Patent US8121362B2 successfully refreshed from Google Patents.
US8121362B2
Active / In Force
US
⏱️ 1.5 years remaining
Registration of medical images using learned-based matching functions
Assignee: Siemens Medical Solutions USA Inc
•
Filing: 2008-04-28
•
Granted: 2012-02-21
•
Expires: 2028-04-28
Abstract
Abstract A method for registering a medical image includes acquiring a first medical image of a subject. One or more simulated medical images are synthesized based on the acquired first medical image. One or more matching functions are trained using the first medical image and the simulated medical images. A second medical image of the subject is acquired. The first medical image and the second medical image are registered using the one or more trained matching functions.
Inventors
- 👤 Yiqiang Zhan
- 👤 Xiang Zhou
- 👤 Arun Krishnan
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-09 13:53
CPC Classifications
G; G06; G06T; G06T7/00; G06T7/30; G06T7/33; G06V; G06V10/00; G06V10/20; G06V10/24; G06V2201/00; G06V2201/03
Claims Analysis
Total Claims: 23
Claim 1
Independent Claim
88 words
A method for registering a medical image, comprising: acquiring a first medical image of a subject; placing one or more anchor points in the first medical image; synthesizing one or more simulated medical images by transforming the acquired first medical image about the anchor points; training one or more matching functions using the first medical image and the simulated medical images; acquiring a second medical image of the subject; and registering the first medical image and the second medical image using the one or more trained matching functions.
Claim 2
Independent Claim
88 words
A method for registering a medical image, comprising: acquiring a first medical image of a subject; placing one or more anchor points in the first medical image; synthesizing one or more simulated medical images by transforming the acquired first medical image about the anchor points; training one or more matching functions using the first medical image and the simulated medical images; acquiring a second medical image of the subject; and registering the first medical image and the second medical image using the one or more trained matching functions.
Claim 3
Dependent Claim
18 words
The method of claim 1 , wherein the first medical image is a CT, MRI, or PET image.
Claim 4
Dependent Claim
60 words
The method of claim 1 , wherein transforming the acquired first medical image includes applying one or more spatial transformations to the first medical image to generate one or more of the one or more simulated medical images, wherein the one or more spatial transformations comprise a rigid transformation, a similarity transformation, a spline transformation or a free form deformation.
Claim 5
Dependent Claim
36 words
The method of claim 3 , wherein a transformation function is used to transform the first medical image and the transformation function includes a randomly generated value to randomize each of the resulting simulated medical images.
Claim 6
Dependent Claim
26 words
The method of claim 3 , wherein the first medical image is transformed with a non-rigid deformation to synthesize the one or more simulated medical images.
Claim 7
Dependent Claim
20 words
The method of claim 1 , wherein the quantity, diversity and complexity of the simulated medical images are user selectable.
Claim 8
Dependent Claim
20 words
The method of claim 1 , wherein the synthesis of the one or more simulated medical images is performed offline.
Claim 9
Dependent Claim
46 words
The method of claim 1 , wherein transforming the acquired first medical image includes applying an intensity transformation to the first medical image to generate one or more simulated medical images, wherein the intensity transformation comprises a linear transformation, a polynomial transformation or a kernel-based transformation.
Claim 10
Dependent Claim
94 words
The method of claim 1 , wherein the one or more matching functions are trained using positive and negative training data, wherein positive training data includes a portion of the first medical image and a matching portion of one of the simulated medical images and negative training data includes a portion of the first medical image and a non-matching portion of one of the simulated medical images, wherein the positive training data is cropped such that a location of at least one of the one or more anchor points is consistent in the portions.
Claim 11
Dependent Claim
46 words
The method of claim 1 , additionally comprising performing the following steps prior to training the one or more matching functions: locating the one or more anchor points within the acquired first medical image; and locating corresponding anchor points within each of the simulated medical images.
Claim 12
Dependent Claim
22 words
The method of claim 10 , wherein the anchor points are automatically detected within the first medical image using learning-based landmark detection.
Claim 13
Dependent Claim
21 words
The method of claim 10 , wherein each of the one or more matching functions corresponds to a particular anchor point.
