Neurophysiology

Brain signals such as EEG, intracranial EEG and MEG are information-rich but high-dimensional, and their clinical interpretation is often qualitative. We develop quantitative methods to extract clinically meaningful information from these recordings.

We have used quantitative iEEG to characterise how seizures vary within individual patients — finding that seizure network pathways change on circadian and slower timescales in all patients studied. Building on this, we have investigated how anti-seizure medications modulate seizure activity, showing that they act through two distinct mechanisms: suppressing specific seizure activity patterns in an all-or-nothing fashion, and curtailing the duration of others.

We have also developed normative maps of interictal iEEG activity to identify pathological tissue. We recently produced a large multi-centre normative map spanning 502 subjects across 15 centres, accounting for age, sex and recording site effects. Using such normative approaches alongside MRI, we have shown that quantitative multimodal integration of MRI and iEEG abnormalities can help localise epileptogenic tissue and predict surgical outcome in temporal lobe epilepsy. More broadly, we have used quantitative iEEG analysis to test long-held clinical assumptions — for example, showing that incomplete resection of the iEEG-defined seizure onset zone is not necessarily associated with worse surgical outcome.