Method and system for statistical modeling of data using a quadratic likelihood functional
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G06T-005/00
G06T-007/00
G06F-017/18
출원번호
US-0490577
(2014-09-18)
등록번호
US-10068327
(2018-09-04)
발명자
/ 주소
Yahil, Amos
출원인 / 주소
Siemens Medical Solutions USA, Inc.
인용정보
피인용 횟수 :
0인용 특허 :
12
초록▼
A method and system are provided for constructing a model of a target object in a computer processor by receiving an input signal from a source, the input signal containing data describing the target object and a plurality of parameters, the input signal having a noise portion; selecting a group of
A method and system are provided for constructing a model of a target object in a computer processor by receiving an input signal from a source, the input signal containing data describing the target object and a plurality of parameters, the input signal having a noise portion; selecting a group of initial parameters, estimating a nonparametric probability distribution function (pdf) comprising a linear combination of a set of square-integrable basis functions, computing a quadratic likelihood functional (QLF), evaluating a fit of the initial parameters to the data based on the QLF, iteratively optimizing the QLF by selecting a new group of parameters and evaluating the fit of the new group of parameters into a predetermined condition is achieved. Once an acceptable fit is achieved, an output of a model of the target object constructed using optimized parameters can be displayed.
대표청구항▼
1. A computer-implemented method for removing noise from a signal of a target object, the computer-implemented method comprising: receiving an input signal from a source system, the input signal comprising image data identifying a target object and a plurality of parameters, and further comprising a
1. A computer-implemented method for removing noise from a signal of a target object, the computer-implemented method comprising: receiving an input signal from a source system, the input signal comprising image data identifying a target object and a plurality of parameters, and further comprising a noise portion, wherein the image data is selected from a group consisting of X-ray, CT, emission tomography, SPECT and PET, and the target object comprises image data of a body part of a patient;selecting an initial subset of parameters from the plurality of parameters;estimating a nonparametric probability distribution function (pdf) from the received input signal which comprises a linear combination of functions;generating a quadratic likelihood function (QLF) based on the nonparametric PDF estimated including the linear combination of functions;determining a fit of the initially selected subset of parameters to the data identifying the target object, based on the QLF;in response to the determined fit of the initially selected subset of parameters being below a predetermined threshold, iteratively optimizing parameters by selecting a new subset of parameters from the plurality of parameters and determining a fit of the new subset of parameters, until the determined fit satisfies the predetermined threshold; andgenerating an output signal comprising a reconstructed signal of the target object constructed using the new subset of iteratively optimized parameters. 2. The computer-implemented method of claim 1, wherein the data comprises weights wi and the QLF has the form L=∫dx[12f(x)2-f(x)∑iwiδ(x-xi)]. 3. The computer-implemented method of claim 1, further comprising calculating a source term using the data and basis functions. 4. The computer-implemented method of claim 3, wherein the QLF is obtained by: computing a Gram matrix using the basis functions; andcombining the Gram matrix, the parameters and the source tem′ to produce the QLF. 5. The computer-implemented method of claim 1, wherein the output comprises a 2D, 3D, or 4D image representation of the target object displayed at a graphical user interface. 6. The computer-implemented method of claim 1, wherein the image data is taken in at least two planes, and wherein the output comprises a 3D representation. 7. The computer-implemented method of claim 1, wherein the image data is taken in a least two planes and further comprises time, and wherein the output comprises a 4D representation. 