An image processing device includes an image acquisition section that acquires an image that has been acquired by imaging a tissue using an endoscope apparatus, an in vivo position identification information acquisition section that acquires in vivo position identification information that specifies an in vivo position of the endoscope apparatus when the image has been acquired, a in vivo model acquisition section that acquires a in vivo model that is a model of the tissue, an on-model position determination section that specifies an on-model position that corresponds to the position specified by the in vivo position identification information on the acquired in vivo model, and a linking section that links information about the acquired image to the specified on-model position.
