Method for image reconstruction using sparsity-constrained correction
View Patent ↗An image reconstruction method applicable to a number of different imaging modalities including magnetic resonance imaging (MRI), x-ray computed tomography (CT), positron emission tomography (PET), and single photon emission computed tomography (SPECT) is disclosed. A sparsifying image is reconstructed from a series of acquired undersampled data to provide a priori knowledge of a subject being imaged. An iterative reconstruction process is further employed to iteratively determine a correction image for a given image frame that, when subtracted from the sparsifying image, produces a quality image for the image frame.
1. A method for producing an image of a subject positioned in a field of view (FOV) of a medical imaging system, the steps comprising:
a) acquiring, with a motion monitoring system, a signal indicative of motion in the subject;
b) acquiring, with the medical imaging system, a plurality of image data sets of the subject positioned in the FOV at a corresponding plurality of time frames;
c) reconstructing a sparsifying image by combining data from the plurality of acquired image data sets;
d) producing a sparsifying image data set by transforming the sparsifying image reconstructed in step c);
e) producing a difference image data set by subtracting the image data set acquired in a selected time frame from the sparsifying image data set produced in step d), wherein the selected time frame is selected using the signal indicative of motion in the subject acquired in step a);
f) reconstructing a correction image using the difference image data set produced in step e); and
g) subtracting the correction image from the sparsifying image reconstructed in step c) to produce an image of the subject at the selected time frame.
2. The method as recited in claim 1 in which step d) includes regridding the sparsifying image to a sampling pattern of the image data set acquired at a selected time frame.
3. The method as recited in claim 1 in which the medical imaging system is an x-ray computed tomography (CT) imaging system and the image data sets acquired in step a) are interleaved x-ray projection space data sets.
4. The method as recited in claim 1 in which the plurality of image data sets acquired in step a) include undersampled image data sets.
5. The method as recited in claim 1 in which the medical imaging system is an image-guided radiation therapy (IGRT) system and the image data sets acquired in step b) are x-ray projection space data sets.
6. The method as recited in claim 5 in which:
the motion monitoring systems includes a respiration monitoring device;
the acquired signal indicative of motion in the subject is a signal indicative of respiratory motion in the subject; and further including:
h) reconstructing an image of the subject using a PICCS image reconstruction method.
7. The method as recited in claim 6 in which step e) includes producing a respiratory phase image data set from the plurality of image data sets acquired in step a) using the acquired signal indicative of respiratory motion.
8. The method as recited in claim 7 in which the correction image produced in step f) is indicative of a respiratory phase and step h) includes using the image reconstructed in step g) as a prior image.
9. The method as recited in claim 7 further including:
i) updating a radiation delivery plan using the image reconstructed in step h).