Granted Patent B2

US 12,597,504 · App. 18/512,315 · Granted Apr 7, 2026

Inventors: Daphne Yu (Yardley, PA); Yamini Varadan (East Brunswick, NJ); Poikavila Ullaskrishnan (Lebanon, NH)

Assignee: Siemens Healthineers AG

G16H30/20G06F3/0484G06V10/40G06V10/75G06V10/945G10L15/1822G06V2201/03

Abstract

Claims (59)

1 . A computer-implemented method comprising:

receiving user input comprising 1) a description of desired medical images and 2) viewing preferences for the desired medical images;

determining one or more nodes of a graph of a clinical knowledge ontological database that match the description of the desired medical images, the graph comprising 1) a plurality of nodes each associated with clinical nomenclature of medical images and 2) edges connecting at least some of the plurality of nodes, wherein at least one of the plurality of nodes is associated with at least one medical image that corresponds to its associated clinical nomenclature, wherein the one or more nodes of the graph are determined by:

extracting a vector representation of the description of the desired medical images from the user input,

performing a vector search between the vector representation of the description of the desired medical images and a vector representation of each of the plurality of nodes of the graph of the clinical knowledge ontological database, and

identifying the one or more nodes as matching the description of the desired medical images based on results of the vector search;

generating a display layout of one or more medical images associated with the one or more nodes based on the viewing preferences; and

outputting the display layout.

2 . The computer-implemented method of claim 1 , wherein:

receiving user input comprising 1) a description of desired medical images and 2) viewing preferences for the desired medical images comprises receiving the user input further comprising a temporal description of the desired medical images; and

determining one or more nodes of a graph of a clinical knowledge ontological database that match the description of the desired medical images comprises determining the one or more nodes of the graph of the clinical knowledge ontological database that match the description of the desired medical images based on the temporal description of the desired medical images.

3 . The computer-implemented method of claim 1 , wherein identifying the one or more nodes as matching the description of the desired medical images based on results of the vector search comprises:

identifying one or more highest ranking nodes that satisfy a ranking threshold as the one or more nodes.

4 . The computer-implemented method of claim 1 , further comprising generating the clinical knowledge ontological database by:

extracting features from the at least one medical image using one or more machine learning models;

associating the extracted features with corresponding ones of the plurality of nodes of the graph of the clinical knowledge ontological database; and

outputting the clinical knowledge ontological database with the extracted features associated with the corresponding nodes.

5 . The computer-implemented method of claim 1 , wherein the description of desired medical images comprises a description of an anatomical object of interest to which to navigate within one or more medical images and wherein the one or more matching nodes associated with one or more medical images in the graph of the clinical knowledge ontological database comprises one or more matching nodes associated with coordinates of the anatomical object of interest in the one or more medical images.

6 . The computer-implemented method of claim 1 , wherein the description of the desired medical images is defined based on at least one of imaging modality, acquisition parameter, acquisition orientation, image appearance, anatomical field of view, or classifications and detections.

7 . The computer-implemented method of claim 1 , wherein the viewing preferences are defined based on at least one of anatomical display orientation, windowing, or rendering mode.

8 . The computer-implemented method of claim 1 , wherein generating a display layout of the one or more medical images based on the viewing preferences comprises:

translating the viewing preferences to rendering parameters using a language model.

9 . The computer-implemented method of claim 1 , wherein outputting the display layout comprises:

displaying the display layout on a display device.

10 . An apparatus comprising:

means for receiving user input comprising 1) a description of desired medical images and 2) viewing preferences for the desired medical images;

means for determining one or more nodes of a graph of a clinical knowledge ontological database that match the description of the desired medical images, the graph comprising 1) a plurality of nodes each associated with clinical nomenclature of medical images and 2) edges connecting at least some of the plurality of nodes, wherein at least one of the plurality of nodes is associated with at least one medical image that corresponds to its associated clinical nomenclature, wherein the one or more nodes of the graph are determined by

extracting a vector representation of the description of the desired medical images from the user input,

performing a vector search between the vector representation of the description of the desired medical images and a vector representation of each of the plurality of nodes of the graph of the clinical knowledge ontological database, and

identifying the one or more nodes as matching the description of the desired medical images based on results of the vector search;

means for generating a display layout of one or more medical images associated with the one or more nodes based on the viewing preferences; and

means for outputting the display layout.

