Method and device for extracting major vessel region on basis of vessel image
A method for extracting a major vessel region from a vessel image by a processor may comprise the steps of: extracting an entire vessel region from a vessel image; extracting a major vessel region from the vessel image on the basis of a machine learning model which extracts a major vessel region; and revising the major vessel region by connecting separated vessel portions on the basis of the entire vessel region.
1. A method for extracting a major vessel region from a blood vessel image performed by a processor, the method comprising:
extracting an entire vessel region from the blood vessel image;
extracting a major vessel region from the blood vessel image based on a machine learning model which extracts a major vessel region;
detecting the separated portion of the blood vessel in the major vessel region; and
correcting the major vessel region by connecting a separated portion of the blood vessel based on the entire vessel region,
wherein the detecting of the separated portion of the blood vessel includes:
determining an area between blobs in which a shortest distance between blobs corresponding to the major vessel region is smaller than a threshold distance as the separated portion of the blood vessel.
2. The major vessel region extracting method of claim 1 , wherein the machine learning model is a machine learning model which is trained for predetermined shapes of a major blood vessel.
3. The major vessel region extracting method of claim 2 , wherein the machine learning model is a machine learning model trained for a shape of at least one blood vessel of a right coronary artery (RCA), a left anterior descending artery (LAD), and a left circumflex artery (LCX).
4. The major vessel region extracting method of claim 1 , wherein the correcting of the major vessel region includes:
generating a connecting line which connects an area between blobs in which a shortest distance between blobs corresponding to the major vessel region is smaller than a threshold distance; and
correcting the major vessel region by connecting the separated portion of the blood vessel in the major vessel region based on an area corresponding to the connecting line in the entire vessel region.
5. The major vessel region extracting method of claim 4 , wherein the correcting of the major vessel region includes:
correcting the major vessel region based on an area having a shortest distance which connects the separated portion of the blood vessel, among a plurality of areas, when there is the plurality of areas which is capable of connecting the separated portion of the blood vessel corresponding to the connecting line in the entire vessel region.
6. The major vessel region extracting method of claim 1 , further comprising:
converting an RGB value of the blood vessel image into a grayscale level; and
normalizing a blood vessel image which is converted into the grayscale level.
7. The major vessel region extracting method of claim 1 , wherein the extracting of the entire vessel region includes:
extracting the entire vessel region based on a partial blood vessel image generated from an entire blood vessel image, and
the extracting of the major vessel region includes:
extracting the major vessel region based on a partial blood vessel image generated from the entire blood vessel image.
8. The major vessel region extracting method of claim 1 , wherein the correcting of the major vessel region includes:
determining the separated portion of the blood vessel in response to a user's input to designate a position of the major vessel region; and
correcting a major vessel region of a target position based on a surrounding major vessel region adjacent to the separated portion of the blood vessel.
9. A non-transitory computer-readable storage medium storing instructions that, when executed by one or more processors, configure the one or more processors to perform the method of claim 1 .
10. A major vessel region extracting apparatus, comprising:
a processor configured to extract an entire vessel region from a blood vessel image, extract a major vessel region from the blood vessel image based on a machine learning model which extracts a major vessel region, detect the separated portion of the blood vessel in the major vessel region, and correct the major vessel region by connecting a separated portion of the blood vessel based on the entire vessel region; and
a memory configured to store at least one of the blood vessel image, the entire vessel region, the major vessel region, and the machine learning model,
wherein the processor determines an area between blobs in which a shortest distance between blobs corresponding to the major vessel region is smaller than a threshold distance as the separated portion of the blood vessel.
11. The major vessel region extracting apparatus of claim 10 , wherein the machine learning model is a machine learning model which is trained for predetermined shapes of a major blood vessel.
12. The major vessel region extracting apparatus of claim 11 , wherein the machine learning model is a machine learning model trained for a shape of at least one blood vessel of a right coronary artery (RCA), a left anterior descending artery (LAD), and a left circumflex artery (LCX).
13. The major vessel region extracting apparatus of claim 10 , wherein the processor generates a connecting line which connects an area between blobs in which a shortest distance between blobs corresponding to the major vessel region is smaller than a threshold distance and corrects the major vessel region by connecting the separated portion of the blood vessel in the major vessel region based on an area corresponding to the connecting line in the entire vessel region.
14. The major vessel region extracting apparatus of claim 13 , wherein when there is a plurality of areas which is capable of connecting the separated portion of the blood vessel corresponding to the connecting line in the entire vessel region, the processor corrects the major vessel region based on an area having a shortest distance which connects the separated portion of the blood vessel, among the plurality of areas.
15. The major vessel region extracting apparatus of claim 10 , wherein the processor converts an RGB value of the blood vessel image into a grayscale level and normalizes a blood vessel image which is converted into the grayscale level.
16. The major vessel region extracting apparatus of claim 10 , wherein the processor extracts the entire vessel region and the major vessel region based on a partial blood vessel image generated from the entire blood vessel image.
17. The major vessel region extracting apparatus of claim 10 , wherein the processor determines the separated portion of the blood vessel in response to a user's input to designate a position of the major vessel region and corrects a major vessel region of a target position based on a surrounding major vessel region adjacent to the separated portion of the blood vessel.