Methods and systems for selective recombination of reconstructed image regions
Various methods and systems are provided for selective recombination of reconstructed nuclear medicine (NM) image regions. The method may include acquiring and sorting imaging scan data that includes data elements having different accuracy properties into two or more different groups using an anticipated accuracy metric, reconstructing a first initial image volume from a first group having a highest anticipated accuracy metric of the two or more different groups, reconstructing a second initial image volume from a combined data group comprising data of the first group and a second group of the two or more different groups, wherein the second group has a lower anticipated accuracy metric compared to the first group, and generating an optimized image volume using corresponding image values from the first initial image volume, the second initial image volume, and an output of a high spatial frequencies detection (HSFD) operator applied to the first initial image volume.
1 . An imaging system, comprising:
a detector array including a plurality of detectors;
a display device; and
a computing device communicatively coupled to the detector array and the display device, and configured with instructions in non-transitory memory that when executed cause the computing device to:
acquire imaging scan data including data elements with different accuracy properties;
sort the imaging scan data into two or more different groups based on an anticipated accuracy metric;
reconstruct a high accuracy image from a first group, the first group comprising data elements with a highest anticipated accuracy metric;
calculate image-based kernels for a guided hybrid-reconstruction algorithm;
select a second group from the two or more different groups, the second group having a next highest accuracy data of a set of image data groups, with respect to the first group;
reconstruct an updated high accuracy image from the high accuracy image, the image-based kernels, and the second group;
update the image-based kernels for the guided hybrid-reconstruction algorithm, based on the updated high accuracy image; and
output the updated high accuracy image for display on the display device.
2 . The imaging system of claim 1 , wherein the imaging system is configured to capture the imaging scan data having two or more distinct ranges of data accuracy, and the computing device is configured to continuously update reconstruction of the updated high accuracy image from each subsequent accuracy group and updated kernel calculation.
3 . The imaging system of claim 1 , wherein the imaging system is a large axial field of view (FOV) positron emission tomography (PET) system.
4 . The imaging system of claim 1 , wherein the imaging system is a single photon emission computed tomography (SPECT) imaging system configured to perform cardiac scans and sort imaging data of a 180-degree acquisition into the first group comprising data elements with the highest anticipated accuracy metric and sort imaging data of a 360-degree acquisition into the second group having a next highest accuracy data of the set of image data groups, with respect to the first group.
5 . The imaging system of claim 1 , wherein the imaging scan data are sorted for different accuracy levels by separating detected coincidence photons by their energy, where photons within a narrow energy window around a peak will be considered as most accurate, and photons with energy in a lower energy window, further from the peak, will be considered as less accurate.