Supervisors info:
Σεραφείμ Νανάς, Καθηγητής, Ιατρική σχολή, ΕΚΠΑ
Ιωάννης Βασιλειάδης, Επίκουρος Καθηγητής, Ιατρική σχολή, ΕΚΠΑ
Νικολέττα Ροβίνα, Καθηγήτρια, Ιατρική σχολή, ΕΚΠΑ
Summary:
In this study, we evaluate three alternative methods of human-computer interaction, considering objective measurements from performance experiments, in which two target groups participated: a) athletes and b) patients with stroke during rehabilitation. We utilised IDEA software, which performs multiparametric performance evaluations when using any point-and-click device that simulates the functions of a common mouse, for experiments and measurements. Using the specific software that actually records and analyses the trajectory of the mouse pointer, we examined the following three alternative interfaces of selecting a target on a computer screen: 1) mouse and left button(click) for target selection, 2) gesture and 3) Tracker Pro camera, that detects head movements. The last two methods of interaction are performed using the Click Gestures Assistant (CGA) selection software. Our goal is to compare the performance of the three different human- computer interfaces in both athletes and stroke patients.
The experiments were executed in one-dimensional and two-dimensional graphical environments of increasing difficulty. The usability of the three interfaces will be measured based on the Greek version of the Systems Usability Scale (SUS) questionnaire. The site of the experiments is the physiotherapist clinic of the student Petrou Dimos-David in the city of Istiaia under the guidance of Dr. Alexandros Pino from the Department of Informatics and Telecommunications of the National and Kapodistrian University of Athens. In the experiments participated thirty healthy athletes and eight patients with stroke, after being evaluated on the Mini-Mental State scale. The collected data were first processed by the IDEA software, that exports the main measurement indicators and then analysed statistically.
According to the results of the research, the group of athletes and the group of patients showed higher Throughput and lower Movement Time in conventional mouse, then in gesture and finally in head movements. In addition, both groups showed lower Missed Clicks and Target Re-entries in gesture compared to head movements.
In all key assessment scores, the athletes performed better than the patients on the mouse and gesture, which are the two interfaces that required upper extremity dexterity. In contrast, both groups performed similarly in the interface of head movements. Therefore, head control after a stroke does not appear to be affected, unlike the upper extremities. Finally, both groups, based on the System Usability Scale questionnaire, rated the common mouse with a high score, while both the interfaces of gesture and head movements are judged to be moderately useful.
This study, although low in participants, is one of the first to compare gestures with head movements in people who have had a stroke and will become a reference point for future studies.
Keywords:
Point and click, Hand gesture, Tracker Pro, SUS, Stroke
