Preliminary Results Comparing Thin Plate Splines with Finite Element Methods for Modeling Brain Deformation during...

Area of Science:

• Neurosurgery
• Medical Imaging
• Computational Mechanics

Background:

• Brain shift, the deformation of the brain during tumor resection, complicates neuro-navigation by altering intraoperative anatomy.
• Accurate mapping of presurgical images to intraoperative patient coordinates is crucial for enhancing surgical precision and utility.

Purpose of the Study:

• To compare the accuracy of two brain deformation modeling techniques: thin plate spline interpolation and finite element method (FEM).
• To evaluate the efficacy of these models in compensating for brain shift during clinical tumor resections.

Main Methods:

• Two computational models, thin plate spline and FEM, were employed to predict brain deformation.
• Displacements for model input were derived from automatically detected features (SIFT-Rank algorithm) around the tumor.
• Deformation prediction accuracy was validated against manually identified landmarks.

Main Results:

• The finite element method (FEM) demonstrated significantly superior performance in predicting landmark deformation compared to the spline-based approach.
• Both methods were capable of modeling brain deformations within practical time frames for operating room use.
• FEM required more extensive preprocessing than the spline method.

Conclusions:

• Finite element method (FEM) models, incorporating biophysical and geometric constraints, offer higher accuracy for brain shift compensation in neurosurgery.
• Advanced modeling techniques like FEM hold promise for improving the accuracy and reliability of image-guided neuro-navigation.