3D Segmentation with an application of level set-method using MRI volumes for image guided surgery

Area of Science:

• Medical Imaging
• Surgical Planning
• Computational Anatomy

Background:

• Image-guided surgery requires accurate anatomical segmentation.
• Manual segmentation is time-consuming and less precise.
• Generic atlases lack patient-specific anatomical detail.

Purpose of the Study:

• To develop and evaluate an innovative, faster image segmentation method for image-guided surgery.
• To enhance pre-surgery planning using patient-specific 3D anatomical models.
• To compare the precision and efficiency of different segmentation techniques.

Main Methods:

• A 3D filtering module using anisotropic diffusion to preserve contours and smooth regions.
• A comparative analysis of Region Growing, Cubic Spline hand-assisted, and Level Set segmentation methods.
• A proposed Level Set-based front propagation method for brain structure reconstruction.

Main Results:

• The Level Set method demonstrated superior performance in segmenting and reconstructing internal brain structures.
• The proposed segmentation approach is faster than manual methods.
• Utilizing the patient as an anatomical reference offers higher precision than generic atlases.

Conclusions:

• The developed methodology provides a precise and efficient tool for 3D information extraction in image-guided surgery.
• The 3D visualization of segmented models significantly aids pre-surgery planning.
• This innovation advances the application of medical imaging in surgical interventions.