Experience-based SEEG planning: from retrospective data to automated electrode trajectories suggestions

Area of Science:

• Neurosurgery
• Medical Imaging
• Computational Neuroscience

Background:

• Stereo-ElectroEncephaloGraphy (SEEG) is crucial for identifying the epileptogenic focus.
• Current SEEG planning is time-consuming and lacks automated initial trajectory generation.
• Existing methods focus on optimizing trajectories, not creating initial sets.

Purpose of the Study:

• To develop a methodology for generating initial SEEG electrode trajectories based on retrospective data.
• To automate and expedite the SEEG planning process.
• To represent clinical practice through average trajectories for new patient cases.

Main Methods:

• Analysis of retrospective SEEG planning data from 75 anonymized patients.
• Creation of average trajectories and exploratory patterns in an average brain space.
• Validation of trajectory mapping and comparison with manual planning.

Main Results:

• Successfully mapped 90% of generated trajectories in a test set.
• Optimized trajectories showed comparable or superior safety metrics (distance from vessels, insertion angle) to manual plans.
• Identified eight distinct neurosurgical planning strategies from the data.

Conclusions:

• The proposed methodology effectively generates initial SEEG electrode trajectories.
• This approach can significantly reduce planning time and potentially improve safety.
• Understanding planning strategies offers insights into neurosurgical decision-making for epilepsy treatment.