Towards determining clinical factors influencing critical structure identification using Artificial Intelligence.

Area of Science:

• Surgical analytics
• Medical imaging analysis
• Artificial intelligence in medicine

Background:

• Manual analysis of surgical videos is time-consuming and limits study scale.
• Granular anatomical annotations are crucial for safety achievement studies.
• Artificial intelligence (AI) offers automated operative video analysis to overcome manual limitations.

Purpose of the Study:

• To demonstrate the real-world utility of AI video analysis in surgical studies.
• To correlate AI-derived surgical analytics with clinical factors.
• To explore the potential of AI in scaling clinical studies through automated video analysis.

Main Methods:

• Analysis of 481 laparoscopic cholecystectomy videos using AI algorithms.
• AI identified key anatomical structures (cystic duct, artery) and operative phases.
• Metrics stratified by surgeon experience and case complexity.

Main Results:

• Operative time correlated positively with increasing operative difficulty.
• Consultant surgeons showed greater proportional anatomy visualization than trainees in complex cases.
• Cystic duct identification typically preceded cystic artery identification, irrespective of complexity.

Conclusions:

• AI-driven surgical video analysis provides significant insights with patient benefits.
• AI analytics correlate with clinical factors, proving practical utility.
• Automated video analysis by AI can overcome limitations of manual review in surgical studies.