Modeling regions of interest of an anatomic structure

A computer-implemented method comprising:

• receiving segmented image volume data for at least one anatomic structure, wherein the segmented image volume data includes a three-dimensional (3-D) image volume that includes the at least one anatomic structure;
• searching outward from a computed centerline for the at least one anatomic structure, the searching outward from the computed centerline comprising:
• implementing a first evaluation of the segmented image volume data to identify one or more portions of the at least one anatomic structure that has a thickness that is equal to or greater than a thickness threshold;
• implementing a second evaluation of the segmented image volume data based on the identified one or more portions of the at least one anatomic structure to detect transitions between an abnormal region of interest and a respective other portion of the at least one anatomic structure based on an evaluation of values of voxels representing the 3-D image volume that includes the at least one anatomic structure, wherein a first detected transition corresponds to a location on a surface of the abnormal region of interest and a second detected transition corresponds to a location on a surface of the respective portion, and wherein the abnormal region of interest comprises one of a calcified region, mineralized region, and plaque region, wherein the abnormal region of interest is part of at least one of the identified one or more portions of the at least one anatomic structure;
• isolating the abnormal region of interest of the at least one anatomic structure in response to the searching;
• and
• generating a region of interest model based on the abnormal region of interest.

A system comprising:

• a non-transitory computer-readable medium configured to store data comprising segmented image volume data for at least one anatomic structure and instructions, the segmented image volume data including a three-dimensional (3-D) image volume that includes the at least one anatomic structure;
• a processor configured to access the non-transitory computer-readable medium and execute the instructions, the instructions comprising a region of interest modeling engine, the region of interest modeling engine being programmed to:
• search outward from a computed centerline for the at least one anatomic structure, wherein the searching outward from the computed centerline comprising:
• implementing a first evaluation of the segmented image volume data to identify one or more portions of the at least one anatomic structure that has a thickness that is equal to or greater than a thickness threshold;
• implementing a second evaluation of the segmented image volume data based on the identified one or more portions of the at least one anatomic structure to detect a surface of one or more regions of interest of the at least one anatomic structure, the surface of the one or more regions of interest being detected based on an evaluation of-Hounsfield unit (HU) values associated with voxels representing the 3-D image volume that includes the at least one anatomic structure relative to a plurality of HU threshold ranges, each HU threshold range being associated with a density of a different material, wherein during the searching a first surface of a first region of interest of the one or more regions of interest is detected based on a first HU threshold range of the plurality of HU threshold ranges, and a second surface of a second region of interest of the one or more regions of interest is detected based on a second HU threshold range of the plurality of HU threshold ranges, the first region of interest having a different material density than the second region of interest, wherein the one or more regions of interest are part of at least one of the identified one or more portions of the at least one anatomic structure;
• isolate the first region of interest of the one or more regions of interest of the at least one anatomic structure in response to the searching by excluding 3-D image volumes for the first region of interest of the one or more regions of interest from the segmented image volume data for the at least one anatomic structure;
• and
• generate a region of interest model by applying a marching cubes algorithm to a voxel dataset for the first region of interest of the one or more regions of interest of the at least one anatomic structure.