Assessment of video tracking usability for training simulators

Area of Science:

• Medical Simulation and Training
• Computer-Assisted Surgery (CAS)
• Image-Guided Interventions

Background:

• Commercial computer-assisted surgery (CAS) systems often rely on expensive tracking components.
• There is a need for cost-effective solutions in surgical training simulators.
• Accurate registration between physical models and virtual 3D representations is crucial for CAS.

Purpose of the Study:

• To develop a simulator mimicking commercial CAS systems for surgical operation tasks.
• To assess the accuracy of a low-cost video-based tracking system for CAS training.
• To determine the suitability of video tracking for achieving accurate registration in a training domain.

Main Methods:

• A simulator was built using low-cost video-based tracking instead of expensive systems.
• Two camera types were evaluated for video tracking accuracy.
• Accuracy was tested using common tests (noise, relative accuracy, pivoting) with different marker sizes (40-mm, 60-mm, 80-mm) at various distances.

Main Results:

• The best camera achieved encouraging results, with accuracies reported in millimeters.
• The 80-mm marker demonstrated superior accuracy across all tests, ranging from 2 mm (noise test) to 8 mm (registration test).
• Smaller markers (60-mm and 40-mm) showed progressively lower accuracy, indicating marker size impacts performance.

Conclusions:

• The evaluated video-based tracking approach provides sufficient accuracy for registration in CAS training.
• This low-cost solution is suitable for developing realistic surgical training simulators.
• Further development can leverage this technology to enhance surgical education and skill acquisition.