# US 8,571,278 B2 · System and methods for multi-object multi-surface segmentation > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 8,571,278 B2 - Document ID: US8571278B2 - Status: Granted - Kind: B2 - Application number: 12/291,803 - Filed: 2008-11-13 - Published: 2009-05-28 - Granted: 2013-10-29 - Earliest claimed: 2005-06-24 - Assignees: University of Iowa Research Foundation - Applicants: Milan Sonka; Xiaodong Wu; Yin Yin; Xiangmin Zhang - Inventors: Milan Sonka; Xiaodong Wu; Yin Yin; Xiangmin Zhang - Web page: https://patentgrep.com/patent/US8571278B2 ## CPC - G06T 7/11: Region-based segmentation - G06F 18/2323: based on graph theory, e.g. minimum spanning trees [MST] or graph cuts - G06T 7/12: Edge-based segmentation - G06V 10/457: {by analysing connectivity, e.g. edge linking, connected component analysis or slices} - G06V 10/7635: {based on graphs, e.g. graph cuts or spectral clustering} - G06T 2207/10072: Tomographic images - G06T 2207/30048: Heart; Cardiac - G06T 2207/30101: Blood vessel; Artery; Vein; Vascular - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A system and methods for the efficient segmentation of globally optimal surfaces representing object boundaries in volumetric datasets is provided. An optical surface detection system and methods are provided that are capable of simultaneously detecting multiple interacting surfaces in which the optimality is controlled by the cost functions designed for individual surfaces and by several geometric constraints defining the surface smoothness and interrelations. The present invention includes surface segmentation using a layered graph-theoretic approach that optimally segments multiple interacting surfaces of a single object using a pre-segmentation step, after which the segmentation of all desired surfaces of the object are performed simultaneously in a single optimization process. ## Claims ### Claim 1 A method for image segmentation comprising the steps of: representing n-dimensional image data as an m-dimensional structure to form an image representation using a graph-search approach, wherein said n-dimensional image data represents an image that depicts multiple objects and each of the multiple objects has one or more surfaces; forming one or more regions of a q-ary object interaction to form a specified q-ary inter-object relationship; calculating a cost functional related to the m-dimensional structure and the one or more regions of the q-ary object interaction; and segmenting the n-dimensional image data with one or more layer surfaces of multiple objects using the cost functional, the image representation, and the specified q-ary inter-object relationship to produce segmented layers, thereby simultaneously segmenting the image that depicts multiple objects. ### Claim 2 (depends on 1) The method of claim 1 further comprising the step of utilizing quantitative analysis of one or more segmented layers. ### Claim 3 (depends on 1) The method of claim 1, wherein the n-dimensional image data represents a medical image. ### Claim 4 (depends on 1) The method of claim 1, wherein the n-dimensional image data represents an orthopedic image. ### Claim 5 (depends on 1) The method of claim 1, wherein the n-dimensional image data represents a volumetric image. ### Claim 6 (depends on 1) The method of claim 1, wherein the cost functional includes cost terms that are obtained from a combination of one or more of the following feature term categories selected from the group comprising of image properties, global smoothness constraints of one or more boundary surfaces of the m-dimensional structure, local smoothness constraints of one or more boundary surfaces of the m-dimensional structure, object-independent smoothness constraints of one or more boundary surfaces of the m-dimensional structure, object-specific smoothness constraints of one or more boundary surfaces of the m-dimensional structure, q-ary object interactions, optimally selected features selection, training examples, regions of object interaction, expert-designed features, imaging modalities, and patient characteristics. ### Claim 7 (depends on 1) The method of claim 1, wherein said calculating step is performed in an optimal fashion. ### Claim 8 (depends on 1) The method of claim 1 further comprising the step of employing one or more graph image representations to describe the m-dimensional structure and q-ary object interaction. ### Claim 9 (depends on 8) The method of claim 8, wherein the one or more graph image representations enforces boundary surface smoothness constraints. ### Claim 10 (depends on 1) The method of claim 1 further comprising the step of identifying individual layer-separating surfaces. ### Claim 11 (depends on 1) The method of claim 1 further comprising the step of establishing correspondence between the surfaces of interacting objects to form individual object graph representations to enforce the q-ary object interactions. ### Claim 12 (depends on 11) The method of claim 11, wherein said corresponding step further comprises the step of calculating an approximating separating surface. ### Claim 13 (depends on 12) The method of claim 12, wherein the approximating separating surface is a medial sheet. ### Claim 14 (depends on 11) The method of claim 11, wherein said establishing step further comprises the step of modifying topology of one or more individual-object graph representations. ### Claim 15 (depends on 1) The method of claim 1, wherein the graph-search approach is an optimal multi-layer graph-search approach. ### Claim 16 (depends on 2) The method of claim 2, wherein said utilizing step includes one or more of the following characteristics selected from the group comprising of morphologic characteristics, textural characteristics, and functional characteristics. ### Claim 17 (depends on 16) The method of claim 16, wherein the characteristics represent properties of the one or more segmented layers. ### Claim 18 (depends on 16) The method of claim 16, wherein the characteristics represent properties of any portion or portions of the one or more segmented layers. ### Claim 19 (depends on 2) The method of claim 2, wherein the quantitative analysis includes one or more characteristics selected from the group comprising of morphologic characteristics, textural characteristics, and functional characteristics. ### Claim 20 (depends on 19) The method of claim 19, wherein the characteristics represent global or local properties of one or more selected from the group comprising of an individual object, an object part, or an object combination. ### Claim 21 (depends on 2) The method of claim 2, wherein said utilizing step further comprises the step of exporting the quantitative analysis as a report. ### Claim 22 (depends on 2) The method of claim 2, wherein said utilizing step further comprises the step of storing the quantitative analysis on a storage medium. ### Claim 23 (depends on 22) The method of claim 22, wherein the storage medium is one selected from the group comprising of paper and computer disk. ### Claim 24 (depends on 2) The method of claim 2, wherein said utilizing step is visualized in a form corresponding to the representation of the segmented layers of said segmenting step. ### Claim 25 (depends on 1) The method of claim 1, wherein the wherein the n-dimensional image data represents a pelvis image.