11 . The apparatus of claim 10 , wherein:

the means for receiving user input comprising 1) a description of desired medical images and 2) viewing preferences for the desired medical images comprises means for receiving the user input further comprising a temporal description of the desired medical images; and

the means for determining one or more nodes of a graph of a clinical knowledge ontological database that match the description of the desired medical images comprises determining the one or more nodes of the graph of the clinical knowledge ontological database that match the description of the desired medical images based on the temporal description of the desired medical images.

12 . The apparatus of claim 10 , wherein the means for identifying the one or more nodes as matching the description of the desired medical images based on results of the vector search comprises:

means for identifying one or more highest ranking nodes that satisfy a ranking threshold as the one or more nodes.

13 . The apparatus of claim 10 , further comprising means for generating the clinical knowledge ontological database by:

means for extracting features from the at least one medical image using one or more machine learning models;

means for associating the extracted features with corresponding ones of the plurality of nodes of the graph of the clinical knowledge ontological database; and

means for outputting the clinical knowledge ontological database with the extracted features associated with the corresponding nodes.

14 . The apparatus of claim 10 , wherein the description of desired medical images comprises a description of an anatomical object of interest to which to navigate within one or more medical images and wherein the one or more matching nodes associated with one or more medical images in the graph of the clinical knowledge ontological database comprises one or more matching nodes associated with coordinates of the anatomical object of interest in the one or more medical images.

15 . A non-transitory computer-readable medium comprising instructions which, when executed by a computer, cause the computer to carry out operations comprising:

receiving user input comprising 1) a description of desired medical images and 2) viewing preferences for the desired medical images;

determining one or more nodes of a graph of a clinical knowledge ontological database that match the description of the desired medical images, the graph comprising 1) a plurality of nodes each associated with clinical nomenclature of medical images and 2) edges connecting at least some of the plurality of nodes, wherein at least one of the plurality of nodes is associated with at least one medical image that corresponds to its associated clinical nomenclature, wherein the one or more nodes of the graph are determined by:

extracting a vector representation of the description of the desired medical images from the user input,

performing a vector search between the vector representation of the description of the desired medical images and a vector representation of each of the plurality of nodes of the graph of the clinical knowledge ontological database, and

identifying the one or more nodes as matching the description of the desired medical images based on results of the vector search;

generating a display layout of one or more medical images associated with the one or more nodes based on the viewing preferences; and

outputting the display layout.

16 . The non-transitory computer-readable medium of claim 15 , wherein:

receiving user input comprising 1) a description of desired medical images and 2) viewing preferences for the desired medical images comprises receiving the user input further comprising a temporal description of the desired medical images; and

determining one or more nodes of a graph of a clinical knowledge ontological database that match the description of the desired medical images comprises determining the one or more nodes of the graph of the clinical knowledge ontological database that match the description of the desired medical images based on the temporal description of the desired medical images.

17 . The non-transitory computer-readable medium of claim 15 , wherein the description of the desired medical images is defined based on at least one of imaging modality, acquisition parameter, acquisition orientation, image appearance, anatomical field of view, or classifications and detections.

18 . The non-transitory computer-readable medium of claim 15 , wherein the viewing preferences are defined based on at least one of anatomical display orientation, windowing, or rendering mode.

19 . The non-transitory computer-readable medium of claim 15 , wherein generating a display layout of the one or more medical images based on the viewing preferences comprises:

translating the viewing preferences to rendering parameters using a language model.

20 . The non-transitory computer-readable medium of claim 15 , wherein outputting the display layout comprises:

displaying the display layout on a display device.

Assignments (3)

Continuity (1)

References Cited (40)