8. A system for modeling of data describing a target object contained in an input signal, the system comprising: a computer-readable medium;a parameter optimization processor coupled to the computer-readable medium; anda communication interface coupled to the parameter optimization processor and adapted to receive and transmit electronic representations of reconstructed models signals to and from the parameter optimization processor, respectively, the computer-readable medium having stored thereon software instructions that, when executed by the parameter optimization processor, cause the parameter optimization processor to perform operations including: receive the input signal from a source system configured to collect object data, the input signal comprising image data identifying a target object and a plurality of parameters, and further comprises noise, wherein the image data is selected from a group consisting of X-ray, CT, emission tomography, SPECT and PET, and the target object comprises image data of a body part of a patient;select an initial subset of parameters corresponding to the target object from the plurality of parameters;estimate a nonparametric probability distribution function comprising (pdf) from the received input signal which comprises a linear combination of functions;generate a quadratic likelihood function (QLF) based on the nonparametric PDF including the linear combination of functions;determine a fit of the initially selected subset of parameters to the data identifying the target object, based on the QLF;in response to the determined fit of the initially selected subset of parameters being below a predetermined threshold, iteratively optimize parameters by selecting a new subset of parameters from the plurality of parameters and determining a fit of the new subset of parameters, until the determined fit satisfies the predetermined threshold; andgenerate an output signal comprising a signal of the target object constructed using the new subset of iteratively optimized parameters. 9. The system of claim 8, wherein the data comprises weights w; and the QLF has the form L=∫dx[12f(x)2-f(x)∑iwiδ(x-xi)]. 10. The system of claim 8, further comprising calculating a source term using the data and basis functions. 11. The system of claim 10, wherein the QLF is obtained by: computing a Gram matrix using the basis functions; andcombining the Gram matrix, the parameters and the source term to produce the QLF. 12. The system of claim 8, wherein the output signal comprises a 2D, 3D or 4D image representation of the target object displayed at a graphical user interface. 13. The system of claim 8, wherein the image data is taken in at least two planes, and wherein the output comprises a 3D representation. 14. The system of claim 8, wherein the image data is taken in a least two planes and further comprises time, and wherein the output comprises a 4D representation. 15. A method of generating a reconstructed image of a target object from an input signal having a data component and a noise component, the method comprising: receiving the input signal from an image source system, the input signal comprising image data identifying a target object and a plurality of parameters, and further comprising a noise portion, wherein the image data is selected from a group consisting of X-ray, CT, emission tomography, SPECT and PET, and the target object comprises image data of a body part of a patient;selecting an initial subset of parameters from the plurality of parameters;estimating a nonparametric probability distribution function (pdf) from the received input signal which comprises a linear combination of functions;generating a quadratic likelihood function (QLF) based on the nonparametric PDF including the linear combination of functions;determining a fit of the initially selected subset of parameters to the data identifying the target object, based on the QLF;in response to the determined fit of the initially selected subset of parameters being below a predetermined threshold, iteratively optimizing parameters by selecting a new subset of parameters from the plurality of parameters and determining a fit of the new subset of parameters, until the determined fit satisfies the predetermined threshold; andgenerating an output signal comprising a display of reconstructed image of the target object based on the new subset of iteratively optimized parameters. 16. The method of claim 15, wherein the input signal comprises first plane image data and second plane image data, and the output comprises displaying a three-dimensional image of the target object. 17. The method of claim 16, wherein the data comprises weights w; and the QLF has the form L=∫dx[12f(x)2-f(x)∑iwiδ(x-xi)]. 18. The method of claim 15, further comprising calculating a source term using the data and basis functions. 19. The method of claim 18, wherein the QLF is obtained by: computing a Gram matrix using the basis functions; andcombining the Gram matrix, the parameters and the source term to produce the QLF. 20. The computer-implemented method of claim 1, wherein the QDF comprises a form L=∫dx[12f(x,θ)2-f(x,θ)∑iδ(x-xi)] where θ represents the parameters, x represents the positions of the observations, and f (x, θ) is the pdf.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (12)
Puetter, Richard; Yahil, Amos; Pi?a, Robert, Accelerated signal encoding and reconstruction using pixon method.
Elbakri, Idris A.; Fessler, Jeffrey A., Method for statistically reconstructing a polyenergetic X-ray computed tomography image and image reconstructor apparatus utilizing the method.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.