Method and system for selecting a region of interest in an image
Inventors
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-11769253-B2
Publication Date
2023-09-26
Expiration Date
Abstract
A system and computer-implemented method for analysing or monitoring a subject, the method comprising: identifying one or more objects of interest that have been segmented from a first image comprising any of a baseline scan, a follow-up scan or a reference image of the subject; identifying predefined landmarks of the objects; determining reference morphometries pertaining to the objects by performing morphometrics on the objects by reference to the landmarks; selecting one or more regions of interest (ROIs) from the objects according to the reference morphometries, comprising identifying respective locations of the ROIs relative to the reference morphometries; performing a first analysis of the one or more ROIs; performing at least one corresponding second analysis of the one or more ROIs in respectively at least one second image comprising any other of a baseline scan, a follow-up scan or a reference image; and generating a comparison of one or more results of the first analysis and one or more corresponding results of the second analysis.
Core Innovation
The computer-implemented method for analysing or monitoring a subject segments one or more objects of interest from a first image that comprises any of a baseline scan, a follow-up scan or a reference image, and identifies predefined landmarks of the objects. Reference morphometries pertaining to the objects are determined by performing morphometrics on the objects by reference to the landmarks, including applying one or more image processing algorithms to the first image to calculate basic morphometries. One or more regions of interest (ROIs) are then selected from the objects according to the reference morphometries by identifying respective locations of the ROIs relative to the reference morphometries.
The method performs a first analysis of the one or more ROIs and performs at least one corresponding second analysis of the one or more ROIs in respectively at least one second image comprising any other of a baseline scan, a follow-up scan or a reference image. A comparison is generated of one or more results of the first analysis and one or more corresponding results of the second analysis, and the comparison is output. The ROI selection is therefore tied to reference morphometries derived from landmark-referenced morphometrics to support consistent ROI placement across images.
In the system form, the invention uses an object and landmark identifier, a morphometer, and a region selector configured to implement the same pipeline: identifying segmented objects of interest from a first image, identifying predefined landmarks, determining reference morphometries using morphometrics relative to the landmarks with image processing algorithms that calculate basic morphometries, and selecting ROIs according to the reference morphometries. The system is configured to perform the first analysis and the corresponding second analysis in at least one second image, generate the comparison of the results, and output the comparison.
Claims Coverage
The document includes two independent claims that cover a computer-implemented method and a corresponding system for analysing or monitoring a subject. Across these independent claims, the core claim coverage includes: segmentation of objects, landmark identification, calculation of reference morphometries from basic morphometries using image processing algorithms, selection of ROIs according to the reference morphometries, performing first and second analyses in different images, and outputting a comparison of analysis results.
Segmented objects and predefined landmarks for reference morphometries
Identifying one or more objects of interest that have been segmented from a first image comprising any of a baseline scan, a follow-up scan or a reference image of the subject, and identifying predefined landmarks of the objects.
Landmark-referenced morphometrics using image processing algorithms to compute basic morphometries
Determining reference morphometries pertaining to the objects by performing morphometrics on the objects by reference to the landmarks, wherein determining the reference morphometries includes applying one or more image processing algorithms to the first image to calculate basic morphometries.
ROI selection based on reference morphometries
Selecting one or more regions of interest (ROIs) from the objects according to the reference morphometries, comprising identifying respective locations of the ROIs relative to the reference morphometries.
First and second analyses with comparison output across baseline/follow-up/reference images
Performing a first analysis of the one or more ROIs; performing at least one corresponding second analysis of the one or more ROIs in respectively at least one second image comprising any other of a baseline scan, a follow-up scan or a reference image; generating a comparison of one or more results of the first analysis and one or more corresponding results of the second analysis; and outputting the comparison.
Object and landmark identifier, morphometer, and region selector configured for morphometry-based ROI selection and comparative analysis
An object and landmark identifier configured to identify segmented objects of interest from a first image comprising any of a baseline scan, a follow-up scan or a reference image and to identify predefined landmarks of the objects; a morphometer configured to determine reference morphometries pertaining to the objects by performing morphometrics on the objects by reference to the landmarks, wherein determining the reference morphometries includes applying one or more image processing algorithms to the first image to calculate basic morphometries; and a region selector configured to select one or more regions of interest (ROIs) from the objects according to the reference morphometries, comprising identifying respective locations of the ROIs and determining respective shapes and sizes of the ROIs relative to the reference morphometries; wherein the system is configured to perform a first analysis, perform at least one corresponding second analysis in at least one second image comprising any other of a baseline scan, a follow-up scan or a reference image, generate a comparison of the results, and output the comparison.
Across both independent claims, the document centers on selecting ROIs whose locations, and in the system also shapes and sizes, are defined relative to reference morphometries computed from landmark-referenced morphometrics on a first image. The invention then analyses the selected ROIs in a first image context and corresponding ROIs in at least one second image, and outputs a comparison of analysis results.
Stated Advantages
Provides a comparison of analysis results between a first analysis of ROIs and corresponding second analysis of ROIs in a second image.
Enables ROI selection according to reference morphometries derived from predefined landmarks, supporting consistent ROI placement relative to morphometric reference.
Documented Applications
Worked example: femoral neck ROI from hip DXA.
Worked example: distal radius ROI from wrist CT.
Density-based analysis/comparisons across images using ROI density based on attenuation of a referent material.
Interested in licensing this patent?