Electronic endoscope system and data processing device

Abstract

An electronic endoscope system comprises an evaluation unit that evaluates the extent of a lesion by processing an image of the inside of an organ imaged by an electronic endoscope. The evaluation unit comprises: an image evaluation value calculation unit that calculates an image evaluation value indicating the extent of a bio-tissue lesion for the captured image; a lesion evaluation unit that uses information of an imaging position inside the organ to calculate a representative evaluation value of image evaluation values from the image evaluation value for each of a plurality of spaces obtained by partitioning the area inside the organ, and that uses the representative evaluation value to evaluate the spread of the lesion in the depth direction; and a lesion evaluation adjustment unit that sets, for an empty space for which a representative evaluation value was not generated, as an estimated representative evaluation value, one image evaluation value for a non-empty space which is further to the back of the organ than the empty space, or a representative evaluation value obtained on the basis of the image evaluation value, and that assigns the estimated representative evaluation value to the empty space.

Description

technical field [0001] The present invention relates to an electronic endoscope system and a data processing device for evaluating the spread of lesions in living tissue inside an organ. Background technique [0002] Lesions in living tissue have varying degrees of severity ranging from thinning, roughened, reddened inflammation of the mucosal layer of living tissue to ulceration with localized defects in the mucosal layer and its underlying layers. For example, in ulcerative colitis (UC: Ulcerative Colitis) lesions, the ulcer site is whitish with white fur and pus-like mucus, while in the inflamed site, it is accompanied by edema and redness that is easy to congest. This lesion can be photographed and observed through an endoscope system. [0003] However, in order for surgeons to recognize normal parts and lesion parts through color differences contained in endoscopic images, long-term training under the guidance of skilled persons is required. In addition, it is not eas...

AI Technical Summary

Problems solved by technology

In addition, it is not easy for even skilled surgeons to identify the lesion from subtle differences in color, so careful work is required

Images