Determination of feature boundaries in a digital representation of an anatomical structure
AssigneeThe United States of America as represented by the Department of Health and Human Services
InventorsJianhua Yao, Ronald M. Summers
Life of this patent
Estimating the term…
Abstract
A virtual anatomical structure can be analyzed to determine enclosing three-dimensional boundaries of features therein. Various techniques can be used to determine tissue types in the virtual anatomical structure. For example, tissue types can be determined via an iso-boundary between lumen and air in the virtual anatomical structure and a fuzzy clustering approach. Based on the tissue type determination, a deformable model approach can be used to determine an enclosing three-dimensional boundary of a feature in the virtual anatomical structure. The enclosing three-dimensional boundary can be used to determine characteristics of the feature and classify it as of interest or not of interest.
Claims46
One or more computer-readable media comprising computer-readable instructions for performing:
- receiving a digital representation of at least a portion of an anatomical structure, wherein the digital representation comprises a representation of at least one feature;
- determining an iso-boundary of the digital representation via an iso-value, wherein the iso-value comprises intensity, attenuation, adjusted measurement of intensity, or a combination thereof;
- classifying digital representation components of the digital representation as being of respective tissue types based at least on proximity of the components to the iso-boundary of the digital representation and a curvature type of the iso-boundary, wherein the iso-boundary is positioned between representations of lumen and wall in the digital representation;
- and
- based on the respective tissue types of the digital representation components, determining an enclosing three-dimensional boundary of the at least one feature within the digital representation.
A method of detecting polyps in a virtual colon, the method comprising:
- receiving an image of a portion of the virtual colon having a software-detected candidate polyp;
- locating an iso-boundary between lumen and wall in the image;
- classifying portions of the iso-boundary according to curvature type;
- adjusting intensities of components of the image based on respective proximities to portions of the iso-boundary and curvature types of the portions of the iso-boundary;
- categorizing the components as a tissue type of one of lumen, polyp, and non-polyp, resulting in tissue-categorized components;
- via a deformable model using one or more forces, converging to an enclosing three-dimensional boundary of the candidate polyp.
One or more computer-readable media comprising computer-executable instructions for performing a method comprising:
- receiving an image of a portion of the virtual colon having a software-detected candidate polyp;
- locating an iso-boundary between lumen and wall in the image;
- classifying portions of the iso-boundary according to curvature type;
- adjusting intensities of components of the image based on respective proximities to portions of the iso-boundary and curvature types of the portions of the iso-boundary;
- categorizing the components as a tissue type of one of lumen, polyp, and non-polyp, resulting in tissue-categorized components;
- and
- via a deformable model using one or more forces, converging to an enclosing three-dimensional boundary of the candidate polyp.
One or more computer-readable media comprising computer-executable instructions for performing:
- receiving a digital representation of at least a portion of an anatomical structure;
- based on curvature of an iso-boundary between representations of lumen and anatomical structure wall in the digital representation, categorizing at least a portion of the iso-boundary as a curvature type;
- and
- based at least on proximity of a digital representation component to the iso-boundary and the curvature type of the iso-boundary, adjusting intensity of the digital representation component, resulting in an adjusted intensity of the digital representation component;
- wherein the adjusting comprises:
- determining a score for the component based on proximity to the iso-boundary and a category of the iso-boundary;
- and
- adjusting intensity of the component based on the score.
One or more computer-readable media comprising computer-executable instructions for generating a software user interface comprising:
- a depiction of a feature of a virtual anatomical structure, wherein the depiction comprises components having software-adjusted intensities for image components proximate to an iso-boundary between lumen and air in the depiction, based at least on proximity of the image components to the iso-boundary and a curvature type of the iso-boundary.
A system for detecting a polyp in an anatomical structure, the system comprising:
- means for receiving a digital representation of at least a portion of the anatomical structure, wherein the digital representation comprises a representation of at least one feature;
- means for determining an iso-boundary of the digital representation via an iso-value means, wherein the iso-value means comprises intensity, adjusted measurement of intensity, attenuation, or a combination thereof;
- means for classifying digital representation components of the digital representation as being of respective tissue types based at least on proximity of the components to the iso-boundary of the digital representation and a curvature type of the iso-boundary, wherein the iso-boundary is positioned between representations of lumen and wall in the digital representation;
- means for determining an enclosing three-dimensional boundary of the at least one feature within the digital representation based on the respective tissue types of the digital representation components;
- means for calculating one or more characteristics based at least on the three-dimensional boundary;
- and
- means for classifying the feature as a polyp based at least on the one or more characteristics.
Dependencies are inferred from the claim text.