US 12,437,452 B2
Noise suppression using deep convolutional networks
William J. Sehnert, Fairport, NY (US); Karin Toepfer, Rochester, NY (US); Levon O. Vogelsang, Webster, NY (US); and Lori L. Barski, Pittsford, NY (US)
Assigned to Carestream Health, Inc., Rochester, NY (US)
Appl. No. 18/040,982
Filed by CARESTREAM HEALTH, INC., Rochester, NY (US)
PCT Filed Aug. 30, 2021, PCT No. PCT/US2021/048142
§ 371(c)(1), (2) Date Feb. 8, 2023,
PCT Pub. No. WO2022/051199, PCT Pub. Date Mar. 10, 2022.
Claims priority of provisional application 63/074,129, filed on Sep. 3, 2020.
Prior Publication US 2023/0306657 A1, Sep. 28, 2023
Int. Cl. G06T 5/70 (2024.01); G06T 11/00 (2006.01)
CPC G06T 11/006 (2013.01) [G06T 5/70 (2024.01); G06T 2207/10124 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2211/424 (2013.01)] 15 Claims
OG exemplary drawing
 
1. A computer-implemented method of generating a noise suppressed radiographic image, the method comprising the steps of:
training a machine learning network to generate a noise field image from a current radiographic image by:
accessing a plurality of previously acquired standard exposure radiographic images;
conditioning each of the accessed plurality of previously acquired standard exposure radiographic images with simulated noise content to form a plurality of simulated low-exposure images,
associating each simulated low-exposure image with its corresponding previously acquired standard exposure radiographic image to form a plurality of learning pairs of radiographic images; and
training the machine learning network to generate a noise field image using the plurality of learning pairs of radiographic images;
capturing a current radiographic image of an object and generating therefrom a corresponding noise field image using the trained machine learning network;
suppressing noise in the current radiographic image of the object including applying a scaling factor to at least a portion of the corresponding noise field image and combining the scaled noise field image and the current radiographic image of the object; and
displaying, storing, or transmitting the noise suppressed radiographic image of the object.