US 12,488,446 B2
Apparatus and methods for unsupervised image denoising using double over-parameterization
Tiantian Li, Oakland, CA (US); Zhaoheng Xie, Oakland, CA (US); Wenyuan Qi, Vernon Hills, IL (US); Li Yang, Vernon Hills, IL (US); Evren Asma, Vernon Hills, IL (US); and Jinyi Qi, Oakland, CA (US)
Assigned to The Regents of the University of California, Oakland, CA (US); and Canon Medical Systems Corporation, Otawara (JP)
Filed by The Regents of the University of California, Oakland, CA (US); and Canon Medical Systems Corporation, Otawara (JP)
Filed on Oct. 6, 2022, as Appl. No. 17/961,365.
Claims priority of provisional application 63/302,449, filed on Jan. 24, 2022.
Prior Publication US 2023/0237638 A1, Jul. 27, 2023
Int. Cl. G06T 7/00 (2017.01); G06T 5/70 (2024.01)
CPC G06T 7/0004 (2013.01) [G06T 5/70 (2024.01); G06T 2207/10081 (2013.01); G06T 2207/10088 (2013.01); G06T 2207/10104 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] 14 Claims
OG exemplary drawing
 
1. A method for denoising an image, the method comprising:
receiving a first medical image including a first image of an anatomical structure;
receiving a second medical image including a second image of the anatomical structure; and
training at least one deep image prior (DIP) neural network to produce a denoised image by inputting the second medical image to the at least one DIP neural network and combining a converging noise and an output of the at least one DIP network during the training such that the converging noise combined with the output of the at least one DIP network approximates the first medical image at the end of the training, wherein the output of the DIP network represents the denoised image,
wherein the step of training the at least one DIP neural network comprises
initializing first and second noise vectors;
training the at least one DIP neural network to produce the denoised image by training the first and second noise vectors to be equal to values for which a convolution-based function based on the first and second noise vectors converges to a noise of the first medical image; and
training the at least one DIP neural network to approximate the first medical image minus the convolution-based function, and
wherein the step of training the at least one DIP neural network further comprises training a plurality of DIP neural networks as the at least one DIP neural network using respectively different parameters.