US 12,299,075 B2
Computer-implemented method for parametrizing a function for evaluating a medical image dataset
Alexander Muehlberg, Nuremberg (DE); Oliver Taubmann, Weilersbach (DE); Alexander Katzmann, Fuerth (DE); Felix Denzinger, Nuremberg (DE); Felix Lades, Erlangen (DE); Rainer Kaergel, Stegaurach (DE); Felix Durlak, Langenzenn (DE); and Michael Suehling, Erlangen (DE)
Assigned to Siemens Healthineers AG, Forchheim (DE)
Filed by Siemens Healthcare GmbH, Erlangen (DE)
Filed on Jul. 22, 2021, as Appl. No. 17/382,588.
Claims priority of application No. 20188174 (EP), filed on Jul. 28, 2020.
Prior Publication US 2022/0036136 A1, Feb. 3, 2022
Int. Cl. G06F 18/21 (2023.01); G06F 18/214 (2023.01); G06F 18/24 (2023.01); G06N 3/045 (2023.01); G06V 10/25 (2022.01); G16H 30/40 (2018.01)
CPC G06F 18/217 (2023.01) [G06F 18/214 (2023.01); G06F 18/24765 (2023.01); G06N 3/045 (2023.01); G06V 10/25 (2022.01); G16H 30/40 (2018.01); G06V 2201/03 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method for parametrizing a function for evaluating a medical image dataset concerning a region of interest, the method comprising:
receiving a training dataset including multiple training medical image datasets;
acquiring at least one representation type of the region of interest based on the training medical image datasets;
generating a respective representation parametrization of the training medical image datasets for each of the at least one representation type;
determining, using an optimization algorithm, at least one relevant representation type, of the at least one representation type, based on the respective representation parametrizations, in response to the at least one representation type including more than one representation type;
determining, using the optimization algorithm, a processing algorithm based on at least one of the at least one relevant representation type or the respective representation parametrizations, the processing algorithm configured to determine at least one output parameter based on the respective representation parametrizations; and
providing the at least one relevant representation type and the processing algorithm or processing parameters describing the processing algorithm, wherein the optimization algorithm is configured to optimize performance of the processing algorithm when the processing algorithm operates on a set of training representations generated based on at least a subset of the training medical image datasets using at least one respective representation parametrization of the respective representation parametrizations.