US 12,070,350 B2
Determining CT scan parameters based on machine learning
Brian Teixeira, Lawrence Township, NJ (US); Vivek Singh, Princeton, NJ (US); Ankur Kapoor, Plainsboro, NJ (US); Andreas Prokein, Bubenreuth (DE); and Dorin Comaniciu, Princeton, NJ (US)
Assigned to Siemens Healthineers AG, Forchheim (DE)
Filed by Siemens Healthcare AG, Forchheim (DE)
Filed on Mar. 30, 2022, as Appl. No. 17/657,181.
Claims priority of application No. 21171208 (EP), filed on Apr. 29, 2021.
Prior Publication US 2022/0346742 A1, Nov. 3, 2022
Int. Cl. A61B 6/00 (2006.01); A61B 6/03 (2006.01)
CPC A61B 6/545 (2013.01) [A61B 6/032 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A computer-implemented method for determining computed tomography (CT) scan parameters for performing a CT scan of an anatomical target region of a patient, the method comprising:
determining, based on optical imaging data depicting the patient, an initial set of attenuation curves associated with the anatomical target region of the patient, wherein said determining of the initial set of attenuation curves associated with the anatomical target region of the patient comprises:
obtaining at least one instance of the optical imaging data;
applying a first trained machine-learning algorithm to the at least one instance of the optical imaging data;
generating, by the first trained machine-learning algorithm, at least one latent vector representing the at least one instance of the optical imaging data;
applying a second trained machine-learning algorithm to the at least one latent vector; and
generating, by the second trained machine-learning algorithm, the initial set of the attenuation curves;
determining an initial set of the CT scan parameters based on the initial set of attenuation curves; and
performing the CT scan starting with the initial set of the CT scan parameters.