Method and system for extracting lower limb vasculature

Abstract

The present disclosure relates to systems and methods for extracting a vessel of a lower limb. The methods may include obtaining an original image including a plurality of image data, in some embodiments, each of the plurality of image data may correspond to a pixel (or a voxel), the plurality of image data may include a target data set, the target data set may represent a first structure; extracting a first reference data set from the plurality of image data, in some embodiments, the first reference data set may include the target data set and a second reference data set, the second reference data set may include data of a second structure; extracting the second reference data set from the plurality of image data; and obtaining the target data set based on the first reference data set and the second reference data set.

We claim: 
     
         1 . A method implemented on a computing device having at least one processor and at least one non-transitory storage medium, the method comprising:
 obtaining an original image including a plurality of image data points, the plurality of image data points including a target data set, the target data set representing a first structure;   extracting a first reference data set from the plurality of image data points, the first reference data set including the target data set;   obtaining a first reference mask corresponding to the first structure based on the first reference data set;   determining a first subtraction image corresponding to a second structure based on the first reference data set and the first reference mask;   determining a second reference mask based on the first subtraction image and the original image;   determining a second subtraction image by subtracting the second reference mask from a first image, wherein elements in the first image and data points in the first reference data set are bijective; and   obtaining the target data set based on the second subtraction image.   
     
     
         2 . The method of  claim 1 , wherein the first structure includes at least a portion of a vessel, the second structure includes at least a portion of a skeleton, the target data set includes vessel data, the first reference data set includes vessel data and skeleton data, the second reference mask includes at least a portion of a skeleton, and the second subtraction image includes broken vessel segments. 
     
     
         3 . The method of  claim 2 , wherein the obtaining the target data set based on the second subtraction image includes:
 performing at least one data supplement operation on a frame data set to obtain the target data set, wherein elements in the second subtraction image and data points in the frame data set are bijective.   
     
     
         4 . The method of  claim 3 , wherein the extracting a first reference data set from the plurality of image data includes:
 determining at least one connected domain CD1 in the original image;   determining a first seed point based on the at least one connected domain CD1; and   performing a regional growth on the original image based on the first seed point and a first threshold to obtain the first image.   
     
     
         5 . The method of  claim 4 , wherein the determining a first seed point based on the at least one connected domain CD1 includes:
 determining values of a boundary distance field of the at least one connected domain CD1 to obtain a data set pfield-1;   determining a circularity degree of the at least one connected domain CD1 based on the data set pfield-1;   determining a target connected domain based on the at least one connected domain CD1; and   determining the first seed point based on the target connected domain.   
     
     
         6 . The method of  claim 5 , wherein the determining a circularity degree of the at least one connected domain CD1 based on the data set pfield-1 includes:
 determining a radius of the at least one connected domain CD1 based on the data set pfield-1;   determining a circular area of the at least one connected domain CD1 based on the radius of the at least one connected domain CD1; and   determining the circularity degree of the at least one connected domain CD1 based on the circular area of the at least one connected domain CD1 and an actual area of the at least one connected domain CD1.   
     
     
         7 . The method of  claim 6 , wherein the obtaining a first reference mask corresponding to the first structure includes:
 determining a boundary distance field based on the first image to obtain a data set pfield-2; and   segmenting the first structure based on the data set pfield-2 to obtain the first reference mask.   
     
     
         8 . The method of  claim 7 , wherein the determining the second reference mask based on the first subtraction image and the original image includes:
 segmenting the second structure based on the first subtraction image to obtain a first skeleton mask; and   determining the second reference mask based on the original image and the first skeleton mask.   
     
     
         9 . The method of  claim 5 , wherein the obtaining a first reference mask includes:
 performing a regional growth on the first image based on the first seed point to obtain a first vessel mask; and   dilating the first vessel mask to obtain the first reference mask.   
     
     
         10 . The method of  claim 8 , wherein the segmenting the second structure based on the first subtraction image to obtain a first skeleton mask includes:
 calculating a boundary distance field based on the first subtraction image to obtain a data set pfield-3;   determining a skeleton seed point based on the data set pfield-3; and   performing a regional growth on the first subtraction image based on the skeleton seed point to obtain the first skeleton mask.   
     
