US 12,333,718 B2
Method for generating model by recognizing cross-section regions in units of pixels
Yuki Sakaguchi, Isehara (JP); Yusuke Seki, Tokyo (JP); and Akira Iguchi, Mishima (JP)
Assigned to TERUMO KABUSHIKI KAISHA, Tokyo (JP)
Filed by TERUMO KABUSHIKI KAISHA, Tokyo (JP)
Filed on Sep. 29, 2022, as Appl. No. 17/955,810.
Application 17/955,810 is a continuation of application No. PCT/JP2021/009343, filed on Mar. 9, 2021.
Claims priority of application No. 2020-061513 (JP), filed on Mar. 30, 2020.
Prior Publication US 2023/0020596 A1, Jan. 19, 2023
Int. Cl. G06T 7/00 (2017.01); G06T 7/62 (2017.01)
CPC G06T 7/0012 (2013.01) [G06T 7/62 (2017.01); G06T 2207/10101 (2013.01); G06T 2207/10132 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30101 (2013.01)] 13 Claims
OG exemplary drawing
 
1. A non-transitory computer-readable storage medium storing a computer program which, when executed by a computer of a diagnostic imaging apparatus, causes the computer to perform processing of:
acquiring a plurality of medical images generated based on signals detected by a catheter inserted into a lumen organ while the catheter is moving a sensor along a longitudinal direction of the lumen organ, the lumen organ including a main trunk, a side branch branched from the main trunk, and a bifurcated portion of the main trunk and the side branch; and
recognizing, in units of pixels in the plurality of medical images, a main trunk cross-section region, a side branch cross-section region, and a bifurcated portion cross-section region by inputting the acquired medical images into a learning model configured to recognize the main trunk cross-section region, the side branch cross-section region, and the bifurcated portion cross-section region in units of pixels in the plurality of medical images.
 
9. An information processing device comprising:
an acquisition unit configured to acquire a plurality of medical images generated based on signals detected by a catheter inserted into a lumen organ while the catheter is moving a sensor along a longitudinal direction of the lumen organ, the lumen organ including a main trunk, a side branch branched from the main trunk, and a bifurcated portion of the main trunk and the side branch; and
a learning model configured to recognize, when the acquired medical images are input, a main trunk cross-section region, a side branch cross-section region, and a bifurcated portion cross-section region in units of pixels in the plurality of medical images, and output information indicating the main trunk cross-section region, the side branch cross-section region, and the bifurcated portion cross-section region in units of pixels in the plurality of medical images.
 
13. A method for generating a model, the method comprising performing the following processing in a computer:
generating training data in which data indicating a lumen cross-section is attached to a plurality of medical images each including a main trunk cross-section region, a plurality of medical images each including the main trunk cross-section region and a side branch cross-section region, and a plurality of medical images each including a bifurcated portion cross-section region that are generated based on signals detected by a catheter inserted into a lumen organ while the catheter is moving a sensor along a longitudinal direction of the lumen organ, the lumen organ including a main trunk, a side branch branched from the main trunk, and a bifurcated portion of the main trunk and the side branch; and
generating, based on the generated training data, a learning model configured to recognize, in units of pixels in the plurality of medical images, the main trunk cross-section region, the side branch cross-section region, and the bifurcated portion cross-section region when the medical images are input.