Using movement parallax for 3D laparoscopy.

Abstract

The lack of depth perception hampers the surgeon during laparoscopic operation. Laparoscopes usually are monocular, but binocular ones are currently available. Depth perception, however, does not exclusively rely on binocular disparity. An observer, with only one eye, who is able to move that eye, obtains the same information as an observer who has two eyes. This principle of movement parallax can be applied to laparoscopy by coupling the head movements of the surgeon to the motions of the tip of the laparoscope. In an experiment we investigated if this principle is applicable to laparoscopy. Two groups of testees with no background in surgery were used. The first group was assisted by movement parallax, the second group was viewing a static image. Both groups of testees had to perform an exploration and a manipulation task. Since the amount of space for camera motion within the laparoscope is limited, implementation potential depends on the amount of movements that will be made by the observer. Therefore the movements of the observer performing the exploration task were registered and analysed. Results of the experiment indicate the advantage of movement parallax for the exploration task (performance increases by factor 2 while using only 30% more time) but not for the manipulation task. The analysis of the movements indicates that small movements are sufficient for implementation. Based on these results we concluded that movement parallax is applicable to laparoscopy.