Optimizing registration uncertainty visualization to support intraoperative decision-making during brain tumor...

Area of Science:

• Neurosurgery
• Medical Visualization
• Image-Guided Surgery

Background:

• Brain shift and other factors introduce uncertainty in image-guided surgery.
• Accurate tumor localization and resection are critical to avoid damaging healthy brain tissue.
• Effective communication of surgical uncertainty is essential for informed decision-making.

Purpose of the Study:

• To develop and evaluate a novel visualization software (UVisExplore) for exploring uncertainty in image registration during neurosurgery.
• To assess the effectiveness of various uncertainty visualization techniques in a simulated surgical environment.
• To improve neurosurgeons' understanding and management of navigational uncertainty.

Main Methods:

• Developed UVisExplore, a software module integrating classic and novel uncertainty visualization techniques.
• Introduced a game-based approach to quantitatively evaluate visualization effectiveness in a simulated tumor resection task.
• Conducted a study with clinicians and computer scientists to gather feedback on visualization preferences.

Main Results:

• Surgeon participants favored surgeon-centric visualizations emphasizing uncertainty near the surgical probe.
• Explicit numerical uncertainty measures in millimeters were preferred by surgeons.
• The game provided valuable insights into user preferences and interpretation of uncertainty visualizations.

Conclusions:

• An open-source 3D Slicer module (UVisExplore) and a game for visualizing registration uncertainty were developed.
• UVisExplore serves as a platform for comparing uncertainty visualization techniques in simulated surgical scenarios.
• The tools are valuable for communication and have narrowed down candidate visualizations for future in-vivo testing.