I took a course that introduced me to the creation and evaluation of physical and multimodal human-computer interfaces. CPSC 543 focused on the control and/or display of virtual environments through the sense of touch for the purpose of human-system communication, as well as introducing perceptual/attentional foundations.

My term project focused on the design and evaluation of user input devices for Radiologists’ PACS workstations. In essence, we were asking the question “The mouse/keyboard you’re using isn’t optimized for the task you’re performing. What would work better?”

Some of my contributions to this project are outlined on a website I maintained for the course. One of the people I collaborated with on this project would take the research further and turned it into a conference abstract, to be presented at the 2013 Society for Imaging Informatics in Medicine conference. You can find Louise’s abstract titled “Enhancing Medical Image Interaction by Specializing the Mouse” listed in the final program.

The switch from hard-copy film to soft-copy digital formats in Radiology has resulted in a ballooning number of images to be viewed for diagnosis. Despite the rapid innovation in imaging technology, Radiologists are still using relatively unchanged user-input devices (UIDs) and it has been suggested that they’re a limiting factor in workflow efficiency.

Our project dissected the interactions of radiologists with diagnostic scans into a distinct array of tasks and designed a UID around this importance/frequency of these tasks.

collab1

Our final prototype, shown left during a field testing session with radiologists at UBC Hospital, used a touch interface from an iPod combined with a thumbstick and a custom shaped rocker. This allowed the user to have the majority of their most commonly used interactions mapped to hardware on the UID, rather than buried in drop-down menus on the PACS interface.

You can read more detail in our final report here: CPSC 543 Final Report