US 12,114,832 B2
Medical image processing device, endoscope system, medical image processing method, and program
Shumpei Kamon, Kanagawa (JP)
Assigned to FUJIFILM Corporation, Tokyo (JP)
Filed by FUJIFILM Corporation, Tokyo (JP)
Filed on Nov. 3, 2022, as Appl. No. 18/052,556.
Application 18/052,556 is a continuation of application No. 16/905,888, filed on Jun. 18, 2020, granted, now 11,526,986.
Application 16/905,888 is a continuation of application No. PCT/JP2018/045953, filed on Dec. 13, 2018.
Claims priority of application No. 2018-002005 (JP), filed on Jan. 10, 2018.
Prior Publication US 2023/0086972 A1, Mar. 23, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06K 9/00 (2022.01); A61B 1/00 (2006.01); A61B 1/05 (2006.01); A61B 1/06 (2006.01); A61B 8/00 (2006.01); G06T 7/00 (2017.01)
CPC A61B 1/000094 (2022.02) [A61B 1/000096 (2022.02); A61B 1/0005 (2013.01); A61B 1/05 (2013.01); A61B 1/0655 (2022.02); A61B 8/4245 (2013.01); A61B 8/463 (2013.01); A61B 8/469 (2013.01); G06T 7/0012 (2013.01); G06T 2207/10068 (2013.01); G06T 2207/10132 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30096 (2013.01)] 18 Claims
OG exemplary drawing
 
1. An endoscope system comprising:
an endoscope configured to be inserted into a body cavity of a subject; and one or more processors configured to:
acquire images at in-vivo positions of the subject by sequentially capturing the images through the endoscope;
acquire positional information indicating the in-vivo positions of the acquired images;
select a learned model that is a model learned using images corresponding to a position indicated by the acquired positional information among a plurality of learned models that are models learned using data sets including images at different in-vivo positions respectively and are configured to detect a region of interest;
cause the selected learned model to detect a region of interest in the acquired images;
and cause a display to display the position indicated by the acquired positional information, wherein-the learned models are models learned using data sets including images of mucous membranes at different in-vivo positions, respectively.