Medical image processing
AssigneeThe General Hospital Corporation
InventorsWilliam C. Karl, Zhuangli Liang, Homer Pien, Thomas J. Brady
Life of this patent
Estimating the term…
Abstract
In one aspect, the invention is a method to reduce a blooming effect of a bright object in a medical image generated from a low-dose imaging system. The method includes characterizing an ideal image as a composition of a first component and a second component. The first component is characterized by a first function and the second component is characterized by a second function. The method also includes minimizing the first function and the second function using a smoothing process and generating the medical image based on the minimizing.
Claims18
An apparatus comprising:
- a memory that stores executable instructions to reduce a blooming effect of a bright object in a medical image generated from a low-dose imaging system;
- and
- a processor that executes the instructions to:
- characterize an ideal image as a composition of a first component and a second component, the first component being characterized by a first function and the second component being characterized by a second function;
- minimize the first function and the second function using a smoothing process;
- and
- generate the medical image based on the instructions to minimize;
- wherein the instructions to minimize the first function and the second function using the smoothing process comprises instructions to implement a smoothing process in accordance with an equation:
- min
- f
- ′
-
- (
- α
-
-
- g
- ′
- -
- f
- ′
-
- 2
- +
- β
-
-
- ∇
- f
- ′
-
- 2
- )
- ,
- where f′ represents at least a portion of the ideal image, g′ represents an image from the low-dose imaging system and α and β determine a relative weighting of the fidelity of f′ to the image, g′.
The apparatus of claim l wherein f′ represents an image of the second component.
An apparatus comprising:
- a memory that stores executable instructions to reduce a blooming effect of a bright object in a medical image generated from a low-dose imaging system;
- and
- a processor that executes the instructions to:
- characterize an ideal image as a composition of a first component and a second component, the first component being characterized by a first function and the second component being characterized by a second function;
- minimize the first function and the second function using a smoothing process;
- and
- generate the medical image based on the instructions to minimize;
- wherein the instructions to minimize the first function and the second function using the smoothing process comprises instructions to implement a smoothing process in accordance with an equation:
- min
- f
- '
-
- (
- α
-
-
- g
- '
- -
- f
- '
-
- 2
- +
- β
-
-
- ∇
- f
- '
-
- 2
-
- (
- 1
- -
- v
- )
- 2
- +
- ρ
- 2
-
-
- ∇
- v
-
- +
- v
- 2
- 2
-
- ρ
- )
- ,
- where f′ represents at least a portion of the ideal image, g′ represents an image from the low-dose imaging system and α and β determine a relative weighting of the fidelity of f′ to the image, g′, u represents a probability of a presence of an edge and ρ represents a thickness of the edge.
An article comprising a machine-readable medium that stores executable instructions to reduce a blooming effect of a bright object in a medical image generated from a low-dose imaging system, the instructions causing a machine to:
- characterize an ideal image as a composition of a first component and a second component, the first component being characterized by a first function and the second component being characterized by a second function;
- minimize the first function and the second function using a smoothing process;
- and
- generate the medical image based on the instructions causing a machine to minimize;
- wherein the instructions causing the machine to minimize the first function and the second function using the smoothing process comprises instructions causing the machine to use an equation:
- min
- f
- '
-
- (
- α
-
-
- g
- '
- -
- f
- '
-
- 2
- +
- β
-
-
- ∇
- f
- '
-
- 2
-
- (
- 1
- -
- v
- )
- 2
- +
- ρ
- 2
-
-
- ∇
- v
-
- +
- v
- 2
- 2
-
- ρ
- )
- ,
- where f′ represents at least a portion of the ideal image, g′ represents an image from the low-dose imaging system and α and β determine a relative weighting of the fidelity of f′ to the image, g′, u represents a probability of a presence of an edge and ρ represents a thickness of the edge.
An article comprising a machine-readable medium that stores executable instructions to reduce a blooming effect of a bright object in a medical image generated from a low-dose imaging system, the instructions causing a machine to:
- characterize an ideal image as a composition of a first component and a second component, the first component being characterized by a first function and the second component being characterized by a second function;
- minimize the first function and the second function using a smoothing process;
- and generate the medical image based on the instructions causing a machine to minimize;
- wherein the instructions causing the machine to minimize the first function and the second function using the smoothing process comprises instructions causing the machine to use an equation:
- min
- f
- ′
-
- (
- α
-
-
- g
- ′
- -
- f
- ′
-
- 2
- +
- β
-
-
- ∇
- f
- ′
-
- 2
- )
- ,
- where f′ represents at least a portion of the ideal image, g′ represents an image from the low-dose imaging system and α and β determine a relative weighting of the fidelity of f′ to the image, g′.
A method for producing a CT image comprising:
- a) acquiring CT data of a region-of-interest (ROI) in a subject;
- b) reconstructing the CT data into a set of component images, each having a known characteristic, by iteratively minimizing a model that relates the acquired CT data to estimates of each of the component images and includes constraints related to each of the known characteristics;
- c) performing a selected noise suppression operation on each of the component images to produce a corresponding set of smooth component images;
- and
- d) combining the smooth component images to produce a smooth CT image of the subject.
Dependencies are inferred from the claim text.