MRI and CT Fusion in Stereotactic Electroencephalography (SEEG)

Area of Science:

• Neurology
• Medical Imaging
• Image Processing

Background:

• Epilepsy affects millions, with a significant portion resistant to anti-epileptic drugs.
• Surgical intervention offers an alternative for intractable epilepsy, demanding precise localization of epileptogenic zones.
• Stereotactic Electroencephalography (SEEG) is crucial for pre-surgical planning, relying on multi-modal image fusion for accurate electrode placement.

Purpose of the Study:

• To develop and validate an improved image fusion method for Stereotactic Electroencephalography (SEEG).
• To address challenges in multi-modal image registration for SEEG by accounting for implanted electrodes.
• To enhance the accuracy and reliability of epileptogenic tissue localization in epilepsy surgery planning.

Main Methods:

• Proposed a novel image fusion technique for SEEG incorporating electrode segmentation from computed tomography (CT) as a sampling mask during image registration.
• Validated the method using eight image pairs from the Retrospective Image Registration Evaluation Project (RIRE).
• Assessed efficacy by comparing Euclidean distances of reference points between registrations with and without the CT-derived sampling mask.

Main Results:

• The proposed method demonstrated comparable average registration error to methods without a sampling mask.
• Crucially, the fusion method significantly reduced the dispersion of registration errors, lowering the standard deviation from 5.25 to 0.86.
• This indicates enhanced precision and reliability in electrode localization.

Conclusions:

• The developed image fusion method effectively improves the precision of electrode localization in SEEG.
• Incorporating electrode segmentation as a sampling mask is a viable strategy to mitigate registration artifacts and misregistrations.
• This advancement holds promise for more accurate pre-surgical planning in epilepsy cases requiring surgical intervention.