Functional Neuroimaging in Temporal Lobe Epilepsy

Functional neuroimaging has transformed our understanding of temporal lobe epilepsy (TLE) by revealing aberrant brain activity and network disruptions that extend beyond the epileptogenic focus. Techniques such as resting-state functional magnetic resonance imaging (rs-fMRI), positron emission tomography and electroencephalography-fMRI fusion permit non-invasive mapping of functional connectivity, metabolic activity and haemodynamic responses. These modalities elucidate both static and dynamic interactions among hippocampus, neocortex, thalamus and large-scale networks including the default mode and salience systems. By characterising alterations in network topology, temporal dynamics and regional coupling, functional imaging informs diagnosis, guides surgical planning and predicts postoperative outcome. Recent advances in machine learning and deep learning have further refined the ability to detect subtle neuroimaging signatures, stratify patients into biologically meaningful subtypes and offer objective biomarkers for prognostication and personalised treatment.

Research from Nature Portfolio

Machine-learning analysis of structural MRI data has identified four distinct TLE phenotypes characterised by differential patterns of hippocampal atrophy, neocortical involvement and amygdala enlargement. These data-driven biotypes show divergent clinical trajectories and varying seizure outcomes following surgery, suggesting a framework for precision medicine in epilepsy care. In parallel, convolutional neural network algorithms applied to routine T1-weighted MRI have achieved high accuracy in distinguishing TLE from Alzheimer’s disease and healthy controls. Feature-visualisation methods highlight subtle abnormalities in hippocampal subfields and adjacent cortex that escape conventional radiological assessment, underscoring the potential of deep learning to augment clinical decision-making and identify non-lesional TLE cases.

Functional Neuroimaging in Temporal Lobe Epilepsy publication trend

The graph below shows the total number of articles in functional neuroimaging in temporal lobe epilepsy across all publications each year (not limited to Nature Index journals).

Technical terms

Resting-state functional MRI: A technique measuring spontaneous brain activity fluctuations when a subject is not performing a task.

Functional connectivity: Statistical dependencies between activity time-series in different brain regions, indicating network interactions.

Dynamic functional connectivity: Temporal variations in functional connectivity patterns over the course of a scan.

Hippocampal sclerosis: Neuronal loss and gliosis in the hippocampus, a common pathological substrate in TLE.

References

1. Identification of four biotypes in temporal lobe epilepsy via machine learning on brain images. Nature Communications (2024).
2. MRI-based deep learning can discriminate between temporal lobe epilepsy, Alzheimer’s disease, and healthy controls. Communications Medicine (2023).
3. Dynamic functional connectivity and gene expression correlates in temporal lobe epilepsy: insights from hidden markov models. Journal of Translational Medicine (2024).
4. Altered Functional Connectivity and Small-World in Mesial Temporal Lobe Epilepsy. PLOS ONE (2010).