VOLUMETRIC ANALYSIS OF PATHOLOGIES

• US 20140119624A1
• Filed: 03/26/2013
• Published: 05/01/2014
• Est. Priority Date: 03/26/2012
• Status: Active Grant

• Current Assignee: Cleveland Clinic Foundation
• Original Assignee: Cleveland Clinic Foundation
• Inventors: Ehlers, Justis P., Dupps, William J., Srivastava, Sunil K., Xu, David, Kaiser, Peter K.
• Granted Patent: US 9,299,139 B2
• Time in Patent Office: Days
• US Class Current: 382/131
• CPC Class Codes: A61B 5/0066 Optical coherence imaging
• CPC Class Codes: G06T 17/20 Finite element generation, e.g. wire-frame surface description, tesselation
• CPC Class Codes: G06T 2207/10101 Optical tomography; Optical coherence tomography [OCT]
• CPC Class Codes: G06T 2207/20016 Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
• CPC Class Codes: G06T 2207/20072 Graph-based image processing
• CPC Class Codes: G06T 2207/30041 Eye; Retina; Ophthalmic
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/0016 involving temporal comparison
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/12 Edge-based segmentation
• CPC Class Codes: G06T 7/162 involving graph-based methods
• CPC Class Codes: G06T 7/62 of area, perimeter, diameter or volume

Abstract

Systems and methods are provided for volumetric analysis of pathologies. A segmentation component is configured to determine, for each of a series of images of a region of interest containing a pathological feature, a set of segmentation boundaries within the image representing a cross-section of the pathological feature. A mesh generation component is configured to link the sets of segmentation boundaries from adjacent images in the series of images to generate a polygonal mesh representing a volumetric reconstruction of the pathological feature. A volumetric measurement component is configured to calculate volumetric parameters from the volumetric reconstruction representing the pathological feature. A user interface is configured to provide the calculated volumetric parameters to an associated display.

Claims

1. A non-transitory computer readable medium storing executable instructions comprising:

a coarse segmentation component configured to determine, for each of a series of B-scan images, a set of constraint boundaries, based on natural contours within a region of tissue, from an optical coherence tomography (OCT) B-scan image and truncate all image area outside of the constraint boundaries to provide a truncated image;

a fine segmentation component configured to determine, for each truncated image, a set of pathology boundaries within the truncated image representing a cross-section of the pathological feature;

a mesh generation component configured to link the sets of pathology boundaries from adjacent images in the series of B-scan images to generate a polygonal mesh representing a volumetric reconstruction of the pathological feature; and

a user interface configured to provide the polygonal mesh to an associated display.

2. The non-transitory computer readable medium of claim 1, further comprising a volumetric measurement component configured to calculate volumetric parameters from the volumetric reconstruction representing the pathological feature.

3. The non-transitory computer readable medium of claim 2, further comprising a predictive model configured to predict a clinical outcome for a patient according to a plurality of predictors, the plurality of predictors including at least one volumetric parameter.

4. The non-transitory computer readable medium of claim 2, further comprising a predictive model configured to predict a clinical outcome for a patient according to a plurality of predictors, the plurality of predictors including at least one predictor that is a function of a first volumetric parameter, derived from an OCT scan taken prior to a start of a therapeutic intervention, and a second volumetric parameter, derived from an OCT scan taken after the start of a therapeutic intervention.

5. The non-transitory computer readable medium of claim 1, wherein the fine segmentation component is configured to determine the set of pathology boundaries by defining an undirected graph having vertices representing image pixels and edges between the vertices having respective weights defined from a set of at least one image characteristic and find an optimal segmentation for the defined graph.

6. The non-transitory computer readable medium of claim 5, wherein the set of at least one image characteristic is a first set of at least one image characteristic and the fine segmentation component is configured to define the undirected graph such that each pixel has defined edges to each of a source and a sink vertex, the edges having respective weights based on a second set of at least one image characteristic and find the optimal segmentation for the defined graph as an optimal solution to a maximum-flow, minimum-cut problem defined by the undirected graph.

7. The non-transitory computer readable medium of claim 6, the second set of at least one image characteristic comprising a set of pixel intensity values and histogram similarity metrics representing regions of the image.

