Inference process visualization system for medical scans

Patent metadata

Field	Value
Publication number	US-12061994-B2
Application number	US-202016990086-A
Country	US
Kind code	B2
Filing date	Aug 11, 2020
Priority date	Aug 11, 2020
Publication date	Aug 13, 2024
Grant date	Aug 13, 2024

Abstract

An inference process visualization system is configured to generate inference process visualization data for a medical scan indicating an inference process flow of plurality of sub-models applied to the medical scan and further indicating a plurality of inference data for the medical scan generated by applying the plurality of sub-models in accordance with the inference process flow. The inference process visualization system is further configured to facilitate display of the inference process visualization data via an interactive interface.

First claim

What is claimed is:

1. An inference process visualization system, comprising:
a processing system that includes a processor; and
a memory device that stores executable instructions that, when executed by the processing system, configure the processor to perform operations comprising:
generating inference process visualization data for a medical scan indicating an inference process flow of a plurality of sub-models applied to said medical scan and further indicating a plurality of inference data for said medical scan generated by applying said plurality of sub-models in accordance with said inference process flow comprising the steps of:
presenting a set of possible sub-models via an interactive interface;
receiving sub-model selection data based on user input to said interactive interface indicating said plurality of sub-models as a proper subset of said set of possible sub-models, wherein said plurality of sub-models are applied to said medical scan based on said sub-model selection data;
presenting a prompt via said interactive interface to select said inference process flow of said set of possible sub-models;
receiving inference process flow selection data based on user input to said interactive interface indicating said inference process flow of said plurality of sub-models, wherein said plurality of sub-models are applied to said medical scan in accordance with said inference process flow based on said inference process flow selection data;

facilitating display of said inference process visualization data via said interactive interface.

2. The inference process visualization system of claim 1 , wherein the executable instructions, when executed by the processing system, further configure the processor to perform operations comprising:
generating the plurality of inference data for the medical scan by applying the plurality of sub-models in accordance with said inference process flow.

3. The inference process visualization system of claim 1 , wherein the executable instructions, when executed by the processing system, further configure the processor to perform operations comprising:
automatically selecting said plurality of sub-models as a proper subset of a set of possible sub-models.

4. The inference process visualization system of claim 3 , wherein said inference process flow of the plurality of sub-models includes at least one: of a serialized ordering of at least two of said plurality of sub-models, or at least two of said plurality of sub-models to be applied in parallel.

5. The inference process visualization system of claim 1 , wherein a first set of said plurality of sub-models is automatically selected based on an output of a second set of said plurality of sub-models, wherein said inference process visualization data indicates said first set of said plurality of sub-models is indicated after said second set of said plurality of sub-models in a serialized ordering of said inference process flow, and wherein said inference process visualization data indicates said first set of said plurality of inference data, generated by applying said second set of said plurality of sub-models, indicates selection of said first set of said plurality of sub-models.

6. The inference process visualization system of claim 1 , wherein said plurality of sub-models includes at least one triage model, at least one detection model, at least one measurement model, at least one characterizing model, and at least one longitudinal tracking model.

7. The inference process visualization system of claim 6 , wherein said inference process flow of said plurality of sub-models further includes at least one of: at least one longitudinal characterizing model or at least one longitudinal measurement model.

8. The inference process visualization system of claim 6 , wherein at least one of: said at least one triage model is selected from a plurality of triage model options, said at least one detection model is selected from a plurality of detection model options, said at least one measurement model is selected from a plurality of measurement model options, said at least one characterizing model is selected from a plurality of characterizing model options, or said at least one longitudinal tracking model is selected from a plurality of characterizing model options.

9. The inference process visualization system of claim 1 , wherein said plurality of sub-models includes at least one of: at least one body part classifier model, at least one view classifier model, or at least one demographic classifier model.

10. The inference process visualization system of claim 1 , wherein said inference process visualization data includes model description data for each of said plurality of sub-models.

11. The inference process visualization system of claim 1 , wherein display of the inference process visualization data include display of at least one of said plurality of inference data as corresponding abnormality annotation data displayed in conjunction with display of said medical scan.

12. The inference process visualization system of claim 1 , wherein said plurality of sub-models corresponds to one of a plurality of medical scan classification categories corresponding to said medical scan.

13. The inference process visualization system of claim 12 , wherein said plurality of medical scan classification categories corresponds to at least one of: a plurality of body parts or a plurality of view types.

14. A method, comprising:
generating inference process visualization data for a medical scan indicating an inference process flow of a plurality of sub-models applied to said medical scan and further indicating a plurality of inference data for said medical scan generated by applying said plurality of sub-models in accordance with the inference process flow comprising the steps of:
presenting a set of possible sub-models via an interactive interface;
receiving sub-model selection data based on user input to said interactive interface indicating said plurality of sub-models as a proper subset of said set of possible sub-models, wherein said plurality of sub-models are applied to said medical scan based on said sub-model selection data;
presenting a prompt via said interactive interface to select said inference process flow of said set of possible sub-models;
receiving inference process flow selection data based on user input to said interactive interface indicating said inference process flow of said plurality of sub-models, wherein said plurality of sub-models are applied to said medical scan in accordance with said inference process flow based on said inference process flow selection data;

facilitating display of the inference process visualization data via said interactive interface.

15. The method of claim 14 , further comprising:
generating the plurality of inference data for said medical scan by applying said plurality of sub-models in accordance with the inference process flow.

16. The method of claim 14 , further comprising:
automatically selecting said plurality of sub-models as a proper subset of a set of possible sub-models.

17. The method of claim 14 , wherein said process flow of said plurality of sub-models includes a process flow of at least one triage model, at least one detection model, at least one measurement model, at least one characterizing model, and at least one longitudinal tracking model.

18. The method of claim 14 , wherein said process flow of said plurality of sub-models includes a process flow of at least one of: at least one body part classifier model, at

Assignees

• Enlitic Inc

Inventors

• Lyman Kevin
• Upton Anthony
• Francis Jordan
• Li Vicky

Classifications

• G06N3/09

  Supervised learning · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G16H30/20

  for handling medical images, e.g. DICOM, HL7 or PACS · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title