Same Patient, Same Space, Divergent Needs: Revealing Gaps and Design Opportunities in Surgeon–Anesthesiologist Collaboration

Paper Title

Same Patient, Same Space, Divergent Needs: Revealing Gaps and Design Opportunities in Surgeon–Anesthesiologist Collaboration

Publication Info

• Topic area: Intraoperative collaboration and AI-supported system design for surgeon–anesthesiologist teamwork.
• Keywords: Surgeon–anesthesiologist collaboration, intraoperative communication, AI in surgery, operating room teamwork, multimodal systems, HCI in healthcare, surgical phase recognition, risk assessment, intelligent OR technologies.

Background and Problem

• Problem / challenge: Surgeon–anesthesiologist collaboration in the operating room is hindered by communication breakdowns, asymmetrical information exchange, and differing work models. Existing HCI research has not adequately addressed these interspecialty dynamics or explored AI-driven solutions.
• Significance: Effective collaboration between surgeons and anesthesiologists is critical for patient safety, surgical outcomes, and workflow efficiency. Addressing these gaps can reduce adverse events and improve intraoperative decision-making.
• Motivation and related work: Prior studies have documented miscommunication in the OR, with up to 30% of communication attempts failing. Existing technological solutions focus on single specialties or intra-specialty collaboration, neglecting the unique needs of the surgeon–anesthesiologist dyad. Advances in AI, such as surgical phase recognition and multimodal sensing, offer opportunities to bridge this gap.

Solution

• Proposed approach: A multimodal AI-supported system tailored to the divergent needs of surgeons and anesthesiologists, providing role-specific user interfaces (UIs) and real-time situational awareness.
• Novelty:
  1. Empirical identification of seven key challenges in surgeon–anesthesiologist collaboration through focus groups and observations of 45 surgeries.
  2. Conceptualization of role-specific UIs addressing asymmetrical information needs and cognitive loads.
  3. Integration of state-of-the-art AI methods for surgical phase recognition, event prioritization, and natural language interaction.
  4. Introduction of a triage-based event notification system for real-time decision support.
• Procedure and key techniques:
  • Conducted focus groups with six surgeons and six anesthesiologists to identify challenges and design needs.
  • Observed 45 surgeries (open, laparoscopic, robotic) to analyze intraoperative dynamics.
  • Developed UI sketches and conceptualized backend AI modules, including surgical scene understanding, event triage, and language-based interaction.

Results

• Concrete findings:
  • Identified seven challenges: differing work models, information overload, superficial information exchange, underestimation of surgical risks, inter-specialty knowledge gaps, missed information due to split attention, and verbal communication challenges during cognitive overload.
  • Proposed role-specific UIs: detailed surgical progress and risk assessment tools for anesthesiologists; concise patient status and timing feedback for surgeons.
• Advantage over baselines:
  • Enhanced situational awareness and decision-making through real-time updates and predictive insights.
  • Reduced communication friction by aligning information delivery with each role’s needs.
  • Improved anticipation of surgical events and complications via AI-driven risk assessment and event prioritization.
• Experiments / evaluation:
  • Focus groups and observations provided qualitative insights into intraoperative collaboration.
  • UI components and backend modules were conceptually designed but require future simulation-based evaluation.
• Limitations and future work:
  • Findings are based on general and visceral surgery; applicability to other specialties needs exploration.
  • Proposed system has not been tested in real or simulated OR environments.
  • Future work will focus on evaluating system impact on decision timing, collaboration, and communication.

Summary

This study investigates the challenges in surgeon–anesthesiologist collaboration and proposes a multimodal AI-supported system to address them. Through focus groups and observations of 45 surgeries, seven key challenges were identified, including asymmetrical information exchange and cognitive overload. The proposed system features role-specific UIs and AI-driven modules for surgical phase recognition, event triage, and risk assessment. These innovations aim to enhance situational awareness, reduce communication friction, and improve intraoperative decision-making. Future work will evaluate the system in simulated environments to assess its impact on collaboration and patient safety.