US 12,086,908 B2
Reconstruction with magnetic resonance compressed sensing
Boris Mailhe, Plainsboro, NJ (US); Charles Millard, Oxford (GB); and Mariappan S. Nadar, Plainsboro, NJ (US)
Assigned to Siemens Healthineers AG, Forchheim (DE)
Filed by Siemens Healthineers AG, Forchheim (DE)
Filed on Jun. 8, 2021, as Appl. No. 17/303,790.
Claims priority of provisional application 63/199,091, filed on Dec. 7, 2020.
Claims priority of provisional application 63/199,090, filed on Dec. 7, 2020.
Prior Publication US 2022/0180574 A1, Jun. 9, 2022
Int. Cl. G06T 11/00 (2006.01); G01R 33/3415 (2006.01); G01R 33/385 (2006.01); G01R 33/483 (2006.01); G01R 33/56 (2006.01); G01R 33/561 (2006.01)
CPC G06T 11/005 (2013.01) [G01R 33/3415 (2013.01); G01R 33/3854 (2013.01); G01R 33/483 (2013.01); G01R 33/5608 (2013.01); G01R 33/5611 (2013.01); G06T 11/006 (2013.01); G06T 11/008 (2013.01); G06T 2210/41 (2013.01); G06T 2211/424 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method for reconstructing medical image data, the method comprising:
acquiring, by a magnetic resonance imaging device with multiple coils, compressed sensing data;
transforming, by a processor, the compressed sensing data into an estimated image;
transforming, by the processor, the estimated image into wavelets;
estimating, by the processor, a covariance of an error between the wavelets and a target image with a spatial modulation of variance within each wavelet subband;
denoising, by the processor, the estimated image based on a noise model from the modulated covariance, the denoising resulting in an updated image; and
outputting an output image, the output image being based on the updated image.