Robotic and artificial intelligence for keyhole neurosurgery: the ROBOCAST project, a multi-modal autonomous path...

Area of Science:

• Robotics
• Artificial Intelligence
• Neurosurgery
• Medical Imaging

Background:

• Computer-assisted surgery requires advanced robotic and sensor integration.
• Keyhole neurosurgery presents challenges in precision and safety for procedures like tumor biopsy and drug delivery.

Purpose of the Study:

• To develop and evaluate an intelligent trajectory planner for robot-assisted keyhole neurosurgery.
• To integrate pre-operative and intra-operative data for enhanced surgical guidance.
• To improve surgeon assistance through an intelligent high-level controller (HLC).

Main Methods:

• The ROBOCAST project utilizes an intelligent high-level controller (HLC).
• The HLC integrates data from surgeons, diagnostic images, and on-field sensors.
• An intelligent trajectory planner based on risk estimation and human criticism was developed and tested.

Main Results:

• The paper details the overall architecture and focuses on the intelligent trajectory planner.
• Initial tests using a real image stack and risk descriptor phantom demonstrate the planner's functionality.
• Fuzzy risk description allows for on-field knowledge upgrades without expert intervention.

Conclusions:

• The ROBOCAST project advances robot-assisted neurosurgery through intelligent control and planning.
• The intelligent trajectory planner enhances safety and precision in keyhole procedures.
• Fuzzy logic offers a flexible approach to risk assessment and system adaptation in surgical robotics.