Objective: Whether auto-segmentation matches anatomy?

Whether auto-segmentation volume difference in either side matches UPDRS score difference between two sides?

whether auto-segmentation volume depends on MRI sequence?

Whether auto-segmentation and MER matches?

Whether auto-segmentation helps programming?

Background: Judgment of STN spatial extent and subsequent targeting is often based on a mixture of individual experiences and single or multimodal visibility of the STN or neighboring structures. Targeting is often mostly based on 2-dimensional (2D) characterization of the STN rather than utilizing manual or (semi-) automatic 3D volumetric segmentations.  Such segmentation, if accurate and patient specific, can be of great help in preop planning and postop programming.

We discussed our experience in patients who underwent STN DBS (N=40) and GPI DBS (N=10)

Method: MRI was obtained using standard protocol.  DBS surgery was performed in standard way.  Electrophysiologic data was obtained.  Electrode was placed based on electrophysiology.

The electrode position was plotted in the auto-segmentation.  The results were corroborated.

The finding of monopolar review during DBS programming was corroborated with auto segmentation finding.

Results: Volume of STN and GPI based on auto-segmentation varies slightly based on sequence of MRI.

Volume of STN and GPI can be of great help and difference of volume co relates with laterality of UPDRS  score

For STN, Intra-Op MER was different from auto segmentation.  The difference was more obvious in Z axis especially for exit from STN.  Planning based on auto segmentation and without auto segmentation was comparable.

For GPI, planning based on auto segmentation and without auto segmentation was comparable as well.

Conclusion: MRI-based automatic segmentation of the STN is a viable, patient-specific targeting approach that can be used alongside traditional targeting methods in deep brain

stimulation to support preoperative planning and visualization of target structures and aid postoperative optimization of programming.

References: Reinacher PC, Várkuti B, Krüger MT, Piroth T, Egger K, Roelz R, Coenen VA.
Automatic Segmentation of the Subthalamic Nucleus: A Viable Option to Support
Planning and Visualization of Patient-Specific Targeting in Deep Brain
Stimulation. Oper Neurosurg (Hagerstown). 2019 Nov 1;17(5):497-502.

« Back to 2026 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/automatic-segmentation-of-the-subthalamic-nucleus-and-globus-pallidus-internus-can-it-support-planning-and-programming-in-deep-brain-stimulation/