Method of performing measurements on digital images
AssigneeAgfa Healthcare NV
InventorPiet Dewaele
Life of this patent
Estimating the term…
Abstract
A first and a second image are expressed in a common coordinate system by applying a geometric transformation to the second image so as to map a structure in the second image onto a corresponding structure in the first image in a common coordinate system. Starting from initial values, the parameters of the geometric transformation are updated taking into account the result of an evaluation of a cost function. Measurements are performed in the common coordinate system.
Claims18
A method of performing a geometric measurement in a digital image comprising the steps of:
- (1) obtaining a digital signal representation of an image A,
- (2) obtaining a digital signal representation of a reference image B having the same dimension as image A,
- (3) applying at least said digital signal representation of image A to a display device coupled to a computer and displaying image A,
- (4) applying a geometric transform to image B so as to map at least a geometric structure in image B onto a corresponding geometric structure in image A in a common coordinate system,
- (5) displaying transformed image B, and
- (6) performing a geometric measurement in said common coordinate system using corresponding pixels in image A and transformed image B, the geometric measurement comprising one of a measurement of distance and a measurement of angle.
A method of determining the extent to which the position of a structure in an image A is rotated relative to a reference position in a reference image comprising the steps of:
- (1) obtaining a digital signal representation of image A,
- (2) obtaining a digital reference image B pertaining to the same structure as in the image A and having the same dimension as image A,
- (3) applying said digital signal representation of image A to a display device coupled to a computer and displaying image A,
- (4) selecting said structure in displayed image A,
- (5) computing parameters of a geometric transform comprising at least a rotational component, said geometric transformation mapping said structure from said reference image B onto a corresponding structure in image A in common coordinate system, and
- (6) aligning said images by applying said geometric transform with the computed parameters to image B, wherein the rotational component of said transform determines the extent to which the position of said structure is rotated.
A method of identifying a vertebra in a spine image comprising the steps of:
- (1) obtaining a digital signal representation of a spine image A,
- (2) obtaining a digital signal representation of a reference image B,
- (3) applying said digital signal representation of image A to a display device coupled to a computer and displaying image A,
- (4) selecting a vertebra in displayed image A,
- (5) selecting a corresponding vertebra in the reference image B,
- (6) applying a geometric transform to image B so as to map at least said vertebra in image B onto a corresponding vertebra in image A in a common coordinate system, and
- (7) identifying the selected vertebra in spine image A by assigning a label identifying the vertebra in image B to a corresponding vertebra in image A.
Dependencies are inferred from the claim text.