US 11,918,398 B2
Analysis method and analysis unit for determining radiological result data
Peter Huber, Windsbach (DE); Bernhard Krauss, Burgthann (DE); Sebastian Schmidt, Weisendorf (DE); and Bernhard Schmidt, Fuerth (DE)
Assigned to SIEMENS HEALTHINEERS AG, Erlangen (DE)
Filed by Siemens Healthcare GmbH, Erlangen (DE)
Filed on Mar. 20, 2019, as Appl. No. 16/358,790.
Claims priority of application No. 18165007 (EP), filed on Mar. 29, 2018.
Prior Publication US 2019/0298288 A1, Oct. 3, 2019
Int. Cl. A61B 6/00 (2006.01); A61B 5/00 (2006.01); A61B 6/03 (2006.01); G06N 3/04 (2023.01); G06N 3/045 (2023.01); G06N 3/08 (2023.01); G06T 7/00 (2017.01); G06N 3/044 (2023.01)
CPC A61B 6/482 (2013.01) [A61B 5/7267 (2013.01); A61B 6/032 (2013.01); A61B 6/4241 (2013.01); A61B 6/54 (2013.01); G06N 3/04 (2013.01); G06N 3/045 (2023.01); G06N 3/08 (2013.01); G06T 7/0012 (2013.01); G06N 3/044 (2023.01); G06T 2207/10081 (2013.01)] 21 Claims
OG exemplary drawing
 
1. An analysis method for automatically determining radiological result data from radiology data sets, the analysis method comprising:
provisioning a first radiology data set, the first radiology data set being at least based on X-ray data of a first X-ray energy spectrum;
provisioning at least one second radiology data set, the at least one second radiology data set being at least based on X-ray data of a second X-ray energy spectrum;
provisioning an analysis unit, the analysis unit including a neural network configured to analyze the first radiology data set and the at least one second radiology data set, the neural network including
an input layer with a plurality of cells, wherein a number of the plurality of cells of the input layer are each assigned only one pixel value or only one raw data value of the first radiology data set or the at least one second radiology data set,
a number of intermediate layers, and
an output layer representing the radiological result data;
dividing the first radiology data set into a first subset and a second subset;
dividing the at least one second radiology data set into a first subset and a second subset;
analyzing, via the neural network in a first analyzing step, the first subset of the first radiology data set and the first subset of the at least one second radiology data set, the first subset of the first radiology data set being assigned to a first set of cells of the input layer and the first subset of the at least one second radiology data set being assigned to a second set of cells of the input layer, for joint processing of the first subset of the first radiology data set and the first subset of the at least one second radiology data set in the neural network, the first set of cells being different than the second set of cells;
analyzing, via the neural network in a second analyzing step, the second subset of the first radiology data set and the second subset of the at least one second radiology data set, the second subset of the first radiology data set being assigned to the first set of cells of the input layer and the second subset of the at least one second radiology data set being assigned to the second set of cells of the input layer, for joint processing of the second subset of the first radiology data set and the second subset of the at least one second radiology data set in the neural network, the first set of cells being different than the second set of cells; and
acquiring the radiological result data on the output layer.