Prediction of major adverse cardiovascular events (MACE) from AI analysis of pericoronary fat in CT images

Patent metadata

Field	Value
Publication number	US-12573036-B2
Application number	US-202318354921-A
Country	US
Kind code	B2
Filing date	Jul 19, 2023
Priority date	Nov 22, 2022
Publication date	Mar 10, 2026
Grant date	Mar 10, 2026

Abstract

The present disclosure, in some embodiments, relates to a method of generating a prognosis for a patient. The method includes accessing automatically segmented pericoronary adipose tissue (PCAT) corresponding to a patient within an electronic memory. A plurality of non-confounding PCAT features are generated by measuring values of Hounsfield units for an imaging unit within the PCAT. The measured values of the Hounsfield units are predominately free of iodine confounding and artifacts. The plurality of non-confounding PCAT features are provided to a regression model. The regression model is configured to generate a prognosis for the patient using the plurality of non-confounding PCAT features

First claim

What is claimed is:

1 . A non-transitory computer-readable medium storing computer-executable instructions that, when executed, cause a processor to perform operations, comprising:
automatically segmenting one or more computed tomography calcium score (CTCS) images of a patient and identifying automatically segmented pericoronary adipose tissue (PCAT) within the CTCS images of the patient; extracting a plurality of non-confounding PCAT features from the automatically segmented PCAT, wherein measurement of the plurality of non-confounding PCAT features is predominantly free of iodine confounding and artifacts; and providing the plurality of non-confounding PCAT features to a regression model, the regression model being configured to generate a major adverse cardiovascular event (MACE) risk score for the patient using the plurality of non-confounding PCAT features.

2 . The non-transitory computer-readable medium of claim 1 , wherein the operations further comprise:
providing an Agatston score to the regression model, wherein the regression model is configured to generate the MACE risk score using the plurality of non-confounding PCAT features and the Agatston score.

3 . The non-transitory computer-readable medium of claim 1 , wherein extracting the plurality of non-confounding PCAT features includes measuring values of Hounsfield units for a pixel or a voxel within the automatically segmented PCAT.

4 . The non-transitory computer-readable medium of claim 1 , wherein the plurality of non-confounding PCAT features comprise intensity features, morphology features, and texture features.

5 . The non-transitory computer-readable medium of claim 1 , wherein the operations further comprise:
automatically segmenting one or more cardiac computed tomography perfusion (CCTP) images to identify automatically segmented CCTP PCAT; determining potential iodine confounds from the automatically segmented CCTP PCAT; automatically segmenting one or more coronary computed tomography angiogram (CCTA) images to identify automatically segmented CCTA PCAT; generating a plurality of CCTA PCAT features from the automatically segmented CCTA PCAT, wherein the plurality of CCTA PCAT features are modified based upon the potential iodine confounds; and utilizing the plurality of CCTA PCAT features to aid in selection of the plurality of non-confounding PCAT features.

6 . The non-transitory computer-readable medium of claim 1 , wherein the operations further comprise:
automatically segmenting one or more CCTA images to identify CCTA PCAT and coronary arteries; and registering the coronary arteries to the one or more CTCS images.

7 . An assessment apparatus, comprising:
a memory configured to store one or more automatically segmented non-contrast images, the one or more automatically segmented non-contrast images identifying automatically segmented pericoronary adipose tissue (PCAT) within computed tomography calcium score (CTCS) images of a patient; a feature extraction circuit configured to extract a plurality of non-confounding PCAT features from the automatically segmented PCAT, wherein the plurality of non-confounding PCAT features are predominately free of iodine confounding and artifacts; and a regression circuit configured to generate a major adverse cardiovascular event (MACE) risk score for the patient using the plurality of non-confounding PCAT features.

8 . The assessment apparatus of claim 7 , wherein the regression circuit is configured to generate a MACE risk prediction using the plurality of non-confounding PCAT features and an Agatston score of the patient.

9 . The assessment apparatus of claim 7 , wherein extracting the plurality of non-confounding PCAT features includes measuring values of Hounsfield units for a pixel or a voxel within the automatically segmented PCAT.

10 . The assessment apparatus of claim 7 , wherein the plurality of non-confounding PCAT features comprise intensity features, morphology features, and texture features.

11 . The assessment apparatus of claim 7 , wherein the feature extraction circuit is configured to utilize a plurality of CCTA PCAT features extracted from automatically segmented CCTA PCAT within one or more coronary computed tomography angiogram (CCTA) images to aid in selection of the plurality of non-confounding PCAT features.

12 . A method, comprising:
automatically segmenting one or more computed tomography calcium score (CTCS) images of a patient and identifying automatically segmented pericoronary adipose tissue (PCAT) within the CTCS images of the patient; extracting a plurality of non-confounding PCAT features from the automatically segmented PCAT, wherein measurement of the plurality of non-confounding PCAT features is predominantly free of iodine confounding and artifacts; and providing the plurality of non-confounding PCAT features to a regression model, the regression model being configured to generate a major adverse cardiovascular event (MACE) risk score for the patient using the plurality of non-confounding PCAT features.

13 . The method of claim 12 , further comprising:
providing an Agatston score to the regression model, wherein the regression model is configured to generate the MACE risk score using the plurality of non-confounding PCAT features and the Agatston score.

14 . The method of claim 12 , wherein extracting the plurality of non-confounding PCAT features includes measuring values of Hounsfield units for a pixel or a voxel within the automatically segmented PCAT.

15 . The method of claim 12 , wherein the plurality of non-confounding PCAT features comprise intensity features, morphology features, and texture features.

16 . The method of claim 12 , further comprising:
automatically segmenting one or more cardiac computed tomography perfusion (CCTP) images to identify automatically segmented CCTP PCAT; determining potential iodine confounds from the automatically segmented CCTP PCAT; automatically segmenting one or more coronary computed tomography angiogram (CCTA) images to identify automatically segmented CCTA PCAT; generating a plurality of CCTA PCAT features from the automatically segmented CCTA PCAT, wherein the plurality of CCTA PCAT features are modified based upon the potential iodine confounds; and utilizing the plurality of CCTA PCAT features to aid in selection of the plurality of non-confounding PCAT features.

17 . The method of claim 12 , further comprise:
automatically segmenting one or more CCTA images to identify CCTA PCAT and coronary arteries; and registering the coronary arteries to the one or more CTCS images.

Assignees

• Univ Case Western Reserve
• Univ Hospitals Cleveland Medical Center

Inventors

• Wilson David L
• Hoori Ammar
• Hu Tao
• Song Yingnan
• Wu Hao
• Lee Juhwan
• Al-Kindi Sadeer
• Rajagopalan Sanjay

Classifications

• G06V10/26

  Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion · CPC title

• G06T2207/20081

  Training; Learning · CPC title

• G06T2207/10081

  Computed x-ray tomography [CT] · CPC title

• G06T2207/30048

  Heart; Cardiac · CPC title

• G06T7/30

  Determination of transform parameters for the alignment of images, i.e. image registration · CPC title