Posterior image sampling using statistical learning model
First Claim
1. A method of image processing using image processor circuitry applying a trained statistical learning model, the method comprising:
- obtaining imaging data of a human subject acquired at least in part using a medical imaging modality;
applying at least one of a reconstruction or a segmentation to the imaging data to generate at least one reconstructed or segmented initial image;
providing the initial image and an image randomness component to a generator neural network, the generator neural network applying a previously trained conditional generative statistical learning model;
generating, with the generator neural network, a plurality of posterior distribution simulated images from the initial image and the image randomness component;
analyzing the posterior distribution simulated images, to identify at least one indication of an image error associated with the initial image; and
providing the indication of the image error associated with the initial image for displaying or further processing.
1 Assignment
0 Petitions
Accused Products
Abstract
Image reconstruction can include using a statistical or machine learning, MAP estimator, or other reconstruction technique to produce a reconstructed image from acquired imaging data. A Conditional Generative Adversarial Network (CGAN) technique can be used to train a Generator, using a Discriminator, to generate posterior distribution sampled images that can be displayed or further processed such as to help provide uncertainty information about a mean reconstruction image. Such uncertainty information can be useful to help understand or even visually modify the mean reconstruction image. Similar techniques can be used in a segmentation use-case, instead of a reconstruction use case. The uncertainty information can also be useful for other post-processing techniques.
7 Citations
20 Claims
-
1. A method of image processing using image processor circuitry applying a trained statistical learning model, the method comprising:
-
obtaining imaging data of a human subject acquired at least in part using a medical imaging modality; applying at least one of a reconstruction or a segmentation to the imaging data to generate at least one reconstructed or segmented initial image; providing the initial image and an image randomness component to a generator neural network, the generator neural network applying a previously trained conditional generative statistical learning model; generating, with the generator neural network, a plurality of posterior distribution simulated images from the initial image and the image randomness component; analyzing the posterior distribution simulated images, to identify at least one indication of an image error associated with the initial image; and providing the indication of the image error associated with the initial image for displaying or further processing. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17)
-
-
18. A method of training image processor circuitry using statistical learning, the method comprising:
-
providing at least one reconstructed or segmented initial image of an anatomical object captured from a human subject to a discriminator neural network and a generator neural network; providing at least one corresponding deemed true image of the anatomical object to the discriminator neural network; using the generator neural network to generate, from the initial image and an image randomness component, posterior distribution simulated images, conditioned on the initial image, provided to the discriminator neural network; and using the discriminator neural network and the initial image to distinguish between the posterior distribution simulated images and the deemed true image, causing training of a statistical learning model, the statistical learning model adapted to be implemented by the generator neural network in generative operations for generating posterior distribution simulated images; wherein analysis of the posterior distribution simulated images generated from a subsequently obtained at least one reconstructed or segmented initial image enables identification of an image error associated with the subsequently obtained at least one reconstructed or segmented initial image. - View Dependent Claims (19, 20)
-
Specification