Adaptive NMR angiographic reprojection method

• US 5,204,627 A
• Filed: 03/14/1991
• Issued: 04/20/1993
• Est. Priority Date: 03/14/1991
• Status: Expired due to Fees

• Current Assignee: Wisconsin Alumni Research Foundation (University of Wisconsin System)
• Original Assignee: Wisconsin Alumni Research Foundation (University of Wisconsin System)
• Inventors: Mistretta, Charles A., Weber, David M., Korosec, Frank R.
• Primary Examiner(s): Tokar, Michael J.
• Application Number:
  US07/669,143
• Time in Patent Office: 768 Days
• Field of Search: 128/653.3, 324/300, 324/307, 324/309, 324/318, 324/322, 324/306
• US Class Current: 324/309
• CPC Class Codes: G01R 33/56 Image enhancement or correction, e.g. subtraction or averaging techniques , e.g. improvement of signal-to-noise ratio and resolution
• CPC Class Codes: G01R 33/563 of moving material, e.g. flow contrast angiography

Abstract

A magnetic resonance angiogram is produced by projecting a 3D array of motion sensitized NMR data. A mask which locates the vessels in the 3D array is produced by thresholding the NMR data, and this mask is combined with the 3D NMR data set to exclude signals produced by surrounding stationary tissues. An integration projection technique is used to produce the angiogram from the masked data set.

Claims

1. A method for producing a magnetic resonance angiogram, the steps comprising:

acquiring a 3D NMR data set comprised of a set of values which indicates by their magnitude the translational movement of spins and which indicates by their location in the 3D NMR data set the physical location of said spins;

producing a binary mask having a set of binary values in which there is a binary value corresponding to each value in the 3D NMR data set and the binary values are determined by a) locating regions in the 3D NMR data set in which the magnitude of its values exceed a first threshold value, b) enlarging those regions and c) setting the binary values in the binary mask which correspond to said enlarged regions to one of their binary values;

combining the 3D NMR data set with the binary mask to produce a 3D adapted data set in which values in the 3D NMR data set which correspond to binary values in the binary mask located outside the enlarged regions are removed from the 3D NMR data set; and

producing a 2D magnetic resonance angiogram from the 3D adapted data set in which a brightness value of each pixel in the 2D magnetic resonance angiogram is determined by the sum of all the data points therein that are disposed along the pixel'"'"'s projection ray.

2. The method as recited in claim 1 in which the regions are enlarged by convolving each data point located within said regions with a kernel.

3. The method as recited in claim 1 in which the binary mask is produced by:

locating the regions in successive 2D slices of the 3D NMR data set in which the magnitude of the 3D NMR data set values exceed the first threshold value;

enlarging those located regions in the plane of the 2D slice by convolving each 2D slice with a first kernel; and

enlarging those located regions in the direction perpendicular to the planes of the 2D slices by convolving with a second kernel.

4. The method as recited in claim 3 in which the first kernel has a center pixel and surrounding pixels, and the surrounding pixels decrease in value as a function of distance from the center pixel.

5. The method as recited in claim 1 in which the brightness of each pixel in the 2D magnetic resonance angiogram is produced by combining the brightness value (S) as determined by the sum of all the data points in the 3D adapted data set that are disposed along the pixel'"'"'s projection ray with a brightness value (MIP) as determined by the largest data point in the 3D NMR data set which is disposed along the pixel'"'"'s projection ray.

6. The method as recited in claim 5 in which the brightness value (S) is multiplied by a weighting factor (W) before being combined with the other brightness value (MIP).

7. The method as recited in claim 5 in which the brightness values (S) and (MIP) are combined to produce the magnetic resonance angiogram pixel brightness value (MRA) as follows:

space="preserve" listing-type="equation">MRA=(MIP+(W*S))/(W+1),

where W is a weighting factor.

8. The method as recited in claim 7 in which the weighting factor W is calculated as follows:

space="preserve" listing-type="equation">W=30 (S-S.sub.min).sup.2 /(S.sub.max -S.sub.min).sup.2,

where S_min is the smallest value in the 3D adapted data set and S_max is the largest value in the 3D adapted data set.

9. The method as recited in claim 1 in which the first threshold value is determined by producing a first intensity histogram from data in the 3D NMR data set which includes both background data and vessel data, producing a second intensity histogram from the data in the 3D NMR data set which includes background data, and setting the first threshold value to the lowest intensity level which has less than a predetermined number of occurrences in the first intensity histogram and is greater than a maximum intensity level which occurs in the second intensity histogram.