Gaze2Segment: A Pilot Study for Integrating Eye-Tracking Technology into Medical Image Segmentation

This is the authors' abstract. We don't add key points for this paper.

This study introduced a novel system, called Gaze2Segment, integrating and computer vision techniques to support radiologists' reading with an automatic image segmentation task. During diagnostic of lung CT scans, the radiologists' gaze information were used to a visual attention map. This map was then combined with a-derived saliency map, extracted from the gray-scale CT images. The attention map was used as an input for indicating roughly the location a object of interest. With computer-derived saliency information, on the hand, we aimed at finding foreground and background cues for the object interest. At the final step, these cues were used to initiate a seed-based process. Segmentation accuracy of the proposed Gaze2Segment was to be 86% with dice similarity coefficient and 1.45 mm with Hausdorff. To the best of our knowledge, Gaze2Segment is the first true of eye-tracking technology into a medical image segmentation task the need for any further user-interaction.