8. The non-transitory computer readable medium of claim 5, the set of at least one image characteristic for a given edge comprising a gradient between pixels linked by the edge and a neighborhood similarity metric representing the similarity of pixels connected to a given pixel associated with the edge.

9. A non-transitory computer readable medium storing executable instructions comprising:

a segmentation component configured to determine, for each of a series of images of a region of interest containing a pathological feature, a set of segmentation boundaries within the image representing a cross-section of the pathological feature;

a mesh generation component configured to link the sets of segmentation boundaries from adjacent images in the series of images to generate a polygonal mesh representing a volumetric reconstruction of the pathological feature;

a volumetric measurement component configured to calculate volumetric parameters from the volumetric reconstruction representing the pathological feature; and

a user interface configured to provide the calculated volumetric parameters to an associated display.

10. The non-transitory computer readable medium of claim 9, further comprising a predictive model configured to predict a clinical outcome for a patient according to a plurality of predictors, the plurality of predictors including at least one of the calculated volumetric parameters.

11. The non-transitory computer readable medium of claim 9, further comprising a predictive model configured to predict a clinical outcome for a patient according to a plurality of predictors, the plurality of predictors including at least one predictor that is a function of a first volumetric parameter, derived from a set of images taken prior to a start of a therapeutic intervention, and a second volumetric parameter, derived from a set of images taken after the start of a therapeutic intervention.

12. The non-transitory computer readable medium of claim 9, the segmentation component comprising:

a coarse segmentation component configured to determine, for each of the series of images, a set of constraint boundaries, based on natural contours within a region of eye tissue, from an optical coherence tomography (OCT) B-scan image and truncate all image area outside of the constraint boundaries to provide a truncated image; and

a fine segmentation component configured to determine, for each truncated image, the set of segmentation boundaries representing a cross-section of the pathological feature.

13. A system comprising:

an OCT imager to provide the a series of images of the region of interest;

the non-transitory computer readable medium of claim 9;

a processor operatively connected to the non-transitory computer readable medium; and

the display associated with the user interface.

14. A method comprising:

imaging a region of interest containing a pathological feature with an associated imager prior to a start of a therapeutic intervention to provide a set of pre-intervention scan data;

calculating a first value of a volumetric parameter from the set of pre-intervention scan data;

imaging the region of interest with the imager after the start of the therapeutic intervention to provide a set of post-intervention scan data;

calculating a second value of the volumetric parameter from the set of post-intervention scan data;

predicting a clinical outcome associated with the therapeutic intervention as at least a function of the first value of the volumetric parameter and the second value of the volumetric parameter; and

displaying the predicted clinical outcome to a user at an associated output device.

15. The method of claim 14, wherein imaging the region of interest comprises imaging a region of eye tissue using an OCT imager to provide a series of B-scan images.

16. The method of claim 14, wherein calculating a first value of a volumetric parameter from the set of pre-intervention scan data comprises:

determining, for each of a series of images in the pre-intervention scan data, a set of segmentation boundaries within each image representing a cross-section of the pathological feature;

linking the sets of segmentation boundaries from adjacent images in the series of images to generate a polygonal mesh representing a volumetric reconstruction of the pathological feature; and

calculating the first value of the volumetric parameter from the volumetric reconstruction representing the pathological feature.

17. The method of claim 14, wherein the volumetric parameter can include one of a total volume, a base area, a top area, a maximal base width, and a minimum width of the volume representing the pathological feature.

18. The method of claim 14, wherein predicting a clinical outcome associated with the therapeutic intervention comprises providing a plurality of predictors to a predictive model, the plurality of predictors including one of the function of the first value of the volumetric parameter and the second value of the volumetric parameter and a set including both of the first value of the volumetric parameter and the second value of the volumetric parameter.

19. The method of claim 18, the plurality of predictors comprising at least one biometric parameter representing the patient that is not derived from either of the set of pre-intervention scan data and the set of post-intervention scan data.

20. The method of claim 14, wherein the volumetric parameter represents one of a volume and a basal diameter and the predicted clinical outcome is a minimum period of time for which it is necessary for the patient to remain in a prone position after the therapeutic intervention.