When do they StOP?: A first step toward automatically identifying team communication in the operating room.

Area of Science:

• Computer Vision
• Surgical Workflow Analysis
• Human-Computer Interaction

Background:

• Effective team communication in the operating room (OR) is vital for surgical performance and patient safety.
• Automated identification of OR team communication is crucial for developing advanced surgical workflow analysis and intra-operative support systems.

Purpose of the Study:

• To develop and evaluate a novel approach for detecting group communication activities, specifically the team Time-out and StOP protocols, within surgical operations.
• To create the first dataset with temporal annotations of group activities in the OR to facilitate research in this area.

Main Methods:

• A new dataset, Team-OR, comprising over 100 hours of multi-view surgical videos from real surgeries was generated.
• Temporal annotations for 33 Time-out and 22 StOP protocols were created within the Team-OR dataset.
• A novel group activity detection method was proposed, encoding both scene context and action features using an efficient neural network.

Main Results:

• The proposed group activity detection approach demonstrated superior performance compared to existing state-of-the-art temporal action detection methods on the Team-OR dataset.
• Experimental results highlighted the novelty and significance of the Team-OR dataset due to the limited prior research on group activities in surgical settings.

Conclusions:

• The study successfully investigated team communication protocols (Team Time-Out and StOP) in the OR by introducing the first temporally annotated dataset of group activities.
• A novel group activity detection approach was presented, outperforming existing methods and paving the way for improved computer-assisted surgical analysis.