Introduction: Patients with medically refractory epilepsy often require intracranial electrode implantation for seizure localization. As the use of these procedures continues to increase, the demand for simpler and more efficient methods of electrode localization methods has grown. We developed a semi-automated pipeline to localize electrode contacts using multimodal neuroimaging data.
Methods: Our pipeline co-registers the pre-op MRI and the post-op CT for each patient using a rigid body transformation and scaling to align electrodes in patient-specific anatomy. It then extracts a volumetric brain image to create a mask that represents the dataspace of interest. We wrote customized code to automatically determine image-specific intensity thresholds, distinguish contacts fused by artifact signal, and output contact coordinates. Coordinates are integrated with a sophisticated three-dimensional model for visualization. This pipeline was initially validated in a sample of 6 patients with 785 intracranial electrodes. We evaluated coordinate accuracy by total contact count and visual inspection of coordinates plotted on the post-op CT.
Results: Coordinates for 780/785 (99.4%) contacts were successfully identified. Specificity for contacts was 99.6% due to detection of three anchor bolts in one patient that were within the cranial vault and therefore mis-categorized as contacts. All extra-cranial contacts were successfully excluded. We subsequently used our method to successfully visualize the anatomical segregation of various response types during a Stroop-like task across >1500 contacts in 19 patients.
Conclusions: The method described is an accurate and easily implemented approach to intracranial electrode localization using MATLAB and open access tools. Compared to similar open access methods, our pipeline requires minimal user input and significantly reduces person-hours required for task completion. This method allows for seamless analysis of electrode locations and thus can be widely used in clinical and electrophysiological research.
Patient Care: As mentioned in the introduction, the ability to accurately and efficiently define precise locations of intracranial electrodes is essential for drawing meaningful conclusions from research. From a translational perspective, this technique could be employed for neuroelectrophysiology research to elucidate mechanisms of important cognitive processes, such as memory and mood regulation, that are impaired in pathological states and can inform therapeutic targets.
Learning Objectives: 1) Describe the importance of computer-based image processing for advancing neurological electrophysiology and clinical research
2) Describe a novel and highly accurate method for automatically isolating individual electrode leads within an image and creating patient-specific three dimensional visualizations
References: Dale, A.M., Fischl, B., Sereno, M.I., 1999. Cortical surface-based analysis. I. Segmentation and surface reconstruction. Neuroimage 9, 179-194.
Fedorov A., Beichel R., Kalpathy-Cramer J., Finet J., Fillion-Robin J-C., Pujol S., Bauer C., Jennings D., Fennessy F., Sonka M., Buatti J., Aylward S.R., Miller J.V., Pieper S., Kikinis R. 3D Slicer as an Image Computing Platform for the Quantitative Imaging Network. Magn Reson Imaging. 2012 Nov;30(9):1323-41. PMID: 22770690.
S.M. Smith. Fast robust automated brain extraction. Human Brain Mapping, 17(3):143-155, November 2002.
M. Jenkinson, C.F. Beckmann, T.E. Behrens, M.W. Woolrich, S.M. Smith. FSL. NeuroImage, 62:782-90, 2012.