An atlas-based method to compensate for brain shift: preliminary results.

Area of Science:

• Neurosurgery
• Medical Imaging
• Computational Biology

Background:

• Intraoperative brain shift poses a challenge to image-guided neurosurgery.
• Accurate compensation requires understanding deformation sources like brain sag and edema.
• Computational models offer a promising solution for real-time adjustments.

Purpose of the Study:

• To develop and validate a strategy for compensating intraoperative brain shift.
• To predict distributed brain shift using preoperatively computed deformation models.
• To enhance the accuracy of image-guided neurosurgical systems.

Main Methods:

• A preoperative atlas of brain deformations was computed for various loading conditions.
• A constrained linear inverse model was employed to predict intraoperative brain shift.
• The approach was validated using phantom experiments, in vivo cases, and simulations.

Main Results:

• The method achieved an average of 93% surface shift correction across all tested scenarios.
• Subsurface shift correction reached 85% in simulation and phantom experiments.
• Remaining average errors were 0.7+/-0.3 mm for surface and 1.0+/-0.4 mm for subsurface shift.

Conclusions:

• The proposed inverse finite-element approach effectively predicts and compensates for intraoperative brain shift.
• This method has the potential to significantly improve the precision of image-guided neurosurgery.
• Preoperative modeling of brain deformations enhances the accuracy of surgical navigation systems.