Towards automatic quantification of operating table interaction in operating rooms.

Area of Science:

• Surgical Technology
• Human-Computer Interaction
• Operating Room Workflow Analysis

Background:

• Global perioperative staff shortages necessitate efficient and motivating work environments.
• Increasingly sophisticated operating room technologies are shifting staff focus from patients to technology.
• Objective measurement of this shift is crucial for designing supportive tech and optimizing high-tech procedures.

Purpose of the Study:

• To objectively measure the shift in staff interaction from patients to technology during surgery.
• To compare staff interaction levels across different surgical technology levels: open surgery (OS), minimally invasive surgery (MIS), and robot-assisted surgery (RAS).
• To assess the potential of an algorithm for detecting operating table interaction based on staff posture and movement.

Main Methods:

• Recorded 35 gynecological procedures across OS, MIS, and RAS.
• Annotated staff-patient and staff-operating table interactions.
• Developed and proposed an algorithm to detect operating table interaction from staff posture and movement data.

Main Results:

• The algorithm measured operating table interactions at 70.4% (OS), 70.3% (MIS), and 30.1% (RAS).
• Patient interaction percentages were 37.6% (OS), 38.3% (MIS), and 24.6% (RAS).
• Interaction peaks correlated with anomalous events or workflow phase transitions.

Conclusions:

• Robot-assisted surgery (RAS) demonstrated significantly less operating table and patient interaction compared to OS and MIS.
• The proposed algorithm shows similar trends to manual annotations, indicating potential for workflow phase detection.
• Challenges in pose tracking due to operating room complexity can affect algorithm input quality.