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-11170245-B2
Publication Date
2021-11-09
Expiration Date
Abstract
A computer-implemented method and system for selecting one or more regions of interest (ROIs) in an image. The method comprises: identifying one or more objects of interest that have been segmented from the image; identifying predefined landmarks of the objects; determining reference morphometrics pertaining to the objects by performing morphometrics on the objects by reference to the landmarks; selecting one or more ROIs from the objects according to the reference morphometrics, comprises identifying the location of the ROIs relative to the reference morphometrics; and outputting the selected one or more ROIs.
Core Innovation
The computer-implemented method selects one or more regions of interest (ROIs) in an image by identifying one or more objects of interest that are segmented from the image and by identifying predefined landmarks of the objects. The method determines reference morphometries pertaining to the objects by performing morphometrics on the objects by reference to the landmarks, including determining a shape, a size, or a shape and a size of the one or more objects. The method then selects the one or more ROIs from the objects according to the reference morphometries by identifying respective locations of the ROIs relative to the reference morphometries, and outputs the selected one or more ROIs.
The invention supports determining reference morphometries by using basic morphometries measured from the detected landmarks and determining the reference morphometries either directly from the measured basic morphometries or using one or more trained deep learning reference morphometric models. Selection of the ROIs includes determining respective shapes and sizes of the ROIs relative to the reference morphometries, so that ROI characterization is tied to object-specific reference morphometries.
Segmentation provides the segmented objects used for landmark detection and reference morphometric determination. The method can include image preprocessing using Gaussian blur and/or histogram equalization, and it can include coordinate transformation to normalize objects from an image coordinate system to an anatomical coordinate system prior to performing reference morphometrics.
Claims Coverage
The independent claims cover a computer-implemented method and a corresponding system that select one or more ROIs by anchoring ROI locations, and optionally ROI shapes and sizes, to object-specific reference morphometries derived from predefined landmarks. Across the independent claims, the core inventive features include object segmentation, landmark identification, reference morphometries via landmarks, and ROI location selection relative to the reference morphometries with ROI output.
Landmark-referenced reference morphometries from segmented objects
The reference morphometries pertaining to the objects are determined by performing morphometrics on the objects by reference to the landmarks, where determining the reference morphometries comprises determining a shape, a size, or a shape and a size of the one or more objects.
ROI location selection relative to reference morphometries
One or more ROIs are selected from the objects according to the reference morphometries, comprising identifying respective locations of the ROIs relative to the reference morphometries.
Optional ROI shape and size relative to reference morphometries
Selecting ROIs includes determining respective shapes and sizes of the ROIs relative to the reference morphometries.
Computer-implemented ROI output
The selected one or more ROIs are outputted.
Overall, claim coverage centers on computing reference morphometries of segmented objects using predefined landmarks and selecting ROI locations relative to those reference morphometries, with the invention further characterized by determining ROI shape and size relative to the reference morphometries and outputting the selected ROIs.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Selecting a femoral neck ROI from hip DXA imaging, including reference morphometric features and an example ROI dimension definition.
Selecting a distal radius ROI from wrist CT imaging, including reference morphometric features and an example ROI dimension definition.
Interested in licensing this patent?