Operational framework and training standard requirements for AI-empowered robotic surgery

Area of Science:

• Robotics
• Artificial Intelligence
• Machine Learning
• Surgical Technology

Background:

• Autonomous robot-delivered surgeries require advancements in human-centered artificial intelligence (AI) and transparent machine learning (ML).
• Integration of gross anatomy models and medical imaging data from cadavers is essential for performance evaluation.
• Current limitations necessitate novel approaches for AI and ML in surgical robotics.

Purpose of the Study:

• To review technological advances and state-of-the-art developments in surgical robotics and AI.
• To explore frameworks and standards for AI, emphasizing transparency in machine learning.
• To propose a methodology for enhancing AI-driven surgical procedures.

Main Methods:

• Literature search on surgical robotics, AI, and machine learning.
• Review of technological advancements and documented developments.
• Analysis of agent studies, relevant frameworks, and AI standards, including transparency aspects.

Main Results:

• A "procedure/skill template" is recommended for AI training, adaptable for surgeons.
• Metric-based training methodologies have shown over 40% error reduction in objectively assessed intraoperative procedures.
• Integration of Explainable AI, ML, and novel sensorics enhances robotic guidance and molecular-level tissue classification.

Conclusions:

• Human-centered AI and transparent ML are key for feasible autonomous robotic surgery.
• A "procedure/skill template" can standardize AI training, leading to significant error reduction.
• Explainable AI, ML, and advanced sensorics offer precise robotic guidance and tissue characterization.