Method and device for automatically processing blood vessel image
A method, performed by a processor, for processing a blood vessel image from an angiography image may comprise the steps of: extracting a target blood vessel from a blood vessel image; identifying an error portion from the extraction result of the target blood vessel on the basis of at least one of blood vessel structure data related to the target blood vessel, curvature information of the target blood vessel, diameter information of the target blood vessel, and brightness information of the target blood vessel; and in response to a case where an error portion is identified in the target blood vessel, correcting the identified error portion.
1. A method for processing a blood vessel image performed by a processor comprising the steps of:
extracting a target blood vessel from a blood vessel image;
identifying an error portion of the blood vessel image from the extraction result of the target blood vessel on the basis of brightness information of the target blood vessel, wherein the brightness information of the target blood vessel includes an indication of a concentration of a contrast agent that has been injected into a bloodstream flowing within the target blood vessel; and
correcting the identified error portion of the blood vessel image in response to a case where the error portion of the blood vessel image is identified from the extraction result of the target blood vessel,
wherein the identifying the error portion of the blood vessel image comprises identifying a lower branch as a misidentification portion of the blood vessel image in response to a case where a brightness difference between an upper branch and the lower branch adjacent to each other is a threshold brightness difference or more in the extraction result of the target blood vessel,
wherein the correcting the identified error portion of the blood vessel image comprises replacing the lower branch with another candidate lower branch having a brightness difference from the upper branch of less than the threshold brightness difference, and
wherein the threshold brightness difference is determined based on a type of the target blood vessel.
2. The method for processing the blood vessel image of claim 1 , wherein the identifying of the error portion of the blood vessel image is also on the basis of blood vessel structure data related to the target blood vessel and further comprises
segmenting a blood vessel area and a non-blood vessel area from the blood vessel image;
generating the blood vessel structure data on the basis of a node corresponding to each of blood vessel branches divided based on a branch point in the blood vessel area;
comparing connection data between nodes of the target blood vessel with the blood vessel structure data; and
determining nodes corresponding to a connection error as the error portion of the blood vessel image when the connection error between the nodes is detected on the basis of a topology of the blood vessel structure data.
3. The method for processing the blood vessel image of claim 2 , wherein the correcting of the identified error portion of the blood vessel image comprises
connecting regions corresponding to a discontinuity portion in response to a case where the error portion of the blood vessel image is identified as the discontinuity portion; and
replacing a node corresponding to a misidentification portion with a node matching the blood vessel structure data in response to a case where the error portion of the blood vessel image is identified as the misidentification portion.
4. The method for processing the blood vessel image of claim 1 , wherein:
the identifying of the error portion of the blood vessel image is also on the basis of blood vessel structure data related to the target blood vessel and further comprises
identifying a lower branch as a misidentification portion in response to a case where a curvature between an upper branch and the lower branch adjacent to each other exceeds a threshold curvature in the extraction result of the target blood vessel; and
the correcting of the identified error portion of the blood vessel image comprises
replacing the lower branch with another candidate lower branch having a curvature with the upper branch of the threshold curvature or less with respect to the identified misidentification portion.
5. The method for processing the blood vessel image of claim 1 , wherein
the correcting of the identified error portion of the blood vessel image comprises
generating a new extraction result for the target blood vessel in response to a case where the error portion of the blood vessel image is identified.
6. The method for processing the blood vessel image of claim 1 , wherein
the correcting of the identified error portion of the blood vessel image comprises
providing a user with one or more candidate branches in response to a case where the error portion of the blood vessel image is the misidentification portion; and
replacing a branch corresponding to the error portion of the blood vessel image with the selected branch in response to a case of receiving a selection input for one branch among the one or more candidate branches from the user.
7. The method for processing the blood vessel image of claim 1 , wherein
the correcting of the identified error portion of the blood vessel image comprises
connecting a region corresponding to a blood vessel branch corresponding to a start point of a user input and a region corresponding to a blood vessel branch corresponding to an end point of the user input, in response to a case where the user input is detected in a plurality of points.
8. A method for processing a blood vessel image performed by a processor comprising the steps of:
extracting a target blood vessel from a blood vessel image;
identifying an error portion of the blood vessel image from the extraction result of the target blood vessel on the basis of diameter information of the target blood vessel, and brightness information of the target blood vessel, wherein the diameter information includes identifying a diameter difference between an upper branch and a lower branch adjacent to each other in the blood vessel image is a threshold or more in the extraction result of the target blood vessel and the brightness information of the target blood vessel includes an indication of a concentration of a contrast agent that has been injected into a bloodstream flowing within the target blood vessel; and
correcting the identified error portion of the blood vessel image in response to a case where the error portion of the blood vessel image is identified from the extraction result of the target blood vessel,
wherein the identifying the error portion of the blood vessel image comprises identifying a lower branch as a misidentification portion of the blood vessel image in response to a case where a brightness difference between an upper branch and the lower branch adjacent to each other is a threshold brightness difference or more in the extraction result of the target blood vessel
wherein the correcting the identified error portion of the blood vessel image comprises replacing the lower branch with another candidate lower branch having a brightness difference from the upper branch of less than the threshold brightness difference, and
wherein the threshold brightness difference is determined based on a type of the target blood vessel.
9. The method for processing the blood vessel image of claim 8 , wherein
the correcting of the identified error portion of the blood vessel image comprises
correcting the identified error portion of the blood vessel image by performing at least one of replacing the lower branch on the basis of the diameter difference from the upper branch with respect to the identified error portion of the blood vessel image and generating a new extraction result including new diameter information about the target blood vessel.
10. A device for processing a blood vessel image comprising:
an image receiver for receiving a blood vessel image; and
a processor for extracting a target blood vessel from the blood vessel image, identifying an error portion of the blood vessel image from the extraction result of the target blood vessel on the basis of brightness information of the target blood vessel, wherein the brightness information of the target blood vessel includes an indication of a concentration of a contrast agent that has been injected into a bloodstream flowing within the target blood vessel, and correcting the identified error portion of the blood vessel image in response to a case where the error portion of the blood vessel image is identified from the extraction result of the target blood vessel,
wherein the processor identifies a lower branch as a misidentification portion of the blood vessel image in response to a case where a brightness difference between an upper branch and the lower branch adjacent to each other is a threshold brightness difference or more in the extraction result of the target blood vessel, and replaces the lower branch with another candidate lower branch having a brightness difference from the upper branch of less than the threshold brightness difference,
wherein the threshold brightness difference is determined based on a type of the target blood vessel.