Enhancing Patient Acceptance of Robotic Ultrasound through Conversational Virtual Agent and Immersive Visualizations

Area of Science:

• Medical technology
• Human-computer interaction
• Artificial intelligence

Background:

• Robotic ultrasound systems offer diagnostic potential but face patient acceptance challenges.
• Patient comfort and trust are critical for adopting autonomous medical procedures.

Purpose of the Study:

• To develop and evaluate a novel robotic ultrasound system integrating an AI virtual agent and mixed reality (MR) visualizations.
• To enhance patient comfort, trust, and acceptance during robotic ultrasound examinations.

Main Methods:

• A system combining a large language model (LLM)-powered AI virtual agent with three MR visualizations (AR, AV, VR) was developed.
• Participants underwent agent-guided robotic ultrasound scans comparing MR visualizations against a standard procedure.
• The AI agent provided conversational support and reassurance, animated as controlling the ultrasound probe.

Main Results:

• Significant improvements in patient trust, acceptance, and comfort were observed across all MR visualization conditions compared to the standard robotic ultrasound.
• The LLM enabled natural dialogue, answering patient questions and providing real-time reassurance.
• Different MR modalities (AR, AV, VR) offered varying levels of immersion and patient experience.

Conclusions:

• Integrating AI virtual agents and MR visualizations is a promising strategy to increase patient acceptance of robotic medical procedures.
• Design insights for future MR visualizations and virtual agents can further optimize patient comfort in autonomous healthcare.
• This approach paves the way for more human-centered autonomous diagnostic systems.