US 11,864,939 B2
Apparatus and method that uses deep learning to correct computed tomography (CT) with sinogram completion of projection data
Jian Zhou, Buffalo Grove, IL (US); Ruoqiao Zhang, Arlington Heights, IL (US); Zhou Yu, Wilmette, IL (US); and Yan Liu, Vernon Hills, IL (US)
Assigned to CANON MEDICAL SYSTEMS CORPORATION, Otawara (JP)
Filed by CANON MEDICAL SYSTEMS CORPORATION, Otawara (JP)
Filed on Jun. 4, 2021, as Appl. No. 17/339,093.
Application 17/339,093 is a continuation of application No. 16/227,251, filed on Dec. 20, 2018, granted, now 11,039,806.
Prior Publication US 2021/0290193 A1, Sep. 23, 2021
This patent is subject to a terminal disclaimer.
Int. Cl. A61B 6/00 (2006.01); G06T 11/00 (2006.01); G06T 5/50 (2006.01); G06T 5/20 (2006.01); G06T 5/10 (2006.01); G06N 3/084 (2023.01); G06N 3/045 (2023.01)
CPC A61B 6/5258 (2013.01) [A61B 6/4014 (2013.01); G06N 3/045 (2023.01); G06N 3/084 (2013.01); G06T 5/10 (2013.01); G06T 5/20 (2013.01); G06T 5/50 (2013.01); G06T 11/005 (2013.01); G06T 11/006 (2013.01); G06T 11/008 (2013.01); A61B 6/482 (2013.01); A61B 6/5205 (2013.01); G06T 2207/10081 (2013.01); G06T 2207/20064 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2211/408 (2013.01)] 20 Claims
OG exemplary drawing
 
11. A data processing method for multi-energy CT imaging, comprising:
obtaining a set of projection data including interleaved low-kV and high-kV sparse projection data acquired in an X-ray CT scan of an object to be imaged, wherein during the CT scan, voltage applied to an X-ray source used in the CT scan alternates between a first voltage and a second voltage higher than the first voltage, the first voltage corresponding to the low-kV sparse projection data, the second voltage corresponding to the high-kV sparse projection data; and
obtaining a generated complete set of low-kV projection data and a generated complete set of high-kV projection data based on the obtained set of projection data, using one or more neural networks trained for completion of missing projection data, the one or more neural networks trained with multiple complete sets of low-kV projection data and complete sets of high-kV projection data which are acquired in CT scans.