Stereovision-based integrated system for point cloud reconstruction and simulated brain shift validation.

Area of Science:

• Neurosurgery
• Medical Imaging
• Computer-Aided Surgery

Background:

• Intraoperative brain shift, or soft tissue deformation, significantly impacts the accuracy of image-guided surgery navigation systems.
• Existing methods for compensating for brain shift often rely on complex computational models and sparse data.

Purpose of the Study:

• To present a novel mock environment for acquiring stereoimages from a tracked operating microscope.
• To reconstruct 3D point clouds for intraoperative brain shift compensation.
• To evaluate the accuracy and practicality of using a tracked stereoscopic microscope for data acquisition.

Main Methods:

• Developed a mock environment to capture stereoimages using a microscope tracked by an optical system.
• Reconstructed 3D point clouds from stereoimages and registered them to compute displacements.
• Generated realistic cortical deformations using a mock craniotomy device.
• Validated measurements against an optically tracked stylus, achieving an average displacement error of [Formula: see text].

Main Results:

• Achieved a reconstruction error of 1 mm using a phantom with known geometry.
• Demonstrated that a tracked stereoscopic microscope can effectively measure mock vessel displacements.
• The tracked microscope system showed practical utility in collecting intraoperative data for brain shift correction.

Conclusions:

• A tracked stereoscopic microscope is a viable alternative to laser range scanners for acquiring data to correct brain shift.
• This method offers a practical and potentially cost-effective approach to enhance neurosurgical navigation accuracy.