Facilitating artificial intelligence integration into systems using a distributed learning platform

Patent metadata

Field	Value
Publication number	US-10957442-B2
Application number	US-201816237525-A
Country	US
Kind code	B2
Filing date	Dec 31, 2018
Priority date	Dec 31, 2018
Publication date	Mar 23, 2021
Grant date	Mar 23, 2021

Abstract

Techniques are described that facilitate integrating artificial intelligence (AI) informatics in healthcare systems using a distributed learning platform. In one embodiment, a computer-implemented is provided that comprises interfacing, by a system operatively coupled to a processor, with a medical imaging application that provides for viewing medical image data. The method further comprises, facilitating, by the system, generation of structured diagnostic data according to a defined ontology in association with usage of the imaging application to perform a clinical evaluation of the medical image data. The method further comprises providing, by the system, the structured diagnostic data to one or more machine learning systems, wherein based on the providing, the one or more machine learning systems employ the structured diagnostic data as training data to generate or train one or more diagnostic models configured to provide artificial intelligence-based diagnostic evaluations of new medical image data.

First claim

What is claimed is:

1. A system, comprising:
a memory that stores computer executable components; and
a processor that executes the computer executable components stored in the memory, wherein the computer executable components comprise:
an interface component that facilitates accessing a medical imaging application that provides for viewing medical image data;
a training data generation component that facilitates generating structured training data according to a defined ontology in association with viewing the medical image data using the imaging application, wherein the structured training data comprises measurement data that defines a measurement for one or more features in the medical image data, and wherein the training data generation component comprises:
a measurement component that generates the measurement data based on a measurement marking applied to the one or more features in the medical image data as displayed via a graphical user interface; and

a data collection component that provides the structured training data to one or more machine learning systems, wherein based on reception of the structured training data, the one or more machine learning systems employ the structured training data as training data to generate or train one or more diagnostic models configured to provide artificial intelligence-based diagnostic evaluations of new medical image data.

2. The system of claim 1 , wherein the structured training data further comprises structured markup data that identifies a location of a feature of interest relative to the medical image data, and wherein the training data generation component further comprises:
a markup component that facilitates applying a graphical marking on the feature of interest present in the medical image data as displayed via a graphical user interface and generates the structured markup data based on the graphical marking.

3. The system of claim 2 , wherein the structured markup data comprises one or more medical images comprising the graphical marking and formatted according to the Digital Imaging and Communications in Medicine (DICOM) standard.

4. The system of claim 1 , wherein the structured training data further comprises structured annotation data that identifies a feature of interest relative to the medical image data, and wherein the training data generation component further comprises:
an annotation component that facilitates applying an annotation to the feature of interest present in the medical image data as displayed via the graphical user interface and generates the structured annotation data based on the annotation, and wherein the annotation component restricts application of the annotation using terms or symbols that adhere to the defined ontology.

5. The system of claim 4 , wherein the annotation component determines the one or more annotation terms in accordance with the defined ontology that are relevant to the medical image data and presents the one or more annotation terms via the graphical user interface for potential manual application to the feature of interest.

6. The system of claim 5 , wherein the annotation component employs artificial intelligence and one or more classifiers to determine the one or more annotation terms.

7. The system of claim 5 , wherein the annotation component restricts manual application of the annotations to the medical image data based on the one or more annotation terms.

8. The system of claim 1 , wherein the measurement component further facilitates applying the measurement marking using one or more defined measurement marking tools of the graphical user interface.

9. The system of claim 1 , wherein the measurement data comprises a dimension of the one or more features, and wherein the measurement component determines the dimension based on the measurement marking.

10. The system of claim 1 , wherein the structured training data further comprises structured segmentation data that defines one or more distinct visual features extracted from the medical image data, and wherein the training data generation component further comprises:
a segmentation component that identifies the one or more distinct visual features in the medical image data, extracts the one or more distinct visual features from the medical image data, and generates structured segmentation data based on the one or more distinct visual features as extracted from the medical image data.

11. The system of claim 1 , wherein the structured training data further comprises structured training data regarding a diagnostic interpretation of the medical image data, and wherein the training data generation component further comprises:
a reporting component that facilitates generating a textual diagnostic report regarding the diagnostic interpretation of the medical image data, wherein the reporting component further generates the structured training data based on the textual diagnostic report.

12. The system of claim 11 , wherein the reporting component determines a subset of medical terms or information categories for potential inclusion in the textual diagnostic report based on the medical image data and the defined ontology and presents the subset of medical terms or information categories via the graphical user interface.

13. The system of claim 12 , wherein the reporting component employs artificial intelligence and one or more classifiers to determine the subset of medical terms or information categories.

14. The system of claim 12 , wherein the reporting component further restricts content for inclusion in the textual diagnostic report based on the subset of the medical terms or information categories.

15. The system of claim 11 , wherein reporting component employs natural language processing to identify one or more semantically relevant terms or phrases included in the textual diagnostic report, and wherein the reporting component further generates the structured textual report data using the semantically relevant terms or phrases.

16. A method, comprising:
interfacing, by a system operatively coupled to a processor, with a medical imaging application that provides for viewing medical image data;
generating, by the system, structured training data according to a defined ontology in association with usage of the imaging application to perform a clinical evaluation of the medical image data, wherein the generating comprises:
generating, by the system measurement data that defines a measurement for one or more features in the medical image data based on a measurement marking applied to the one or more features in the medical image data as displayed via a graphical user interface; and

providing, by the system, the structured training data to one or more machine learning systems, wherein based on the providing, the one or more machine learning systems employ the structured training data as training data to generate or train one or more diagnostic models configured to provide artificial intelligence-based diagnostic evaluations of new medical image data.

17. The method of claim 16 , further comprising:
providing, by the system, a markup tool that facilitates applying a graphical marking on a feature of interest present in the medical image data as displayed via the graphical user interface, and wherein the generating the structured training data further comprises:
generating, by the system, structured markup data that identifies a location of the feature of interest relative to the medical image data based on the graphical marking.

18. The method of claim 16 , further comprising:
providing,

Assignees

• Ge Prec Healthcare Llc

Inventors

• Kalafut John
• Dreyer Keith
• Michalski Mark
• Pomerantz Stuart
• Doyle Sean
• Tenenholtz Neil

Classifications

• G16H50/70

  for mining of medical data, e.g. analysing previous cases of other patients · CPC title

• G06F40/30

  Semantic analysis · CPC title

• G16H50/20

  for computer-aided diagnosis, e.g. based on medical expert systems · CPC title

• G06N20/00

  Machine learning · CPC title

• G16H30/40

  for processing medical images, e.g. editing · CPC title