Device and method for modeling three-dimensional organ by image segmentation
The present disclosure relates to a method for three-dimensionally modeling an organ through image segmentation. The three-dimensional modeling of an organ includes the operations of: receiving one or more pieces of medical image data for a specific bodily organ of a target object; setting a region of interest with respect to the bodily organ based on the one or more pieces of medical image data; forming one or more blocks corresponding to the region of interest, wherein the blocks include a portion of the bodily organ corresponding to the regions of interest; setting a segment algorithm for each of the blocks; generating first image data respectively performing 3D modeling of portions contained in the blocks based on algorithms set to the blocks; and merging the first image data, and generating a three-dimensional section image data with respect to the entire bodily organ.
1. A method for three-dimensionally modeling an organ through image segmentation comprising:
receiving a plurality of pieces of medical image data for a specific bodily organ of a target object;
setting a region of interest with respect to the bodily organ based on the plurality of pieces of medical image data;
forming a plurality of three-dimensional (3D) cube-shaped blocks corresponding to the region of interest, wherein each of the plurality of 3D cube-shaped blocks includes a portion of the bodily organ corresponding to the regions of interest;
setting a respective segment algorithm for a respective 3D cube-shaped block of the plurality of 3D cube-shaped blocks;
generating 3D first image data by simultaneously performing a plurality of segmentation works for the plurality of 3D cube-shaped blocks with segment algorithms and respectively performing 3D modeling of portions contained in the plurality of 3D cube-shaped blocks; and
merging the 3D first image data, and generating 3D section image data with respect to the entire bodily organ.
2. The method according to claim 1 , wherein the setting the region of interest comprises:
recognizing types of the bodily organ based on a deep learning-based first model, and automatically generating the region of interest corresponding to the recognized types of the bodily organ.
3. The method according to claim 1 , wherein the merging comprises:
merging the 3D first image data through image registration performed through at least one among a feature element registration and a template-based registration.
4. The method according to claim 1 , wherein the generating the 3D first image data comprises:
discriminating a body portion contained in an overlapping region of two or more overlapping blocks among the plurality of 3D cube-shaped blocks based on the segmentation algorithms set to the two or more overlapping blocks.
5. The method according to claim 1 , wherein the setting the regions of interest comprises:
segmenting the region of interest into a first region corresponding to a blood area of the bodily organ and a second region corresponding to a muscle area of the bodily organ;
setting the first region and the second region as regions of interest, and
wherein the segmenting includes
identifying, based on the plurality of pieces of medical image data, the first region when a contrast value of the bodily organ of the first region is a predetermined first value or more, and the second region when a contrast value of the second region is a predetermined second value or less.
6. An apparatus for three-dimensionally modeling an organ through image segmentation comprising:
a communication unit for acquiring a plurality of pieces of medical image data for a specific bodily organ of a target object; and
a processor,
wherein the processor sets a region of interest with respect to the bodily organ based on the plurality of pieces of medical image data, forms a plurality of three-dimensional (3D) cube-shaped blocks corresponding to the region of interest, wherein each of the plurality of 3D cube-shaped blocks includes a portion of the bodily organ corresponding to the regions of interest, sets a respective segment algorithm for a respective 3D cube-shaped block of the plurality of 3D cube-shaped blocks; generates 3D first image data by simultaneously performing a plurality of segmentation works for the plurality of 3D cube-shaped blocks with segment algorithms and respectively performing 3D modeling of portions contained in the plurality of 3D cube-shaped blocks; and merges the 3D first image data, and generates 3D section image data with respect to the entire bodily organ.
7. The apparatus according to claim 6 , wherein the processor is further configured to:
discriminate a body portion contained in an overlapping region of two or more overlapping blocks among the plurality of 3D cube-shaped blocks based on the segmentation algorithms set to the two or more overlapping blocks.
8. The apparatus according to claim 6 , wherein the processor is further configured to:
recognize types of the bodily organ based on a deep learning-based first model, and automatically generate the region of interest corresponding to the recognized types of the bodily organ.
9. The apparatus according to claim 6 , wherein the processor is further configured to:
merge the 3D first image data through image registration performed through at least one among a feature element registration and a template-based registration.
10. A non-transitory computer-readable recording medium in which a program for executing the three-dimensional modeling method of an organ through image segmentation according to claim 1 is stored.