Medical imaging stroke model

Patent metadata

Field	Value
Publication number	US-11545266-B2
Application number	US-201916588080-A
Country	US
Kind code	B2
Filing date	Sep 30, 2019
Priority date	Sep 30, 2019
Publication date	Jan 3, 2023
Grant date	Jan 3, 2023

Abstract

Systems and techniques for generating and/or employing a medical imaging stroke model are presented. In one example, a system employs a convolutional neural network to generate output data regarding a brain anatomical region based on diffusion-weighted imaging (DWI) data associated with the brain anatomical region and apparent diffusion coefficient (ADC) data associated with the brain anatomical region. The system also detects presence or absence of a medical stroke condition associated with the brain anatomical region based on the output data.

First claim

What is claimed is:

1. A system, comprising:
a memory that stores computer executable components; and
a processor that executes computer executable components stored in the memory, wherein the computer executable components comprise:
a machine learning component that employs a convolutional neural network to generate output data regarding a brain anatomical region based on diffusion-weighted imaging (DWI) data associated with the brain anatomical region and apparent diffusion coefficient (ADC) data associated with the brain anatomical region, wherein the output data comprises a segmentation mask, generated by the machine learning component, that corresponds to the brain anatomical region, and wherein the convolutional neural network is trained using a segmentation loss generated using a ground truth mask; and
a medical diagnosis component that detects presence or absence of a medical stroke condition associated with the brain anatomical region based on the output data and patient data representative of a state of a patient associated with the brain anatomical region.

2. The system of claim 1 , wherein the medical diagnosis component detects presence or absence of an ischemic stroke condition associated with the brain anatomical region based on the output data.

3. The system of claim 1 , wherein the medical diagnosis component generates a contour mask associated with the medical stroke condition based on the output data.

4. The system of claim 1 , wherein the medical diagnosis component detects presence or absence of an infarct associated with the brain anatomical region based on the output data.

5. The system of claim 1 , wherein the medical diagnosis component determines a size of an infarct associated with the brain anatomical region based on the output data.

6. The system of claim 1 , wherein the medical diagnosis component calculates a volume of an infarct associated with the brain anatomical region based on the output data.

7. The system of claim 1 , wherein the DWI data is associated with darkness information for the brain anatomical region and the ADC data is associated with brightness information for the brain anatomical region.

8. The system of claim 1 , further comprising:
a display component that generates display data associated with the presence or the absence of the medical stroke condition in a human-interpretable format.

9. The system of claim 8 , wherein the display component generates a multi-dimensional visualization associated with the presence or the absence of the medical stroke condition, wherein the multi-dimensional visualization comprises a segmented DWI image, and wherein the machine learning component generates the segmented DWI image, using machine learning, by overlaying the segmentation mask on a DWI image associated with the DWI data.

10. The system of claim 1 , wherein the medical diagnosis component determines a defined therapy or medication treatment regime based on the output data and the patient data.

11. A method, comprising:
employing, by a system comprising a processor, a convolutional neural network to generate output data regarding a brain anatomical region based on diffusion-weighted imaging (DWI) data associated with the brain anatomical region and apparent diffusion coefficient (ADC) data associated with the brain anatomical region, wherein the output data comprises a segmentation mask, generated using machine learning, that corresponds to the brain anatomical region, and wherein the convolutional neural network is trained using a segmentation loss generated using a ground truth mask; and
detecting, by the system, presence or absence of a medical stroke condition associated with the brain anatomical region based on the output data and patient data representative of a state of a patient associated with the brain anatomical region.

12. The method of claim 11 , wherein the detecting comprises detecting presence or absence of an ischemic stroke condition associated with the brain anatomical region based on the output data.

13. The method of claim 11 , wherein the detecting comprises detecting presence or absence of an infarct associated with the brain anatomical region based on the output data.

14. The method of claim 11 , further comprising:
determining, by the system, a size of an infarct associated with the brain anatomical region based on the output data.

15. The method of claim 11 , further comprising:
generating, by the system, display data associated with the presence or the absence of the medical stroke condition in a human-interpretable format.

16. The method of claim 11 , further comprising:
generating, by the system, a multi-dimensional visualization associated with the presence or the absence of the medical stroke condition, wherein the multi-dimensional visualization comprises a segmented DWI image, and wherein the segmented DWI image is generated, using machine learning, by overlaying the segmentation mask on a DWI image associated with the DWI data.

17. A computer readable storage device comprising instructions that, in response to execution, cause a system comprising a processor to perform operations, comprising:
generating, using a convolutional neural network, output data regarding a brain anatomical region based on diffusion-weighted imaging (DWI) data associated with the brain anatomical region and apparent diffusion coefficient (ADC) data associated with the brain anatomical region, wherein the output data comprises a segmentation mask, generated using machine learning, that corresponds to the brain anatomical region, and wherein the convolutional neural network is trained using a segmentation loss generated using a ground truth mask; and
detecting presence or absence of a medical stroke condition associated with the brain anatomical region based on the output data and patient data representative of a state of a patient associated with the brain anatomical region.

18. The computer readable storage device of claim 17 , wherein the detecting comprises detecting presence or absence of an infarct associated with the brain anatomical region based on the output data.

19. The computer readable storage device of claim 17 , wherein the operations further comprise:
determining a size of an infarct associated with the brain anatomical region based on the output data.

20. The computer readable storage device of claim 17 , wherein the operations further comprise:
calculating a volume of an infarct associated with the brain anatomical region based on the output data.

Assignees

• Ge Prec Healthcare Llc
• Partners Healthcare System Inc
• Massachusetts Gen Hospital
• Brigham & Womens Hospital Inc

Inventors

• Kalafut John Francis
• Bizzo Bernardo
• Pedemonte Stefano
• Bridge Christopher
• Tenenholtz Neil
• Gonzalez Ramon Gilberto

Classifications

• G16H40/63

  for local operation · CPC title

• G06T2207/10088

  Magnetic resonance imaging [MRI] · CPC title

• G06N3/08

  Learning methods · CPC title

• G16H15/00

  ICT specially adapted for medical reports, e.g. generation or transmission thereof · CPC title

• G06T2207/20084

  Artificial neural networks [ANN] · CPC title