Claim 14
Dependent Claim
28 words
The method of claim 1 , wherein the one or more trained matching functions are applied hierarchically in registering the first medical image and the second medical image.
Claim 15
Dependent Claim
113 words
The method of claim 13 , wherein registering the first medical image and the second medical image includes matching a first subset of points with a high matching confidence to find corresponding portions of the first and second medical images and later, to the extent necessary, matching a second subset of points with a lower matching confidence to find corresponding portions of the first and second medical images, wherein a relationship between corresponding portions of the first and second medical images found using the first subset of the points are used to facilitate the use of the second subset of the points to find corresponding portions of the first and second medical images.
Claim 16
Independent Claim
154 words
A method for registering a medical image, comprising: acquiring a medical image of a first subject; placing one or more anchor points within the acquired medical image of the first subject; synthesizing one or more simulated medical images by transforming the acquired medical image of the first subject about the anchor points; locating corresponding anchor points within each of the simulated medical images; training one or more matching functions, each corresponding to one of the anchor points, using the medical image of the first subject and the simulated medical images; acquiring a medical image of a second subject; registering the medical image of the first subject and the medical image of the second subject using the trained matching functions; and determining correspondences of points that are not anchor points by interpolating a general relationship between the medical images of the first and second subjects based on the trained matching functions for the anchor points.
Claim 17
Independent Claim
154 words
A method for registering a medical image, comprising: acquiring a medical image of a first subject; placing one or more anchor points within the acquired medical image of the first subject; synthesizing one or more simulated medical images by transforming the acquired medical image of the first subject about the anchor points; locating corresponding anchor points within each of the simulated medical images; training one or more matching functions, each corresponding to one of the anchor points, using the medical image of the first subject and the simulated medical images; acquiring a medical image of a second subject; registering the medical image of the first subject and the medical image of the second subject using the trained matching functions; and determining correspondences of points that are not anchor points by interpolating a general relationship between the medical images of the first and second subjects based on the trained matching functions for the anchor points.
Claim 18
Dependent Claim
47 words
The method of claim 15 , wherein synthesizing the one or more simulated medical images based on the acquired medical image of the first subject includes transforming the medical image of the first subject to generate one or more of the one or more simulated medical images.
Claim 19
Dependent Claim
39 words
The method of claim 16 , wherein a transformation function is used to transform the medical image of the first subject and the deformation function includes a randomly generated value to randomize each of the resulting simulated medical images.
Claim 20
Dependent Claim
34 words
The method of claim 15 , wherein the one or more trained matching functions are applied hierarchically in registering the medical image of the first subject and the medical image of the second subject.
Claim 21
Dependent Claim
134 words
A computer system comprising: a processor; and a non-transitory program storage device readable by the computer system, embodying a program of instructions executable by the processor to perform method steps for registering a medical image, the method comprising: acquiring a first medical image of a subject; placing one or more anchor points within the acquired medical image of the first subject; applying one or both of a spatial transformation and an intensity transformation about the anchor points in the acquired first medical image to generate one or more simulated medical images; training one or more matching functions using the first medical image and the simulated medical images; acquiring a second medical image of the subject; and registering the first medical image and the second medical image using the one or more trained matching functions.
Claim 22
Dependent Claim
134 words
A computer system comprising: a processor; and a non-transitory program storage device readable by the computer system, embodying a program of instructions executable by the processor to perform method steps for registering a medical image, the method comprising: acquiring a first medical image of a subject; placing one or more anchor points within the acquired medical image of the first subject; applying one or both of a spatial transformation and an intensity transformation about the anchor points in the acquired first medical image to generate one or more simulated medical images; training one or more matching functions using the first medical image and the simulated medical images; acquiring a second medical image of the subject; and registering the first medical image and the second medical image using the one or more trained matching functions.
Claim 23
Dependent Claim
21 words
The computer system of claim 19 , wherein the quantity, diversity and complexity of the simulated medical images are user selectable.
References Cited (Backward Citations)
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2015-02-20 | 2015-08-22 | 2016-02-21 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2019-02-20 | 2019-08-22 | 2020-02-21 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2023-02-20 | 2023-08-22 | 2024-02-21 | $7700.00 | Window Closed / Lapsed |