Back to results
US 2026/0253265 A1Published application
Generating synthetic radiologic images
AssigneeBayer Aktiengesellschaft
InventorsDinh Tuan TRUONG, Matthias LENGA, Melanie Ellen-Ruth Hildeg DOHMEN, Ivo Matteo BALTRUSCHAT
Life of this patent
Estimating the term…
Abstract
Systems, methods, and computer programs disclosed herein relate to generating synthetic radiologic images.
Claims20
A computer-implemented method comprising:
- providing a trained conditional generative model,
- providing one or more conditional inputs,
- generating a number N of synthetic radiologic images from different seeds based on the one or more conditional inputs using the trained generative model,
- generating a combined synthetic radiologic image based on the number N of synthetic radiologic images,
- outputting and/or storing the combined synthetic radiologic image, and/or transmitting the combined synthetic radiologic image to a separate computer system.
2 more elements
A computer system comprising:
- a processor;
- and
- a memory storing an application program configured to perform, when executed by the processor,
- an operation, the operation comprising:
- providing a trained conditional generative model,
- providing one or more conditional inputs,
- generating a number of synthetic radiologic images from different seeds based on the one or more conditional inputs using the trained generative model,
- generating a combined synthetic radiologic image based on the number of synthetic radiologic images,
- outputting and/or storing the combined synthetic radiologic image, and/or transmitting the combined synthetic radiologic image to a separate computer system.
6 more elements
A non-transitory computer readable storage medium having stored thereon software instructions that, when executed by a processor of a computer system, cause the computer system to:
- provide a trained conditional generative model,
- provide one or more conditional inputs,
- generate a number of synthetic radiologic images from different seeds based on the one or more conditional inputs using the trained generative model,
- generate a combined synthetic radiologic image based on the number of synthetic radiologic images,
- output and/or store the combined synthetic radiologic image, and/or transmit the combined synthetic radiologic image to a separate computer system.
2 more elements
Dependencies are inferred from the claim text.