US20120321155A1 - Method, apparatus, and program for measuring sizes of tumor regions - Google Patents
Method, apparatus, and program for measuring sizes of tumor regions Download PDFInfo
- Publication number
- US20120321155A1 US20120321155A1 US13/591,067 US201213591067A US2012321155A1 US 20120321155 A1 US20120321155 A1 US 20120321155A1 US 201213591067 A US201213591067 A US 201213591067A US 2012321155 A1 US2012321155 A1 US 2012321155A1
- Authority
- US
- United States
- Prior art keywords
- tumor region
- long axis
- designated
- axis
- designating
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10072—Tomographic images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30096—Tumor; Lesion
Definitions
- the present invention is related to a method, an apparatus, and a program for measuring the sizes of tumor regions that appear within three dimensional medical images.
- the malignancy of tumors have been evaluated by measuring the greatest diameter within a horizontal cross section (a plane that perpendicularly intersects the body axis of the tumor) of a tumor and using the measured value as an index, as disclosed in “New Guidelines to Evaluate the Response to Treatment in Solid Tumors”, P. Therasse et al., RECIST Guidelines, Journal of the National Cancer Institute, Vol. 92, No. 3, pp. 205-216, 2000.
- the indices of evaluation do not appropriately reflect the three dimensional sizes of the tumors, and there are cases in which accurate evaluations cannot be performed.
- the present invention has been developed in view of the foregoing circumstances. It is an object of the present invention to provide a method, an apparatus, and a computer program for measuring the sizes of tumor regions, which are capable of providing long axes and short axes that accurately reflect the three dimensional sizes of tumors.
- a tumor region size measuring method of the present invention comprises the steps of:
- the long axis and the short axis refer to a long axis and a short axis that perpendicularly intersect each other within the tumor region.
- the long axis and the short axis appropriately reflect the three dimensional size of a tumor, to be employed in accurate evaluation thereof.
- the setting of the arbitrary point is performed by setting a plurality of arbitrary points at the centers of gravity of a plurality of cross sections of the tumor region;
- the designation of the long axis is performed by obtaining a line segment having the greatest length from among a plurality of line segments that pass through each of the set arbitrary points and the ends of which are positioned on the outline of the tumor region, and designating the line segment having the greatest length from among the obtained line segments.
- the “line segment having the greatest length from among a plurality of line segments that pass through each of the set arbitrary points and the ends of which are positioned on the outline of the tumor region” does not necessarily refer to a line segment having the absolute greatest length, which is searched for from among line segments that pass through the arbitrary point is all directions.
- the “line segment having the greatest length from among a plurality of line segments that pass through each of the set arbitrary points and the ends of which are positioned on the outline of the tumor region” may be that which has the greatest length from among line segments that extend in a number of directions which have been thinned out to a degree that does not practically deteriorate the evaluation results of the tumor.
- the “cross section having the greatest area”, the “line segment . . . which has the greatest length”, and the phrase “two points . . . having the greatest distance therebetween” may be those which have been searched for among search ranges which have been thinned out to a degree that does not practically deteriorate the evaluation results of the tumor.
- the setting of the arbitrary point is performed by setting a plurality of arbitrary points at the centers of gravity of a plurality of cross sections of the tumor region;
- the designation of the long axis is performed by obtaining a line segment having the greatest length from among a plurality of line segments that pass through each of the set arbitrary points and the ends of which are positioned on the outline of the tumor region, and designating the line segment having the greatest length from among the obtained line segments.
- the designated long axis and the designated short axis may be displayed by a display means.
- the lengths of the designated long axis and the designated short axis may be displayed by the display means.
- both the designated long axis and the designated short axis and the lengths thereof may be displayed by the display means.
- the setting of the arbitrary point may be performed by setting the arbitrary point at the center of gravity of the tumor region.
- a tumor region size measuring apparatus of the present invention comprises:
- tumor region determining means for determining a tumor region from a three dimensional medical image
- axis designating means for designating a long axis and a short axis of the determined tumor region
- a tumor region size measuring program of the present invention causes a computer to execute the procedures of:
- program of the present invention may be provided being recorded on a computer readable medium.
- computer readable media are not limited to any specific type of device, and include, but are not limited to: floppy disks, CD's, RAM's, ROM's, hard disks, magnetic tapes, and Internet downloads, in which computer instructions can be stored and/or transmitted. Transmission of the computer instructions through a network or through wireless transmission means is also within the scope of the present invention. Additionally, computer instructions include, but are not limited to: source, object, and executable code, and can be in any language, including higher level languages, assembly language, and machine language.
- the method, apparatus, and program for measuring the sizes of tumor regions of the present invention determine a tumor region from within a three dimensional medical image, designate a long axis and a short axis of the determined tumor region, and obtain the lengths of the designated long axis and the designated short axis as the size of the tumor region. Thereby, the lengths of a long axis and a short axis that appropriately reflect the three dimensional size of a tumor can be provided.
