Projection-domain material decomposition for spectral imaging

Patent metadata

Field	Value
Publication number	US-12567191-B2
Application number	US-202218276902-A
Country	US
Kind code	B2
Filing date	Feb 15, 2022
Priority date	Feb 19, 2021
Publication date	Mar 3, 2026
Grant date	Mar 3, 2026

Abstract

The present invention relates to a method ( 1 ), resp. a device, system and computer-program product, for material decomposition of spectral imaging projection data. The method comprises receiving ( 2 ) projection data acquired by a spectral imaging system and reducing ( 3 ) noise in the projection data by combining corresponding spectral values for different projection rays to obtain noise-reduced projection data. The method comprises applying ( 6 ) a first projection-domain material decomposition algorithm to the noise-reduced projection data to obtain a first set of material path length estimates, and applying ( 7 ) a second projection-domain material decomposition algorithm to the projection data to obtain a second set of material path length estimates. The second projection-domain material decomposition algorithm comprises an optimization that penalizes a deviation between the second set of material path length estimates being optimized and the first set of material path length estimates.

First claim

The invention claimed is:

1 . A computer-implemented method for material decomposition of spectral imaging projection data, comprising:
receiving projection data acquired by a spectral imaging system; reducing noise in the projection data by combining corresponding spectral values for different projection rays to obtain noise-reduced projection data; applying a first projection-domain material decomposition algorithm to the noise-reduced projection data to obtain a first set of material path length estimates; and applying a second projection-domain material decomposition algorithm to the projection data to obtain a second set of material path length estimates, wherein the second projection-domain material decomposition algorithm comprises an optimization that penalizes a deviation between the second set of material path length estimates being optimized and the first set of material path length estimates.

2 . The method of claim 1 , wherein the first projection-domain material decomposition algorithm comprises an optimization that minimizes a deviation between the noise-reduced projection data and a forward estimate thereof based on the first set of material path length estimates being optimized.

3 . The method of claim 2 , wherein the optimization of the first projection-domain material decomposition algorithm further comprises a Tikhonov regularization of the first set of material path length estimates being optimized.

4 . The method of claim 1 , wherein the optimization of the second projection-domain material decomposition algorithm minimizes a deviation between the projection data and a forward estimate thereof based on the second set of material path length estimates being optimized.

5 . The method of claim 1 , wherein said penalization in the optimization of the second projection-domain material decomposition algorithm comprises a modified Tikhonov regularization of the form Σ m β m ·|l im −{circumflex over (l)} im | n , where m iterates over the number of base-materials of the material decomposition, β m refers to predetermined weight parameters assigned to the respective base-materials, i identifies the projection ray for which the material decomposition is being determined by the optimization, l im refers to the material path length of the i'th projection ray and the m'th base-material in the second set of material path length estimates being optimized, {circumflex over (l)} im refers to the material path length of the i'th projection ray and the m'th base-material as determined by the first set of material path length estimates and n is a predetermined non-zero parameter.

6 . The method of claim 5 , wherein n is a real number or integer, and is greater than or equal to 1.

7 . The method of claim 5 , wherein a cost function of the optimization of the second projection-domain material decomposition algorithm has the form Σ b {circumflex over (p)} ib (l im )−p ib log({circumflex over (p)} ib (l im ))+Σ m β m ·|l im −{circumflex over (l)} im | n , in which pus refers to the projection data of the i'th projection ray and the b'th spectrum or energy bin and {circumflex over (p)} ib (l im ) to a corresponding forward estimate thereof based on the second set of material path length estimates being optimized.

8 . The method of claim 1 , wherein reducing the noise comprises low-pass spatial filtering and/or smoothing the projection data.

9 . The method of claim 8 , wherein low-pass spatial filtering and/or smoothing the projection data comprises applying a boxcar filter.

10 . The method of claim 1 , wherein reducing the noise comprises down-sampling the projection data.

11 . The method of claim 10 , comprising up-sampling and/or interpolating the first set of path length estimates to the same spatial resolution as the projection data for use in the second material decomposition algorithm.

12 . The method of claim 1 , further comprising reducing noise in the projection data by combining corresponding spectral values for different projection rays to obtain further noise-reduced projection data, and applying a further projection-domain material decomposition algorithm to the further noise-reduced projection data to obtain a further set of material path length estimates, wherein the first projection-domain material decomposition algorithm comprises an optimization that penalizes a deviation between the first set of material path length estimates being optimized and the further set of material path length estimates.

13 . A projection data processing device for material decomposition of spectral imaging projection data, comprising:
an input for receiving the projection data acquired by a spectral imaging system, a noise reducer for reducing noise in the projection data by combining corresponding spectral values for different projection rays to obtain noise-reduced projection data; a first optimizer to apply a first projection-domain material decomposition algorithm to the noise-reduced projection data and thus obtain a first set of material path length estimates; and a second optimizer to apply a second projection-domain material decomposition algorithm to the projection data to obtain a second set of material path length estimates, wherein the second projection-domain material decomposition algorithm comprises an optimization that penalizes a deviation between the second set of material path length estimates being optimized and the first set of material path length estimates.

14 . A non-transitory computer-readable medium for storing executable instructions, which cause a method to be performed for material decomposition of spectral imaging projection data, the method comprising:
receiving projection data acquired by a spectral imaging system; reducing noise in the projection data by combining corresponding spectral values for different projection rays to obtain noise-reduced projection data; applying a first projection-domain material decomposition algorithm to the noise-reduced projection data to obtain a first set of material path length estimates; and applying a second projection-domain material decomposition algorithm to the projection data to obtain a second set of material path length estimates, wherein the second projection-domain material decomposition algorithm comprises an optimization that penalizes a deviation between the second set of material path length estimates being optimized and the first set of material path length estimates.

Assignees

• Koninklijke Philips Nv

Inventors

• Brendel Bernhard Johannes
• Erhard Klaus Alfred
• Bontus Claas
• Sossin Artur
• Thran Axel
• Daerr Heiner

Classifications

• G06T2211/408

  Dual energy · CPC title

• G06T5/50

  using two or more images, e.g. averaging or subtraction · CPC title

• G06T5/70

  Denoising; Smoothing · CPC title

• G06T2211/424

  Iterative · CPC title

• G06T12/20 Primary

  Inverse problem, i.e. transformations from projection space into object space · CPC title