Lead-DBS: A toolbox for deep brain stimulation electrode localizations and visualizations
Introduction
Deep brain stimulation surgery (DBS) is a highly efficacious treatment option for patients with severe movement disorders such as Parkinson's disease (PD), dystonia and essential tremor (ET). Improvement in motor symptoms as well as quality of life have been proven in several multicenter studies for PD (Deuschl et al., 2006, Krack et al., 2003, Schupbach, 2005); for dystonia (Kupsch et al., 2006, Mueller et al., 2008, Vidailhet et al., 2005, Volkmann et al., 2012); and for ET (Hariz et al., 2008, Schuurman et al., 2000).
Additionally to the DBS stimulation parameters, the correct anatomical description of the electrode contact placements determines the area that is being stimulated and has a major influence on the clinical outcome of chronic DBS. Studies in PD patients undergoing STN DBS have shown that accuracy of electrode placement within the anatomical target region correlates with the motor improvement during DBS (for example Frankemolle et al., 2010, Welter et al., 2014, Wodarg et al., 2012). Similarly, the degree of motor improvement with DBS in dystonia was related to electrode position within the pallidum (Tisch et al., 2007, Schönecker et al., 2014). Although other factors influence the clinical outcome of DBS (Lumsden et al., 2013, Vidailhet et al., 2005), contact mislocation is considered to be the most common cause of a poor clinical response (Ellis et al., 2008, Marks et al., 2009). Accordingly, knowledge about localizations of contacts can be supportive in clinical routines to evaluate DBS effects and may help to define the best contact for chronic DBS, especially in more complex electrode designs with multiple contacts that will be available in the near future.
Moreover, information on electrode localization is crucial for various scientific approaches using depth recordings from DBS target regions of the basal ganglia (Brown and Williams, 2005) or stimulation protocols to evaluate the involvement of the anatomical DBS target region in information processing (Cavanagh et al., 2011, Frank et al., 2007, Green et al., 2013).
Evaluation of electrode placement can be achieved by postoperative magnetic resonance (MR)- or computer tomography (CT)- imaging. Depending on the DBS center, usually one imaging method is established in clinical routines. Electrode placement is most often denoted relative to the anterior commissure–posterior commissure (AC–PC) stereotactic space (in a systematic review, electrode coordinates in this notation were found in 8 of 13 studies; Caire et al., 2013) or within the standardized MNI space (e.g. see Schönecker et al., 2009, Witt et al., 2013). The former method requires manual measurements that are prone to significant inaccuracies (Pallavaram et al., 2008) and does not take anatomical inter-subject variability into account (Starr et al., 1999, Zhu et al., 2002). On the other hand, warping subject-specific anatomic images into a well-defined standard space as the MNI space makes electrode placements comparable over subjects and DBS centers. This also makes it possible to set electrode contacts into relationship with atlas data of target regions (for examples of available subcortical atlas data, e.g. see Jakab et al., 2012, Keuken et al., 2013, Keuken et al., 2014, Prodoehl et al., 2008, Sarnthein et al., 2013, Tzourio-Mazoyer et al., 2002, Yelnik et al., 2003).
After the normalization of MR-images, a time-consuming and observer-dependent manual survey of electrode contacts has usually to take place to determine their positions. This can be done by manually analyzing the terminal portion of the quadripolar DBS electrodes composed of four metallic (platinium/iridium) non-insulated contacts at equidistant intervals which generate susceptibility artifacts on the postoperative MR image (Schönecker et al., 2009). The centers of the artifacts show hypo-intense and represent the centers of the electrode contacts (Pollo et al., 2004, Yelnik et al., 2003; Fig. 1). Usually, this is performed by using a slice-based (two-dimensional, 2D) MR-viewing software, and fiducial landmarks are manually placed upon the electrode contact artifacts.
To overcome the necessity of a manual survey of electrode contacts and to increase precision in this process, we introduce a toolbox to determine the electrode contact coordinates in a semi-automated design. The primary goal is to enable the user with a tool that provides a good starting-point for fast manual fine adjustment of DBS contact localization.
Section snippets
Patients and imaging
In total, fifty patients (33 male; mean age 44.5 ± 17.45 yrs [mean ± SD], range 13–75 yrs) that underwent DBS surgery were analyzed in this study. Two patients received four, the others received two quadripolar DBS leads bilaterally (Medtronic, Minneapolis, Minnesota, US). To include different electrode target regions, clinical indications and electrode models, patients were randomly selected from a large database of DBS patients treated at our center (~ 550 patients). Forty patients received
Reconstruction results
In MR imaging, using default parameters, 72 of 80 (90%) reconstructions found the correct trajectory. Fifty-five of 80 (69%) electrode contact heights were correctly reconstructed using default parameters. By slightly adjusting individual parameters in cases that yielded faulty reconstructions, a total of 78 (98%) trajectory reconstructions and 61 (76%) electrode contact heights were reconstructed correctly. Such parameter adjustments involved whether the images should be slightly smoothed or
Discussion
In this study, we introduce an automated DBS electrode reconstruction algorithm. Analysis of a large group of patients (n = 50) showed that the algorithm can facilitate and improve the process of electrode and trajectory localization based on post-operative MR- or CT-images.
Acknowledgments
This work was supported by the German Research Foundation (DFG), KFO247. We would like to thank Dr. Evgeniya Kirilina and Thomas Schönecker for their help and thoughtful comments. The toolbox works as an add-on to SPM8 and uses subroutines authored by Jimmy Shen (“Tools for NIfTI and ANALYZE image”), Francois Bouffard (“Draggable”), John D'Errico (“Inhull”), Anders Brun (“Myaa — My Anti-Alias for Matlab”) and Yi Cao (“Rangesearch”) that are covered by a BSD-license allowing for modification and
References (66)
- et al.
Basal ganglia local field potential activity: character and functional significance in the human
Clin. Neurophysiol.
(2005) - et al.
Current steering to control the volume of tissue activated during deep brain stimulation
Brain Stimul.
(2008) - et al.
Mood and behavioural effects of subthalamic stimulation in Parkinson's disease
Lancet Neurol.
(2014) - et al.
Whole brain segmentation: automated labeling of neuroanatomical structures in the human brain
Neuron
(2002) - et al.
Unbiased average age-appropriate atlases for pediatric studies
NeuroImage
(2011) - et al.
Reduction of influence of task difficulty on perceptual decision making by STN deep brain stimulation
Curr. Biol.
(2013) - et al.
Quantifying inter-individual anatomical variability in the subcortex using 7 T structural MRI
NeuroImage
(2014) - et al.
Evaluation of 14 nonlinear deformation algorithms applied to human brain MRI registration
NeuroImage
(2009) - et al.
A mean three-dimensional atlas of the human thalamus: generation from multiple histological data
NeuroImage
(2010) - et al.
Confirmation of functional zones within the human subthalamic nucleus: patterns of connectivity and sub-parcellation using diffusion weighted imaging
NeuroImage
(2012)