Preference elicitation: Obtaining gestural guidelines for PACS in neurosurgery

Area of Science:

• Neurosurgery
• Human-Computer Interaction
• Medical Imaging

Background:

• Accessing medical records during neurosurgery is crucial.
• Touchless gestural interfaces can reduce infection risk in the Operating Room (OR) by enabling hands-free browsing of Picture Archiving and Communication Systems (PACS).

Purpose of the Study:

• To elicit preferred gestures from neurosurgeons for PACS interaction.
• To develop heuristic guidelines for designing gestural interfaces.
• To create a gesture lexicon optimized for surgeon preferences.

Main Methods:

• A gesture elicitation study involving nine neurosurgeons.
• Surgeons outlined gestures for PACS commands and performed imaging tasks using gestures.
• K-nearest neighbor approach used to derive a preferred gesture lexicon.

Main Results:

• Nine gesture lexicons, each with 28 gestures, were generated.
• A significant agreement (p < 0.001) was found on preferred gesture descriptors (89.5%) compared to overall gesture popularity (19.4%).
• Heuristics and a surgeon-preference-compliant lexicon were developed.

Conclusions:

• Neurosurgeons demonstrate consensus on fundamental gesture characteristics for image manipulation.
• Developed heuristics can guide future development of touchless gestural interfaces for PACS in the OR.