Speech and motion control for interventional radiology: requirements and feasibility

Area of Science:

• Medical Imaging
• Human-Computer Interaction
• Radiology

Background:

• Interventional radiology requires sterile environments, necessitating hands-free control for image review.
• Existing methods for image manipulation can be cumbersome in sterile settings, impacting efficiency.

Purpose of the Study:

• To evaluate a speech and motion control system for image review in interventional radiology.
• To define the requirements and limitations of such a system in a clinical test bed.

Main Methods:

• Implemented motion control using Microsoft Kinect framework.
• Tested system with 10 participants using voice and gesture commands under varying light conditions.
• Identified essential voice/gesture commands for interventional radiology procedures.

Main Results:

• Achieved a 93% command recognition rate, with speech commands outperforming gestures.
• No significant impact of lighting conditions on recognition accuracy was observed.
• Accidental command triggering occurred in 30% of cases; 60% of participants would consider routine use.

Conclusions:

• Speech and motion control offer a viable no-touch interface for radiological image handling in sterile environments.
• High and stable command recognition rates were demonstrated across different lighting.
• Reducing accidental command activation is crucial for widespread clinical adoption.