Anatomical Region Segmentation for Objective Surgical Skill Assessment with Operating Room Motion Data

Area of Science:

• Medical Imaging
• Surgical Technology
• Computational Anatomy

Background:

• Existing surgical motion analysis methods are limited to specific tasks or surgery types.
• Analyzing instrument motion relative to anatomical structures requires precise segmentation.
• Anatomical region segmentation is crucial for advanced surgical motion analysis.

Purpose of the Study:

• To develop an automated anatomical region segmentation algorithm for skull base and sinus surgery.
• To enable objective analysis of surgical instrument motion data based on anatomical context.
• To improve the efficiency and accuracy of surgical motion data clustering.

Main Methods:

• Manual segmentation of a skull base atlas into nine key anatomical regions.
• Utilizing six fiducial features for initial alignment of computed tomography (CT) scans.
• Employing B-spline deformable registration and automatic bony boundary extraction for precise segmentation.
• Applying the deformation field to the atlas and clustering motion data accordingly.

Main Results:

• Successfully segmented eight maxillofacial CT scans using the proposed method.
• Clustered motion data into nine distinct groups per dataset, effectively filtering outliers.
• Demonstrated improved efficiency in motion data clustering with limited manual input.

Conclusions:

• The developed algorithm enhances the efficiency of surgical motion data clustering.
• Anatomical region segmentation effectively isolates and categorizes motion data relevant to surgical sites.
• The method offers a more objective and anatomically informed approach to surgical motion analysis.