Semi-automatic stereotactic coordinate identification algorithm for routine localization of Deep Brain Stimulation electrodes

• Published in: Journal of neuroscience methods Vol. 187; no. 1; pp. 114 - 119
• Main Authors: Hebb, Adam O., Miller, Kai J.
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 15.03.2010
• Subjects: Algorithms; Automation; Brain - diagnostic imaging; Brain - pathology; Computed Tomography; Computer image processing; Computerized image analysis; Deep Brain Stimulation; Deep Brain Stimulation - instrumentation; Deep Brain Stimulation - methods; Electrodes, Implanted; Extended Hounsfield unit; Humans; Information Theory; Magnetic Resonance Imaging; Movement disorders; Stereotaxic Techniques; Stereotaxy; Tomography, X-Ray Computed - methods; Computerized image analysis; Deep Brain Stimulation; Magnetic Resonance imaging; Computed Tomography; Computer image processing; Stereotaxy; Movement disorders; Extended Hounsfield unit
• ISSN: 0165-0270; 1872-678X; 1872-678X
• DOI: 10.1016/j.jneumeth.2009.12.016

Deep Brain Stimulation (DBS) is a routine therapy for movement disorders, and has several emerging indications. We present a novel protocol to define the stereotactic coordinates of metallic DBS implants that may be routinely employed for validating therapeutic anatomical targets. Patients were referred for troubleshooting or new DBS implantation. A volumetric MRI of the brain obtained prior to or during this protocol was formatted to the Anterior Commissure–Posterior Commissure (AC–PC) coordinate system. Patients underwent a CT scan of the brain in an extended Hounsfield unit (EHU) mode. A semi-automatic detection algorithm based on a Normalized Mutual Information (NMI) co-registration method was implemented to measure the AC–PC coordinates of each DBS contact. This algorithm was validated using manual DBS contact identification. Fifty MRI-CT image pairs were available in 39 patients with a total of 336 DBS electrodes. The median and mean Euclidean distance errors for automatic identification of electrode locations were 0.20 mm and 0.22 mm, respectively. This method is an accurate method of localization of active DBS contacts within the sub-cortical region. As the investigational indications of DBS expand, this method may be used for verification of final implant coordinates, critical for understanding clinical benefit and comparing efficacy between subjects.