US 12,169,948 B2
Processor for endoscope, endoscope system, information processing apparatus, non-transitory computer-readable storage medium, and information processing method
Kenta Kosugi, Tokyo (JP)
Assigned to HOYA CORPORATION, Tokyo (JP)
Appl. No. 17/634,394
Filed by HOYA CORPORATION, Tokyo (JP)
PCT Filed Aug. 16, 2019, PCT No. PCT/JP2019/032131
§ 371(c)(1), (2) Date Feb. 10, 2022,
PCT Pub. No. WO2021/033215, PCT Pub. Date Feb. 25, 2021.
Prior Publication US 2022/0327738 A1, Oct. 13, 2022
Int. Cl. G06T 7/80 (2017.01); A61B 1/00 (2006.01); G06T 7/90 (2017.01); G06V 10/70 (2022.01); H04N 23/50 (2023.01); H04N 23/74 (2023.01)
CPC G06T 7/80 (2017.01) [A61B 1/000096 (2022.02); G06T 7/90 (2017.01); G06V 10/70 (2022.01); A61B 1/000094 (2022.02); A61B 1/000095 (2022.02); G06T 2207/10024 (2013.01); G06T 2207/10068 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30004 (2013.01); G06V 2201/03 (2022.01); H04N 23/555 (2023.01); H04N 23/74 (2023.01)] 10 Claims
OG exemplary drawing
 
1. A processor for an endoscope for determining whether current endoscope settings are the same as a user's favorite endoscope settings, and if not, changing the current endoscope settings to the user's favorite endoscope settings, comprising:
a controller executing program code to perform:
acquiring, by the controller, an endoscopic image of a part of a subject captured using current endoscope settings including first system information setting at least two of
an intensity of a color of light produced by a light source of the endoscope to image the part of the subject,
a brightness of the light source, and
an enhancement mode for enhancing the endoscopic image;
identifying an organ of the subject contained in the acquired endoscopic image with the current endoscope settings using a first learning model that outputs a discrimination result identifying the organ of the subject in a case in which the acquired endoscopic image is input;
acquiring, by the controller, a setting image of the identified organ of the subject stored in advance in a database before the endoscopic image is acquired,
produced by the endoscope using the user's favorite endoscope settings, and
associated with the discrimination result output by the first learning model;
outputting second system information
containing the user's favorite endoscope settings for the identified organ, setting at least two of the intensity of the color of the light produced by the light source, the brightness of the light source, and the enhancement mode, and
produced using a second learning model that outputs the second system information in a case in which the acquired setting image and the part of the subject are input;
comparing the current endoscope settings of the current first system information with the favorite endoscope settings of the second system information;
determining if the difference between the current endoscope settings and the favorite endoscope settings is equal to or greater than a threshold;
changing the endoscope settings to be the favorite endoscope settings when the comparing operation determines that the difference between the current endoscope settings and the favorite endoscope settings is equal to or greater than the threshold;
determining whether the change of the endoscope settings has succeeded;
outputting a failure notification to a display to display the failure notification when the change of endoscope settings has failed;
displaying the failure notification on the display when the failure notification is output to the display;
outputting a success notification to the display to display the success notification when the change of endoscope settings has succeeded; and
displaying the success notification on the display when the success notification is output to the display.