US 11,783,451 B2
Systems and methods for reducing colored noise in medical images using deep neural network
Daniel Litwiller, Denver, CO (US); Xinzeng Wang, Houston, TX (US); Ali Ersoz, Brookfield, WI (US); Robert Marc Lebel, Calgary (CA); Ersin Bayram, Houston, TX (US); and Graeme Colin McKinnon, Hartland, WI (US)
Assigned to GE Precision Healthcare LLC, Milwaukee, WI (US)
Filed by GE Precision Healthcare LLC, Milwaukee, WI (US)
Filed on Mar. 2, 2020, as Appl. No. 16/806,689.
Prior Publication US 2021/0272240 A1, Sep. 2, 2021
Int. Cl. G06T 5/00 (2006.01); G06T 7/00 (2017.01); A61B 6/00 (2006.01)
CPC G06T 5/002 (2013.01) [A61B 6/5258 (2013.01); G06T 7/0012 (2013.01); G06T 2207/10024 (2013.01); G06T 2207/10088 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20182 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A method comprising:
receiving a medical image acquired by an imaging system, wherein the medical image comprises colored noise;
mapping the medical image to a de-noised medical image without colored noise using a trained convolutional neural network (CNN);
displaying the de-noised medical image via a display device;
wherein mapping the medical image to the de-noised medical image using the trained CNN further includes: acquiring one or more noise parameters corresponding to a source of the colored noise and incorporating the one or more noise parameters into the trained CNN; and
wherein the one or more noise parameters are derived from at least one of a k-space sampling pattern used to acquire the medical image or a k-space sampling density used to acquire the medical image.