Scalp EEG records electrical activity through the skull. Stereo-electroencephalography, or SEEG, goes further by placing thin depth electrodes into selected brain regions to record activity directly from within the brain.
Where the method came from
The modern SEEG concept was developed in France by Jean Talairach, Jean Bancaud and colleagues. Their work used stereotactic methods to test a three-dimensional hypothesis about where seizures began and how they spread, rather than simply placing electrodes wherever was convenient.
Why SEEG is used
SEEG is generally considered when non-invasive tests do not give enough certainty for epilepsy surgery. MRI, scalp EEG, seizure semiology, PET, SPECT, MEG and neuropsychology may all contribute to the hypothesis that guides electrode placement.
What it can reveal
Because electrodes can sample deep and distributed structures, SEEG can help distinguish the true seizure-onset network from activity that has already spread. Electrical stimulation through implanted contacts can also help map function and test whether specific regions participate in symptoms.
It is not a universal test
SEEG is invasive, requires specialist planning and samples only the areas where electrodes are placed. It is therefore used as part of a carefully constructed presurgical investigation, not as a routine diagnostic test for everyone with epilepsy.
SEEG begins with a hypothesis
SEEG is not a fishing expedition. Electrode trajectories are designed to test a specific three-dimensional hypothesis about where seizures begin and how they spread, based on semiology, scalp EEG, MRI and other non-invasive investigations.
What SEEG can sample
Depth electrodes can record from deep structures, buried cortex and multiple lobes simultaneously. This makes SEEG especially useful when the suspected network includes mesial temporal, insular, cingulate or other regions that are difficult to assess with surface electrodes.
When SEEG is considered
Common indications include discordant non-invasive findings, MRI-negative focal epilepsy, multiple possible lesions, suspected deep seizure onset and situations where the proposed treatment is close to eloquent cortex.
When SEEG should not be used
SEEG is useful only when there is a plausible hypothesis that can be tested and when the result could change management. Implantation without a coherent hypothesis risks sampling the wrong network.
How electrodes are implanted
Electrodes are placed through small skull openings using stereotactic planning. Modern centres use robotic, frame-based or frameless guidance depending on local expertise and equipment.
What is recorded
Teams analyse background activity, interictal discharges, seizure onset and propagation. Video remains important because electrical changes must be correlated with the person’s actual clinical seizure.
Cortical stimulation
Electrical stimulation through SEEG electrodes can help map function and may reproduce habitual symptoms or seizures. This can strengthen anatomo-electro-clinical interpretation, but stimulation findings are considered alongside spontaneous seizures.
Risks
SEEG is invasive. Important risks include intracranial haemorrhage, infection, neurological deficit and complications related to anaesthesia or electrode placement. Careful vascular imaging and trajectory planning are used to reduce risk.
SEEG and surgery
The goal is not merely to find the first electrical change. The multidisciplinary team uses the full recording to define the epileptogenic network and decide whether resection, disconnection, ablation or neurostimulation can be offered safely.
SEEG-guided radiofrequency thermocoagulation
In selected cases, contacts already implanted for SEEG can be used to create small radiofrequency lesions. This may be therapeutic or may provide additional information about whether disrupting a particular node reduces seizures.
SEEG versus subdural grids
SEEG offers three-dimensional sampling through small entry points and is well suited to deep and bilateral hypotheses. Subdural electrodes provide dense coverage of exposed cortical surfaces and may still be appropriate for particular mapping questions. Choice depends on the hypothesis, not on one technique being universally superior.
Why MRI-negative cases often need SEEG
A 2025 systematic review found that people categorised as MRI-negative undergo SEEG more often than those with visible lesions. This reflects the need for additional localisation when structural imaging does not provide a clear target.
The central point
SEEG is a hypothesis-driven method for reconstructing seizure networks in three dimensions. Its value comes from carefully chosen electrode placement and integration with the entire presurgical evaluation.
Epilepsy Surgery and Presurgical Assessment →
EEG Source Imaging →
Neurology review — SEEG in presurgical evaluation
Sources
Bancaud, Talairach and colleagues — functional stereotaxic exploration
Stereo-encephalographic presurgical evaluation: an evolving science
Stereoelectroencephalography: indication and efficacy
Evidence checked: September 2026.