Brain mapping in stereotactic surgery: a brief overview from the probabilistic targeting to the patient-based anatomic mapping.

CHU Clermont-Ferrand, Hôpital Gabriel Montpied, Service de Neurochirurgie A, Clermont-Ferrand, F-63003, France.
NeuroImage (Impact Factor: 6.13). 02/2007; 37 Suppl 1:S109-15. DOI: 10.1016/j.neuroimage.2007.05.055
Source: PubMed

ABSTRACT In this article, we briefly review the concept of brain mapping in stereotactic surgery taking into account recent advances in stereotactic imaging. The gold standard continues to rely on probabilistic and indirect targeting, relative to a stereotactic reference, i.e., mostly the anterior (AC) and the posterior (PC) commissures. The theoretical position of a target defined on an atlas is transposed into the stereotactic space of a patient's brain; final positioning depends on electrophysiological analysis. The method is also used to analyze final electrode or lesion position for a patient or group of patients, by projection on an atlas. Limitations are precision of definition of the AC-PC line, probabilistic location and reliability of the electrophysiological guidance. Advances in MR imaging, as from 1.5-T machines, make stereotactic references no longer mandatory and allow an anatomic mapping based on an individual patient's brain. Direct targeting is enabled by high-quality images, an advanced anatomic knowledge and dedicated surgical software. Labeling associated with manual segmentation can help for the position analysis along non-conventional, interpolated planes. Analysis of final electrode or lesion position, for a patient or group of patients, could benefit from the concept of membership, the attribution of a weighted membership degree to a contact or a structure according to its level of involvement. In the future, more powerful MRI machines, diffusion tensor imaging, tractography and computational modeling will further the understanding of anatomy and deep brain stimulation effects.

1 Bookmark
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This study aims to evaluate the improvements of cardinal motor symptoms depending on the stimulation site relative to a standardized, reconstructed three-dimensional MRI-defined subthalamic nucleus (STN.) This retrospective, clinical study includes 22 patients with idiopathic Parkinson's disease, who consecutively underwent bilateral subthalamic nucleus stimulation. Intraoperative microelectrode recording and clinical testing were performed. The location of the best stimulation site, found intraoperatively, and the positions of the active electrode contacts 12 months after the operation were correlated to a standardized, reconstructed three-dimensional MRI-defined STN. Further, the impact of the stimulation site on rigidity, tremor and akinesia was analysed. Significant improvement of the contralateral akinesia was observed if the intraoperative stimulation site was located more lateral and superior in the MRI-STN. Furthermore, active electrode contacts located superior to or in the superior part of the MRI-STN had a significantly better effect on the tremor of the contralateral hand than in other locations inside the STN. For rigidity and akinesia, these correlations were statistically not significant. Although we found significantly better results for tremor suppression in superior and lateral aspects of the STN, for overall clinical improvement, several patients fared better with randomly distributed stimulation sites in medial, posterior or inferior parts of the MRI-defined STN. Locations of stimulation sites with the best improvements of motor symptoms were distributed randomly throughout the whole MRI-defined STN, indicating that MRI-based targeting alone is not sufficient, but intraoperative clinical testing is necessary to determine the optimal stimulation site for each individual patient.
    Neurosurgical Review 02/2014; · 1.86 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To assess stable anatomical features of the human thalamus, an unbiased diffusion tensor parcellation approach was used to segment thalamic substructures with similar spatial orientation. We determined localization, size and individual variations of 21 thalamic clusters in a group of 63 healthy human subjects (32 males/31 females). The laterality differences accounted for ±6 % and gender differences for ±4 % of the thalamic volume. Consecutively, five stable clusters in the anterior, medial, lateral and posterior thalamus were selected, which were common to 90 % of all subjects and contained at least 10 voxels. These clusters could be assigned to the anteroventral nucleus (AN) group, the mediodorsal (MD) nucleus, the medial pulvinar (PuM), and the lateral nuclei group. The subcortical and cortical connectivity of these clusters revealed that: (1) the oblique cranio-caudal-oriented fibers of the AN cluster mainly connect to limbic structures, (2) the numerous dorso-frontal-oriented fibers of MD mainly project to the prefrontal cortex and the medial temporal lobe, (3) the fibers of the PuM running in parallel with the x-axis project to medio-occipital and medio-temporal areas and connect visual areas with the hippocampus and amygdala and via intrathalamic pathways with medio-frontal areas, and (4) the oblique caudo-cranial fibers of the two lateral clusters located anteriorly in the motor and posteriorly in the sensory thalamus are routing sensory-motor information from the brain stem via the internal capsule to pre- and peri-central regions of the cortex.
    Brain Structure and Function 03/2014; · 7.84 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: High quality segmentation of diffusion tensor images (DTI) is of key interest in biomedical research and clinical application. In previous studies, most efforts have been made to construct predefined metrics for different DTI segmentation tasks. These methods require adequate prior knowledge and tuning parameters. To overcome these disadvantages, we proposed to automatically learn an adaptive distance metric by a graph based semi-supervised learning model for DTI segmentation. An original discriminative distance vector was first formulated by combining both geometry and orientation distances derived from diffusion tensors. The kernel metric over the original distance and labels of all voxels were then simultaneously optimized in a graph based semi-supervised learning approach. Finally, the optimization task was efficiently solved with an iterative gradient descent method to achieve the optimal solution. With our approach, an adaptive distance metric could be available for each specific segmentation task. Experiments on synthetic and real brain DTI datasets were performed to demonstrate the effectiveness and robustness of the proposed distance metric learning approach. The performance of our approach was compared with three classical metrics in the graph based semi-supervised learning framework.
    PLoS ONE 03/2014; 9(3):e92069. · 3.53 Impact Factor

Full-text (2 Sources)

Available from
May 26, 2014