Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
METHOD AND SYSTEM FOR PRODUCING LEARNING IMAGE DATA
Document Type and Number:
WIPO Patent Application WO/2024/009740
Kind Code:
A1
Abstract:
A method and a system for producing learning image data are provided, whereby it becomes possible to produce learning image data for use in the production of a learning model without being affected by the distortion of a peripheral part of an endoscopic image and the difference in color between a mucosa image and a mask image. A region which is adjacent to the mask image and in which the mucosa image is distorted due to the influence of a lens of an endoscope for observing a mucosa is subjected to a mask elongation processing for producing an elongated region image of the mask image, thereby producing an elongated learning image. The mask image of the elongated learning image is subjected to a color conversion correction procedure to produce mask color-replaced learning image data, then the color hue of the mucosa image is corrected without affecting the learning of the mask color-replaced learning image to produce mucosa color hue-corrected learning image data, and the mucosa color hue-corrected learning image data is used as learning image data.

Inventors:
NOSATO HIROKAZU
SAKANASHI HIDENORI
IWAKI TAKUYA
Application Number:
PCT/JP2023/022550
Publication Date:
January 11, 2024
Filing Date:
June 19, 2023
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
AIST (JP)
International Classes:
A61B1/045; G06T1/00
Domestic Patent References:
WO2022114088A12022-06-02
Foreign References:
JPH06254044A1994-09-13
Other References:
IWAKI, TAKUYA: "Development of an endoscopic diagnosis assistance system for Hunner-type interstitial cystitis using artificial intelligence", HAINYO SHOGAI PURAKUTISU, MEDIKARU REBYUSHA, OSAKA, JP, vol. 30, no. 1, 31 May 2022 (2022-05-31), JP , pages 72 - 75, XP009552132, ISSN: 0919-5750
Attorney, Agent or Firm:
NISHIURA & ASSOCIATES (JP)
Download PDF: