US 12,272,441 B2
Method and systems for predicting a stream of virtual topograms
Brian Teixeira, Princeton, NJ (US); Vivek Singh, Princeton, NJ (US); Ankur Kapoor, Plainsboro, NJ (US); Yao-jen Chang, Princeton, NJ (US); and Birgi Tamersoy, Erlangen (DE)
Assigned to Siemens Healthineers AG, Forchheim (DE)
Filed by Siemens Healthineers AG, Forchheim (DE)
Filed on Sep. 30, 2021, as Appl. No. 17/489,838.
Claims priority of application No. 10 2020 213 489.8 (DE), filed on Oct. 27, 2020.
Prior Publication US 2022/0130524 A1, Apr. 28, 2022
Int. Cl. G16H 30/40 (2018.01); G06F 18/214 (2023.01); G06N 20/00 (2019.01)
CPC G16H 30/40 (2018.01) [G06F 18/2148 (2023.01); G06N 20/00 (2019.01); G06T 2207/10012 (2013.01); G06T 2207/10081 (2013.01); G06T 2207/30004 (2013.01)] 19 Claims
OG exemplary drawing
 
1. A computer-implemented method of predicting a stream of virtual topograms, the method comprising:
continuously receiving sets of image data of a subject; and
continuously generating virtual topograms based on each received set of image data by a trained machine learning algorithm (MLA), the continuously generating comprising:
updating a representation of body landmarks based on each received set of image data by a trained body marker detector (BMD) of the trained MLA; and
predicting, by a trained topogram generator (TG) of the trained MLA, the virtual topogram based on input of the updated representation of body landmarks and on input of the corresponding set of image data to the TG of the trained MLA.