Artificial intelligence-driven real-time assistance in minimally invasive surgery: a technology-oriented narrative...

Area of Science:

• Surgical Technology
• Artificial Intelligence in Medicine
• Minimally Invasive Surgery

Background:

• Minimally invasive surgery offers patient benefits but increases surgeon cognitive load due to limited visualization and tactile feedback.
• Robot-assisted surgery partially addresses these challenges.
• Intraoperative artificial intelligence (AI) is emerging as a key technology for real-time surgical support.

Purpose of the Study:

• To review key AI technologies applied during the intraoperative phase of endo-laparoscopic and robotic surgery.
• To discuss clinical integration challenges and future research directions for intraoperative AI.

Main Methods:

• Review of AI technologies including anatomical and lesion recognition, instrument detection and tracking.
• Analysis of surgical phase and workflow, real-time tissue characterization, and image-guided navigation.
• Examination of AI-assisted instrument control and multimodal event detection.

Main Results:

• AI technologies reviewed cover a range of intraoperative applications in minimally invasive and robotic surgery.
• Key applications include enhanced perception, decision-making, and instrument control.
• Integration challenges and future research in learning paradigms and human-AI collaboration are identified.

Conclusions:

• Intraoperative AI is a pivotal technology for advancing minimally invasive and robotic surgery.
• Future research should focus on foundation and self-supervised learning, and collaborative human-AI systems.
• Addressing clinical integration challenges is crucial for widespread adoption.