     
         11 . The method of  claim 8 , wherein the determining the second reference mask based on the original image and the first skeleton mask includes:
 determining a first skeleton region based on the first skeleton mask;   dilating the first skeleton region to obtain a first temporary skeleton mask; and   determining the second reference mask based on the first temporary skeleton mask.   
     
     
         12 . The method of  claim 11 , wherein the determining the second reference mask based on the first temporary skeleton mask includes:
 performing a regional growth on the original image based on a second threshold to obtain a second image;   filling the second image to obtain a filled second image;   obtaining a superimposition image based on the first temporary skeleton mask and the filled second image; and   performing a closing operation on at least one connected domain CD2 in the superimposition image to obtain the second skeleton mask.   
     
     
         13 . The method of  claim 11 , wherein the determining a first skeleton region includes:
 eroding the first skeleton mask to obtain at least one connected domain CD3; and   determining the first skeleton region based on the at least one connected domain CD3.   
     
     
         14 . The method of  claim 13 , wherein the determining the first skeleton region based on the at least one connected domain CD3 includes:
 designating a connected domain with the maximum area or the maximum volume in the at least one connected domain CD3 as the first skeleton area.   
     
     
         15 . The method of  claim 3 , wherein the performing at least one data supplement operation on the frame data set includes:
 selecting a second seed point from the frame data set; and   performing a regional growth based on a second threshold and the third seed point to obtain the target data set.   
     
     
         16 . The method of  claim 3 , wherein the performing at least one data supplement operation on the frame data set includes:
 extracting a center line of a vessel based on the second subtraction image; and   generating the vessel based on the center line of the vessel   
     
     
         17 . The method of  claim 16 , wherein the extracting a center line of a vessel includes:
 calculating a boundary distance field based on the second subtraction image;   obtaining a first vessel growing point and a second vessel growing point based on the boundary distance field; and   extracting the center line of the vessel using a shortest route algorithm based on the first vessel growing point and the second vessel growing point.   
     
     
         18 . The method of  claim 17 , wherein the extracting the center line of the vessel further includes:
 determining a second skeleton region in the second skeleton mask; and   extracting the center line of the vessel by excluding the second skeleton region   
     
     
         19 . A non-transitory computer readable medium including executable instructions that, when executed by at least one processor, cause the at least one processor to:
 obtaining an original image including a plurality of image data, each of the plurality of image data corresponding to an element that is a pixel or a voxel, the plurality of image data including a target data set, the target data set representing a first structure;   extracting a first reference data set from the plurality of image data, the first reference data set including the target data set;   obtaining a first reference mask corresponding to the first structure based on the first reference data set;   determining a first subtraction image corresponding to a second structure by subtracting based on the first reference data set and the first reference mask;   determining a second reference mask based on the first subtraction image and the original image;   determining a second subtraction image by subtracting the second reference mask from a first image, wherein pixels or voxels in the first image and data in the first reference data set are bijective; and   obtaining the target data set based on the second subtraction image.   
     
     
         20 . A system comprising:
 at least one processor; and   at least one storage device storing executable instructions that, when executed by the at least one processor, cause the at least one processor to effectuate a method comprising:
 obtaining an original image including a plurality of image data points, each of the plurality of image data points corresponding to an element that is a pixel or a voxel, the plurality of image data points including a target data set, the target data set representing a first structure; 
 extracting a first reference data set from the plurality of image data points, the first reference data set including the target data set; 
 obtaining a first reference mask corresponding to the first structure based on the first reference data set; 
 determining a first subtraction image corresponding to a second structure based on the first reference data set and the first reference mask; 
 determining a second reference mask based on the first subtraction image and the original image; 
 determining a second subtraction image by subtracting the second reference mask from a first image, wherein elements in the first image and data points in the first reference data set are bijective; and 
 obtaining the target data set based on the second subtraction image.