Program, image processing method, and image processing device
A non-transitory computer-readable medium storing a computer program executed by a computer, a method, and an image processing device are disclosed that are capable of compensating a missing region in a tomographic image in a state in which a part of a lumen organ is missing. In accordance with the program, a computer acquires a plurality of tomographic images of a cross section of the lumen organ captured at a plurality of places using a catheter. In addition, the computer extracts, from the plurality of tomographic images, a tomographic image in which the part of the lumen organ is missing. Then, the computer compensates a missing region of the lumen organ for the extracted tomographic image.
1 . A non-transitory computer-readable medium storing a program executed by a computer that executes a process comprising:
acquiring a plurality of tomographic images of a cross section of a lumen organ captured at a plurality of places using a catheter, each tomographic image having an imaging range with a center corresponding to a position of the catheter;
extracting, from the plurality of tomographic images, a tomographic image in which a part of the lumen organ extends beyond the imaging range and is missing from the tomographic image; and
compensating a missing region of the lumen organ that extends beyond the imaging range for the extracted tomographic image.
2 . The computer-readable medium according to claim 1 , further comprising:
inputting the plurality of acquired tomographic images into a learning model trained to output information indicating regions of a lumen and a lumen wall of a lumen organ in a tomographic image when the tomographic image is input, and outputting a lumen and a lumen wall of a lumen organ in each of the tomographic images; and
extracting the tomographic image in which the part of the lumen organ is missing when a part of a contour line of the output lumen or lumen wall of the lumen organ matches a contour line of the tomographic image.
3 . The computer-readable medium according to claim 1 , further comprising:
in the extracted tomographic image in which the part of the lumen organ extends beyond the imaging range and is missing, compensating a contour line of the missing region of the lumen organ based on a contour line of the lumen organ which is not missing.
4 . The computer-readable medium according to claim 1 , further comprising:
compensating the missing region of the lumen organ in the extracted tomographic image based on a tomographic image in which the lumen organ is not missing in tomographic images captured in a vicinity of the extracted tomographic image.
5 . The computer-readable medium according to claim 1 , further comprising:
inputting the extracted tomographic image into a second learning model trained to output the missing region of the lumen organ in the extracted tomographic image when the extracted tomographic image in which the part of the lumen organ extends beyond the imaging range and is missing is input, and outputting the missing region of the lumen organ in the extracted tomographic image; and
wherein compensating the missing region of the lumen organ in the extracted tomographic image is based on the output missing region of the lumen organ.
6 . The computer-readable medium according to claim 1 , further comprising:
adding, to the extracted tomographic image in which the part of the lumen organ extends beyond the imaging range and is missing, a contour line of the compensated missing region of the lumen organ in a mode of indicating a compensated place.
7 . The computer-readable medium according to claim 2 , wherein the lumen organ is a blood vessel, the lumen is an intravascular lumen, and the lumen wall is a vessel wall.
8 . An image processing method of executing processes by a computer, the image processing method comprising:
acquiring a plurality of tomographic images of a cross section of a lumen organ captured at a plurality of places using a catheter, each tomographic image having an imaging range with a center corresponding to a position of the catheter;
extracting, from the plurality of tomographic images, a tomographic image in which a part of the lumen organ extends beyond the imaging range and is missing from the tomographic image; and
compensating a missing region of the lumen organ that extends beyond the imaging range for the extracted tomographic image.
9 . The method according to claim 8 , further comprising:
inputting the plurality of acquired tomographic images into a learning model trained to output information indicating regions of a lumen and a lumen wall of a lumen organ in a tomographic image when the tomographic image is input, and outputting a lumen and a lumen wall of a lumen organ in each of the tomographic images; and
extracting the tomographic image in which the part of the lumen organ is missing when a part of a contour line of the output lumen or lumen wall of the lumen organ matches a contour line of the tomographic image.
10 . The method according to claim 8 , further comprising:
in the extracted tomographic image in which the part of the lumen organ extends beyond the imaging range and is missing, compensating a contour line of the missing region of the lumen organ based on a contour line of the lumen organ which is not missing.
11 . The method according to claim 8 , further comprising:
compensating the missing region of the lumen organ in the extracted tomographic image based on a tomographic image in which the lumen organ is not missing in tomographic images captured in a vicinity of the extracted tomographic image.
12 . The method according to claim 8 , further comprising:
inputting the extracted missing tomographic image into a second learning model trained to output the missing region of the lumen organ in the extracted tomographic image when the tomographic image in which the part of the lumen organ extends beyond the imaging range and is missing is input and outputting the missing region of the lumen organ in the extracted tomographic image; and
wherein compensating the missing region of the lumen organ in the extracted tomographic image is based on the output missing region of the lumen organ.
13 . The method according to claim 8 , further comprising:
adding, to the extracted tomographic image in which the part of the lumen organ extends beyond the imaging range and is missing, a contour line of the compensated missing region of the lumen organ in a mode of indicating a compensated place.
14 . The method according to claim 9 , wherein the lumen organ is a blood vessel, the lumen is an intravascular lumen, and the lumen wall is a vessel wall.
15 . An image processing device comprising:
a processor configured to:
acquire a plurality of tomographic images of a cross section of a lumen organ captured at a plurality of places using a catheter, each tomographic image having an imaging range with a center corresponding to a position of the catheter;
extract, from the plurality of tomographic images, a tomographic image in which a part of the lumen organ extends beyond the imaging range and is missing from the tomographic image; and
compensate a missing region of the lumen organ that extends beyond the imaging range for the extracted tomographic image.
16 . The image processing device according to claim 15 , wherein the processor is further configured to:
input the plurality of acquired tomographic images into a learning model trained to output information indicating regions of a lumen and a lumen wall of a lumen organ in a tomographic image when the tomographic image is input, and outputting a lumen and a lumen wall of a lumen organ in each of the tomographic images; and
extract the tomographic image in which the part of the lumen organ is missing when a part of a contour line of the output lumen or lumen wall of the lumen organ matches a contour line of the tomographic image.
17 . The image processing device according to claim 15 , wherein the processor is further configured to:
in the extracted tomographic image in which the part of the lumen organ extends beyond the imaging range and is missing, compensate a contour line of the missing region of the lumen organ based on a contour line of the lumen organ which is not missing.
18 . The image processing device according to claim 15 , wherein the processor is further configured to:
compensate the missing region of the lumen organ in the extracted tomographic image based on a tomographic image in which the lumen organ is not missing in tomographic images captured in a vicinity of the extracted tomographic image.
19 . The image processing device according to claim 15 , wherein the processor is further configured to:
input the extracted missing tomographic image into a second learning model trained to output the missing region of the lumen organ in the extracted tomographic image when the tomographic image in which the part of the lumen organ extends beyond the imaging range and is missing is input, and outputting the missing region of the lumen organ in the extracted tomographic image; and
wherein compensate the missing region of the lumen organ in the extracted tomographic image is based on the output missing region of the lumen organ.
20 . The image processing device according to claim 15 , wherein the processor is further configured to:
add, to the extracted tomographic image in which the part of the lumen organ extends beyond the imaging range and is missing, a contour line of the compensated missing region of the lumen organ in a mode of indicating a compensated place.