An integrated tool for automated visualization of subdural electrodes in epilepsy surgery evaluation.

PURPOSE

An automated tool embedded in image processing and visualization software should be developed to visualize subdural electrodes on the human cortical surface of the brain.

METHODS

For accurate positioning of electrodes on the cortical surface, preoperative magnetic resonance imaging (MRI) and postoperative computed tomography (CT) datasets of pharmacoresistant epilepsy patients were co-registered and segmented. For the brain segmentation of preoperative MRI a new algorithm was implemented. To account for the radial electrode displacement due to cortical deformation, a radial replacement function was developed. The automated tool was evaluated using intraoperative photographs and using a bimodal phantom.

RESULTS

The tool visualizes electrodes fully automated in 1-5 min. The mismatch between calculated and reference electrode position derived from intraoperative photographs was <or=9 mm, and was based upon phantom data <or=1.6 mm.

CONCLUSIONS

Empirical investigations have to be made to quantify the magnitude and the nature of transverse electrode displacement in order to investigate the practicability and to determine the accuracy of the developed tool. Depending on investigation results, a half-automated intraoperative photography based method might be an option to minimize electrode shift.