Method for pixel shift calculation in digital subtraction angiography and X-ray diagnostic imaging system for generating images in digital subtraction angiography

1. A method for generating images in digital subtraction angiography with an X-ray diagnostic imaging system, comprising the following steps:

accessing a mask image frame of a patient under examination;

accessing a series of live image frames of the patient under examination acquired in the same imaging position of the mask frame acquisition, wherein one of the mask image frame and the live image frames is contrast-enhanced;

assuming possible shift vectors in a region of interest and calculating a scoring thereof, each possible shift vector being a difference vector between the mask image frame and a respective live image frame,

determining the possible shift vector with the highest scoring and choosing it as an elected shift vector,

calculating a likelihood p representing a quality value of the elected shift vector,

calculating a modified shift vector depending on the elected shift vector and the likelihood p,

shifting the mask image frame with respect to the respective live image frame by the modified shift vector,

subtracting the shifted mask image frame from the respective live image frame, and

displaying the respective subtracted image frame.

2. The method of claim 1 , wherein the mask image frame is contrast-enhanced.

3. The method of claim 1 , wherein accessing the series of live image frames includes acquisition thereof in order to carry out live fluoroscopic roadmapping.

4. The method of claim 1 , wherein the likelihood p is calculated according to the following formula:

p=a/b,

wherein:

a is the difference between the highest scoring and an average of the surrounding thereof, and

b is the difference between the highest scoring and the lowest scoring of the possible shift vectors.

5. The method of claim 4 , wherein the average of the surrounding is an arithmetic average of the scoring of possible shift vectors within a distance of d pixels, d being preferably about 5.

6. The method of claim 4 , wherein the surrounding is one of a rectangle and a square around the elected shift vector with the highest scoring.

7. The method of claim 4 , wherein shifting of the mask image frame depends on whether the likelihood p is below a threshold value T 1 , which is preferably 0.35.

8. The method of claim 1 , wherein the modified shift vector is calculated by the formula:

( x′,y ′)=( x 0 ,y 0 )+ t (( x,y )−( x 0 ,y 0 ))∥( x,y )−( x 0 ,y 0 )|,

wherein:

(x′, y′) is the modified shift vector for the current live image frame,

(x 0 ,y 0 ) is the elected shift vector for the previous live image frame,

(x,y) is the current elected shift vector,

t is a parameter, 0≦t≦∥(x,y)−x 0 ,y 0 ∥, which is preferably 1.

9. The method of claim 8 , wherein a weighted mean filtering is applied to the modified shift vector according to the formula:

( x″,y ″)=(( x′,y ′)/ k )+(1−1 /k )( x 0 ,y 0 ),

wherein:

(x″, y″) is the filtered shift vector,

k is a parameter depending on the likelihood p.

10. The method of claim 1 , wherein the modified shift vector is calculated by a GGM (gliding geometrical mean) filtering applied to the elected shift vector according to the formula:

( x′″,y ′″)=(( x,y )/ k )+(1−1 /k )( x 0 ,y 0 ),

wherein:

(x′″, y′″) is the current modified shift vector,

(x,y) is the elected shift vector for the current live image frame,

(x 0 ,y 0 ) is the elected shift vector for the previous live image frame,

k is a parameter depending on the likelihood p.

11. The method of claim 9 or 10 , wherein k=3 in case the likelihood p is above a threshold value T 2 , which is preferably 0.35, k=6 in case the likelihood p is on or below the threshold value T 2 .

12. The method of claim 10 , wherein k is defined as:

k=T 2 /p+c,

wherein c is a constant, which is preferably about 2.

13. A non-transitory program storage device readable by machine, tangibly embodying a computer program product of instructions executable by the machine to perform method steps for generating images in digital subtraction angiography with an X-ray diagnostic imaging system, where the method allows for improving the image quality as set forth in claim 1 .

14. An X-ray diagnostic imaging system for generating images in digital subtraction angiography, in particular for conducting live fluoroscopic subtraction imaging, comprising:

an X-ray source for directing X-ray radiation to a patient being examined;

an X-ray imaging device positioned for receiving the X-ray radiation and acquiring images in response thereto;

a processor arranged in communication with the X-ray source and X-ray imaging device to control acquisition of a mask image frame and live image frames, wherein one of the mask image frame and the live image frames is contrast-enhanced, to conduct a pixel shift vector calculation operation based on a region of interest (ROI) to realize a pixel shift vector to correct for motion between live image frames, to shift pixels comprising the mask image frame by pixel shift directions defined by the pixel shift vector, and to subtract the shifted mask image frame from the live image frames to realize improved images, in particular live roadmapping image frames; and

a display for displaying the images,

wherein the processor is configured to:

assume possible shift vectors in the region of interest and calculating a scoring thereof, each possible shift vector being a difference vector between the mask image frame and a respective live image frame,

determine the possible shift vector with the highest scoring and choosing it as an elected shift vector,

calculate a likelihood p representing a quality value of the elected shift vector,

calculate a modified shift vector depending on the elected shift vector and the likelihood p,

shift the mask image frame with respect to the respective live image frame by the modified shift vector, and

subtract the shifted mask image frame from the respective live image frame.

15. The method of claim 1 , wherein a smaller portion of the modified shift vector is used for the respective live image frame for a low likelihood p, and wherein a higher portion of the modified shifty vector is used for the respective live image frame for a high likelihood p.