomplete insight into epilepsy dynamics is reflected by the insufficient treatment options. Approximately 30% of all patients do not respond to anti-epileptic drugs (AEDs) and thus suffers from recurrent seizures despite adequate pharmacological treatments. These pharmacoresistant patients often undergo epilepsy surgery evaluations.

Epilepsy surgery aims to resect the part of the brain that generates the epileptic seizure activity (seizure onset zone, SOZ). Nonetheless, up to 50% of all patients relapse after surgery. This can be due to incomplete mapping of both the SOZ and of other structures that might be involved in seizure initiation and propagation. Such cortical and subcortical structures are collectively referred to as the epileptic network.

Historically, epilepsy was considered to be either a generalized disorder involving the entire brain, or a highly localized, focal, disorder. The modern technological development of both structural and functional neuroimaging has drastically altered this view. This development has made significant contributions to the now prevailing view that both generalized and focal epilepsies arise from more or less widespread pathological network pathways. Visualization of these pathways play an important role in the presurgical planning. Thus, both improved characterization and understanding of such pathways are pivotal in improvement of epilepsy diagnostics and treatments. It is evident that epilepsy research needs to stand on two legs: Both improved understanding of pathological, neurobiological and neurophysiological process, and improved neuroimaging instrumentation.

Epilepsy research do not only span from visualization to understanding of neurophysiological processes, but also from cellular, neuronal, microscopic processes, to dynamical, large-scale network processes. It is well known that neurons involved in epileptic activities exhibit specific, pathological firing patterns. Genetic mutations resulting in neuronal ion channel defects can cause severe, and even lethal, epileptic syndromes in children, clearly illustrating a role for neuron membrane properties in epilepsy. However, cellular processes themselves cannot explain how epileptic seizures can involve, and propagate across, large cortical areas and generate seizure-specific symptomatologies. A strict cellular perspective can neither explain epilepsy-associated pathological interactions between larger distant regions in between seizures. Instead, the dynamical effects of cellular synchronization across both mesoscopic and macroscopic scales also need to be considered. Today, the only means to study such effects in human subjects are by combinations of neuroimaging modalities. However, as all measurement techniques, these exhibit individual limitations that affect the kind of information that can be inferred from these. Thus, once more we reach the conclusion that epilepsy research needs to rest upon both a neurophysiological/neurobiological leg, and a technical/instrumentational leg.

In accordance with this necessity of a dual approach to epilepsy, this thesis covers both neurophysiological aspects of epileptic activity development, as well as functional neuroimaging instrumentation development with focus on epileptic activity detection and localization. Part 1 (neurophysiological part) is concerned with the neurophysiological dynamical changes that underlie development of so called interictal epileptiform discharges (IEDs) with special focus on the role of low-frequency oscillations. To this aim, both conventional magnetoencephalography (MEG) and intracranial electroencephalography (iEEG) with neurostimulation is analyzed. Part 2 (instrumentation part) is concerned with development of cutting-edge, novel on-scalp magnetoencephalography (osMEG) within clinical epilepsy evaluations and research with special focus on IEDs. The theses cover both modeling of osMEG characteristics, as well as the first- ever osMEG recording of a temporal lobe epilepsy patient.

List of scientific papers

I. Westin K, Lundstrom B, Van Gompel J, Cooray G. Neurophysiological effects of continuous cortical stimulation in epilepsy – Spike and spontaneous ECoG activity. Clin Neurophysiol. 2019;130:38–45.
https://doi.org/10.1016/j.clinph.2018.10.009

II. Westin K, Cooray G, Lundqvist D. Interictal epileptiform discharges in focal epilepsy are preceded by a gradual increase in low-frequency oscillations. Clin Neurophysiol. 2022.
https://doi.org/10.1016/j.clinph.2022.02.003

III. Westin K, Pfeiffer C, Andersen LM, Ruffieux S, Cooray G, Kalaboukhov A, et al. Clinical Neurophysiology Detection of interictal epileptiform discharges : A comparison of on-scalp MEG and conventional MEG measurements. Clin Neurophysiol. 2020;131:1711–20.
https://doi.org/10.1016/j.clinph.2020.03.041

IV. Westin K, Beniczky S, Hämäläinen M, Lundqvist D. On the clinical benefit of on-scalp MEG: A modeling study of on-scalp MEG epileptic activity source estimation ability. [Manuscript]