Medical image analysis using mechanical deformation information
AssigneeWake Forest University Health Sciences
InventorsJeongchul Kim, Youngkyoo Jung, Christopher T. Whitlow
Life of this patent
Estimating the term…
Abstract
Post-image acquisition methods, circuits and systems for evaluating medical images of a subject register a region of interest in a first medical image taken at a first point in time to the region of interest in a second image taken before or after the first medical image with voxels from the first and second medical images having a voxel-wise correspondence. The methods, circuits and systems can use line and/or shape changes of defined 3-D finite elements to electronically determine directional, shear and volumetric changes of the voxels in the region of interest between the first and second medical images.
Claims35
A post-image acquisition method of evaluating medical images of a subject, comprising:
- registering a region of interest in a first medical image taken at a first point in time to the region of interest (ROI) in a second image taken before or after the first medical image with voxels from the first and second medical images having a voxel-wise correspondence;
- electronically defining a plurality of three-dimensional (3-D) finite elements in the ROI of the first image, each 3-D finite element defined by connecting lines to points associated with at least four voxels to establish a shape of each 3-D finite element in the ROI of the first image with associated original lengths and lines;
- electronically deforming the 3-D finite elements in the ROI of the second medical image;
- then
- electronically calculating directional, shear and volume changes of the voxels in the ROI between the first and second images based on changes in lengths and angles of the lines in the deformed 3-D finite elements relative to the original lengths and angles.
A method of evaluating longitudinal medical images of a subject, comprising:
- electronically defining three dimensional (3-D) finite elements in a region of interest in a first medical image of a subject, wherein the 3-D finite elements are each associated with at least four neighboring voxels, and wherein a shape of the 3-D finite elements is based on lines connecting points of positions of the at least four voxels;
- electronically identifying or tracking the 3-D finite elements in a second medical image of the region of interest of the subject;
- electronically calculating a change in dimensions of the lines connecting the at least four voxels and/or a change in the shape of the 3-D finite elements in the region of interest in the second medical image;
- and then
- electronically calculating a plurality of morphometric parameters that identify directional and/or shear change in positions of voxels in the region of interest.
An image analysis system comprising:
- a circuit in communication with at least one server and/or at least one processor that:
- registers a region of interest in a first medical image taken at a first point in time to the region of interest (ROI) in a second image taken before or after the first medical image with voxels from the first and second medical images having a voxel-wise correspondence;
- defines a plurality of three-dimensional (3-D) finite elements in the ROI of the first image, each 3-D finite element defined by connecting lines to points associated with at least four voxels to establish a shape of the 3-D finite element in the ROI of the first image with associated original lengths and lines;
- deforms the 3-D finite elements in the ROI of the second medical image;
- then
- calculates directional, shear and volume changes of the voxels in the ROI between the first and second images based on changes in lengths and angles of the lines in the deformed 3-D finite elements relative to the original lengths and angles.
Dependencies are inferred from the claim text.