US 12,119,090 B1
Utilizing masked autoencoder generative models to extract microscopy representation autoencoder embeddings
Oren Zeev Kraus, Toronto (CA); Kian Runnels Kenyon-Dean, Toronto (CA); Mohammadsadegh Saberian, Kitchener (CA); Maryam Fallah, Toronto (CA); Peter Foster McLean, Centerville, UT (US); Jessica Wai Yin Leung, Toronto (CA); Vasudev Sharma, Toronto (CA); Ayla Yasmin Khan, Salt Lake City, UT (US); Jaichitra Balakrishnan, Sudbury, MA (US); Safiye Celik, Sudbury, MA (US); Dominique Beaini, Montreal (CA); Maciej Sypetkowski, Warsaw (PL); Chi Cheng, Salt Lake City, UT (US); Kristen Rose Morse, Cottonwood Heights, UT (US); Maureen Katherine Makes, Salt Lake City, UT (US); Benjamin John Mabey, Millcreek, UT (US); and Berton Allen Earnshaw, Cedar Hills, UT (US)
Assigned to Recursion Pharmaceuticals, Inc., Salt Lake City, UT (US)
Filed by Recursion Pharmaceuticals, Inc., Salt Lake City, UT (US)
Filed on Dec. 19, 2023, as Appl. No. 18/545,399.
Int. Cl. G06T 15/00 (2011.01); G06T 5/73 (2024.01); G16B 20/00 (2019.01); G16B 40/00 (2019.01); G16B 45/00 (2019.01)
CPC G16B 45/00 (2019.02) [G06T 5/73 (2024.01); G16B 20/00 (2019.02); G16B 40/00 (2019.02); G06T 2207/10056 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
generating a masked training phenomic image by applying a mask to a training phenomic image portraying a cell phenotype; and
training a generative machine learning model to generate phenomic perturbation image embeddings from the training phenomic image by:
generating, utilizing the generative machine learning model, a predicted phenomic image from the masked training phenomic image;
generating a Fourier transformation loss between the predicted phenomic image and the training phenomic image; and
modifying parameters of the generative machine learning model utilizing the Fourier transformation loss and a first Fourier weight.