Systems and methods for medical imaging

Patent metadata

Field	Value
Publication number	US-11367193-B2
Application number	US-201916731407-A
Country	US
Kind code	B2
Filing date	Dec 31, 2019
Priority date	Dec 31, 2019
Publication date	Jun 21, 2022
Grant date	Jun 21, 2022

Abstract

A method for medical imaging may include obtaining a plurality of successive images of a region of interest (ROI) including at least a portion of an object's heart. The plurality of successive images may be based on imaging data acquired from the ROI by a scanner without electrocardiography (ECG) gating. The plurality of successive images may be related to one or more cardiac cycles of the object's heart. The method may also include automatically determining, in the plurality of successive images, target images that correspond to at least one of the one or more cardiac cycles of the object's heart.

First claim

What is claimed is:

1. A system for medical imaging, comprising:
at least one storage device including a set of instructions; and
at least one processor in communication with the at least one storage device, wherein when executing the set of instructions, the at least one processor is directed to cause the system to:
obtain a plurality of successive images of a region of interest (ROI) including at least a portion of an object's heart, wherein:
the plurality of successive images are based on imaging data acquired from the ROI by a scanner without electrocardiography (ECG) gating, and
the plurality of successive images are related to one or more cardiac cycles of the object's heart;

automatically determine, in the plurality of successive images, target images that correspond to at least one of the one or more cardiac cycles of the object's heart; and
for at least one of the target images, determine corresponding time information based on a temporal resolution of the plurality of successive images.

2. The system of claim 1 , wherein to automatically determine, in the plurality of successive images, the target images that correspond to the at least one of the one or more cardiac cycles of the object's heart, the at least one processor is further directed to cause the system to:
determine a reference image based on the plurality of successive images, the reference image relating to cardiac diastole of the object's heart;
determine correlations between the reference image and at least multiple of the plurality of successive images;
determine, in the plurality of successive images, diastolic images relating to the cardiac diastole based on the correlations;
determine two or more terminal images in the diastolic images; and
determine the target images by including images between two neighboring terminal images of the two or more terminal images in the plurality of successive images.

3. The system of claim 2 , wherein to determine the reference image based on the plurality of successive images, the at least one processor is further directed to cause the system to:
divide the plurality of successive images into two or more first groups;
select one of the two or more first groups, a count of images in the selected first group being greatest among the two or more first groups;
determine a median image based on the images of the selected first group, a pixel value of each pixel in the median image being a median of pixel values of corresponding pixels of the images in the selected first group; and
determine the reference image based on the median image.

4. The system of claim 3 , wherein to determine the reference image based on the median image, the at least one processor is further directed to cause the system to:
determine the median image as the reference image; or
wherein to determine the reference image based on the median image, the at least one processor is further directed to cause the system to:
determine a similarity degree between the median image and at least one of the images in the selected first group by comparing the pixel value of each pixel of the median image with the pixel value of the corresponding pixel in the at least one of the images in the selected first group; and
determine one of the images in the selected first group as the reference image based on the similarity degree.

5. The system of claim 2 , wherein to determine, in the plurality of successive images, the diastolic images relating to the cardiac diastole based on the correlations, the at least one processor is further directed to cause the system to:
determine a median of the correlations of the at least multiple of the plurality of successive images; and
determine the diastolic images based on the median of the correlations, the correlations of the diastolic images being larger than the median of the correlations of the at least multiple of the plurality of successive images.

6. The system of claim 2 , wherein to determine the two or more terminal images in the diastolic images, the at least one processor is further directed to cause the system to:
divide the diastolic images into two or more second groups using K-means clustering;
determine a number for each of the diastolic images based on a time order of the diastolic images; and
in each of the two or more second groups, determine the image the number of which is a median of the numbers of the diastolic images in the second group as the terminal image.

7. The system of claim 1 , wherein to automatically determine, in the plurality of successive images, the target images that correspond to the at least one of the one or more cardiac cycles of the object's heart, the at least one processor is further directed to cause the system to:
determine the target images using machine learning.

8. The system of claim 7 , wherein to determine the target images using machine learning, the at least one processor is further directed to cause the system to:
determine two or more terminal images in the plurality of successive images based on a first trained machine learning model; and
determine the target images by including images between two neighboring terminal images of the two or more terminal images in the plurality of successive images.

9. The system of claim 7 , wherein to determine the target images using machine learning, the at least one processor is further directed to cause the system to:
determine diastolic images relating to cardiac diastole of the object's heart or systolic images relating to cardiac systole of the object's heart in the plurality of successive images using a second trained machine learning model;
determine two or more terminal images in the diastolic images or the systolic images; and
determine the target images by including images between two neighboring terminal images of the two or more terminal images in the plurality of successive images.

10. The system of claim 1 , wherein the plurality of successive images are generated based on MR data using compressed sensing.

11. A method for medical imaging implemented on a machine having one or more processors and one or more storage devices, the method comprising:
obtaining a plurality of successive images of a region of interest (ROI) including at least a portion of an object's heart, wherein:
the plurality of successive images are based on imaging data acquired from the ROI by a scanner without electrocardiography (ECG) gating, and
the plurality of successive images are related to one or more cardiac cycles of the object's heart;

automatically determining, in the plurality of successive images, target images that correspond to at least one of the one or more cardiac cycles of the object's heart; and
for at least one of the target images, determining corresponding time information based on a temporal resolution of the plurality of successive images.

12. The method of claim 11 , wherein the automatically determining, in the plurality of successive images, the target images that correspond to the at least one of the one or more cardiac cycles of the object's heart includes:
determining a reference image based on the plurality of successive images, the reference image relating to cardiac diastole of the object's heart;
determining correlations between the reference image and at least multiple of the plurality of successive images;
determining, in the plurality of successive images, diastolic images relating to the cardiac diastole based on the correlations;
determining two or more terminal images in the diastolic images; and
determining the target images by including images between two neighboring terminal images of the two or mor

Assignees

• Uih America Inc

Inventors

• Lyu Jingyuan
• Ding Yu
• Liu Qi
• Xu Jian

Classifications

• A61B5/0044 Primary

  for the heart · CPC title

• G06T7/0016 Primary

  involving temporal comparison · CPC title

• G06F18/24147

  Distances to closest patterns, e.g. nearest neighbour classification · CPC title

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title

• A61B6/503

  for diagnosis of the heart · CPC title