US 11,776,124 B1
Transforming multispectral images to enhanced resolution images enabled by machine learning
Ali Behrooz, San Mateo, CA (US); and Cheng-Hsun Wu, San Bruno, CA (US)
Assigned to VERILY LIFE SCIENCES LLC, South San Francisco, CA (US)
Filed by Verily Life Sciences LLC, South San Francisco, CA (US)
Filed on May 26, 2022, as Appl. No. 17/804,187.
Application 17/804,187 is a continuation of application No. 16/586,205, filed on Sep. 27, 2019, granted, now 11,354,804.
Int. Cl. G06T 7/00 (2017.01); G06N 20/00 (2019.01); G06T 3/00 (2006.01); G06T 3/40 (2006.01); G06N 3/08 (2023.01)
CPC G06T 7/0014 (2013.01) [G06N 3/08 (2013.01); G06N 20/00 (2019.01); G06T 3/0075 (2013.01); G06T 3/40 (2013.01); G06T 2207/10036 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method, comprising:
receiving an input image of a first biological sample, wherein the input image has a first spatial resolution and a plurality of spectral images, and wherein each spectral image of the plurality of spectral images includes data from a different wavelength band at a different spectral channel;
applying a trained artificial neural network to the input image, the trained artificial neural network trained using an image training data set including a plurality of image pairs, wherein each image pair of the plurality of image pairs includes a first image of a biological sample acquired when the biological sample is unstained, wherein the first image includes a first plurality of spectral images of the biological sample, and wherein each spectral image of the first plurality of spectral images includes data from a different wavelength band; and a second image of the biological sample acquired when the biological sample is stained;
generating an output image at a second spatial resolution, and wherein the output image includes a fewer number of spectral channels than the plurality of spectral images included in the input image; and
outputting the output image.