A research team developed a software tool to speed up the process of processing eye-tracking data for the analysis of prosthetic hand function. The program used a standardized set up, including analysis of activity of daily living tasks, and a computer identified the key areas of interest and tracks them in the visual field with minimal human intervention. The system proved to be sufficiently reliable for much of the process to be automated, the study concluded, and the testing found that the prosthesis users did less looking ahead to the next phase of the task.
Gaze-tracking, where a subject’s point of regard is mapped onto the image of the scene the subject sees, can be employed to study the visual attention of the users of prosthetic hands. It can show whether the user pays greater attention to the actions of their prosthetic hand as they use it to perform manipulation tasks, compared with the general population.
Conventional analysis of the video data requires a human operator to identify the key areas of interest in every frame of the video. Computer vision techniques can assist with this process, but fully automatic systems require large training sets, and prosthetic investigations tend to be limited in numbers, according to the study’s authors, but if the assessment task is well-controlled, the much simpler system is effective.
The tool employed color separation and edge detection on images of the visual field to identify the objects to be tracked. To simplify the computer’s task further, the test used the Southampton Hand Assessment Procedure cutting task to define the activity spatially and temporarily, reducing the search space for the computer.
Gaze was successfully tracked for 14 unimpaired subjects and was compared with the gaze of four users of myoelectric hands. In a test of 100 gaze video sequences, the semiautomated tracker only lost the various targets when the visor device that held the pupil and scene cameras lost track of the gaze. There were 18 users with six tasks, and eight sequences where the fixation data was missing.
The open-access study, “A tool to assist in the analysis of gaze patterns in upper-limb prosthetic use,” was published in the journal Prosthesis.
