Systems and methods for plaque identification, plaque composition analysis, and plaque stability detection
AssigneeShanghai United Imaging Healthcare Co., Ltd.
InventorsZhenhuan Gong, Yufei Mao, Xiong Yang, Saisai Su
Life of this patent
Estimating the term…
Abstract
The embodiments of the present disclosure provides methods for processing a plaque implemented on at least one machine each of which has at least one processor and at least one storage device for. The method may include: obtaining a plurality of images corresponding to a target vessel; processing the plurality of images; and determining plaque information based on a processing result. The plaque information may include at least one of an identification result of a target plaque, plaque composition distribution, or a detection result of plaque stability.
Claims20
A method implemented on at least one machine each of which has at least one processor and at least one storage device for identifying a plaque, comprising:
- obtaining an image set corresponding to a target image, the image set including a plurality of images, the target image being one of the plurality of images;
- obtaining a histogram equalized image corresponding to each image of the plurality of images in the image set by performing histogram equalization processing on the each image;
- and
- obtaining a plaque identification result of the target image by inputting the plurality of images and/or the plurality of histogram equalized images corresponding to the plurality of images into a trained segmentation model, wherein the trained segmentation model includes a first plaque segmentation model, and the obtaining a plaque identification result of the target image by inputting the plurality of images and/or the plurality of histogram equalized images corresponding to the plurality of images into a trained segmentation model includes:
- obtaining a vessel wall mask image corresponding to the target image;
- and
- obtaining a first plaque identification result of the target image by inputting the vessel wall mask image, the plurality of images, and the plurality of histogram equalized images corresponding to the plurality of images into the first plaque segmentation model.
A plaque identification system, comprising:
- at least one processor;
- and
- executable instructions, wherein when executed by the at least one processor, the executable instructions direct the at least one processor to implement a method including:
- obtaining an image set corresponding to a target image, the image set including a plurality of images, the target image being one of the plurality of images;
- obtaining a histogram equalized image corresponding to each image of the plurality of images in the image set by performing histogram equalization processing on the each image;
- and
- obtaining a plaque identification result of the target image by inputting the plurality of images and/or the plurality of histogram equalized images corresponding to the plurality of images into a trained segmentation model, wherein the trained segmentation model includes a first plaque segmentation model, and the obtaining a plaque identification result of the target image by inputting the plurality of images and/or the plurality of histogram equalized images corresponding to the plurality of images into a trained segmentation model includes:
- obtaining a vessel wall mask image corresponding to the target image;
- and
- obtaining a first plaque identification result of the target image by inputting the vessel wall mask image, the plurality of images, and the plurality of histogram equalized images corresponding to the plurality of images into the first plaque segmentation model.
A non-transitory computer-readable storage medium storing computer instructions, wherein when executed by at least one processor, the executable instructions direct the at least one processor to implement a method including:
- obtaining an image set corresponding to a target image, the image set including a plurality of images, the target image being one of the plurality of images;
- obtaining a histogram equalized image corresponding to each image of the plurality of images in the image set by performing histogram equalization processing on the each image;
- and
- obtaining a plaque identification result of the target image by inputting the plurality of images and/or the plurality of histogram equalized images corresponding to the plurality of images into a trained segmentation model, wherein the trained segmentation model includes a first plaque segmentation model, and the obtaining a plaque identification result of the target image by inputting the plurality of images and/or the plurality of histogram equalized images corresponding to the plurality of images into a trained segmentation model includes:
- obtaining a vessel wall mask image corresponding to the target image;
- and
- obtaining a first plaque identification result of the target image by inputting the vessel wall mask image, the plurality of images, and the plurality of histogram equalized images corresponding to the plurality of images into the first plaque segmentation model.
Dependencies are inferred from the claim text.