- a configuration may be adopted wherein the designation of the long axis and the short axis comprises the steps of: setting an arbitrary point within the determined tumor region; designating a line segment having the greatest length from among a plurality of line segments that pass through the set arbitrary point and the ends of which are positioned on the outline of the tumor region, as the long axis; detecting a cross section of the tumor region having the greatest area from among cross sections of the tumor region that include the long axis; and designating a line segment having the greatest length from among a plurality of line segments within the detected cross section that perpendicularly intersect the designated long axis and the ends of which are positioned on the outline of the tumor region, as the short axis.
- a long axis and a short axis which are appropriate as indices for evaluating a tumor can be designated.
- the designation of the long axis and the short axis comprises the steps of: setting an arbitrary point within the determined tumor region;
- a cross section of the tumor region having the greatest area from among cross sections of the tumor region that pass through the point; designating a line segment having the greatest length from among a plurality of line segments within the detected cross section and the ends of which are positioned on the outline of the tumor region, as the long axis; and designating a line segment having the greatest length from among a plurality of line segments that perpendicularly intersect the designated long axis and the ends of which are positioned on the outline of the tumor region, as the short axis.
- a long axis and a short axis which are appropriate as indices for evaluating a tumor can be designated.
- the designation of the long axis and the short axis comprises the steps of: designating a line segment that connects two points on the outline of the tumor region having the greatest distance therebetween as the long axis; detecting a cross section of the tumor region having the greatest area from among cross sections of the tumor region that include the long axis; and designating a line segment having the greatest length from among a plurality of line segments within the detected cross section that perpendicularly intersect the designated long axis and the ends of which are positioned on the outline of the tumor region, as the short axis.
- a long axis and a short axis which are appropriate as indices for evaluating a tumor can be designated.
- a configuration may be adopted wherein the designation of the long axis and the short axis comprises the steps of: designating a line segment that connects two points on the outline of the tumor region having the greatest distance therebetween as the long axis; and designating a line segment having the greatest length from among a plurality of line segments that perpendicularly intersect the designated long axis and the ends of which are positioned on the outline of the tumor region, as the short axis.
- a long axis and a short axis which are appropriate as indices for evaluating a tumor can be designated.
- FIG. 1 is a block diagram that illustrates the configuration of a tumor region size measuring apparatus of the present invention.
- FIG. 2 is a diagram that illustrates the configuration of a tumor region determining section.
- FIG. 3 is a diagram for explaining the method by which a long axis is designated by a long axis/short axis designating section in a process according to a first embodiment of the present invention.
- FIG. 4 is a diagram for explaining the method by which a desired cross section is detected by a long axis/short axis designating section in a process according to the first embodiment of the present invention.
- FIG. 5 is a diagram for explaining the method by which a short axis is designated by a long axis/short axis designating section in a process according to the first embodiment of the present invention.
- FIG. 6 is a diagram for explaining the method by which a desired cross section is detected by a long axis/short axis designating section in a process according to a second embodiment of the present invention.
- FIG. 7 is a diagram for explaining the method by which a long axis is designated by a long axis/short axis designating section in a process according to a third embodiment of the present invention.
- FIG. 8 is a diagram for explaining the method by which a short axis is designated by a long axis/short axis designating section in a process according to a fourth embodiment of the present invention.
- FIG. 9 is a diagram that illustrates an example of a display screen which is displayed by a display section.
- the configuration of the tumor region size measuring apparatus 1 illustrated in FIG. 1 is realized by executing an image processing program, which is read out form an auxiliary memory device of a computer (such as a personal computer).
- the image processing program may be distributed being recorded in recording media such as CD-ROM's, or via a network such as the Internet, and installed in the computer.
- the tumor region size measuring apparatus is equipped with: a tumor region determining section 10 ; a long axis/short axis designating section 20 ; a length obtaining section 30 ; and a display section 40 .
- the tumor region determining section 10 determines tumor regions within three dimensional medical images.
- the long axis/short axis designating section 20 designates the long axis and the short axis of the determined tumor region.
- the length obtaining section 30 obtains the lengths of the designated long axis and the designated short axis as the size of the tumor region.
- the display section 40 displays the obtained lengths of the long axis and the short axis.
- the tumor region determining section 10 determines tumor regions from within three dimensional medical images (hereinafter, referred to as “input images”) which are groups of great numbers of two dimensional images obtained by CT apparatuses, MRI apparatuses, ultrasound apparatuses and the like. As illustrated in FIG. 2 , the tumor region determining section 10 is equipped with: an evaluating function obtaining section 11 ; a point setting section 12 ; a region setting section 13 ; a feature obtaining section 14 ; an evaluation value calculating section 15 ; a region dividing section 16 ; and the like.
- the evaluating function obtaining section 11 performs machine learning to learn the features of voxels of a plurality of three dimensional medical images (hereinafter, referred to as “sample images”) that include tumor regions of which the outlines are known, employing machine learning techniques such as the Adaboosting algorithm technique, the neural network technique, and the SVM (Support Vector Machine) technique. Thereby, the evaluating function obtaining section 11 obtains evaluation functions that evaluate whether a voxel is a voxel that represents an outline of a tumor, based on the features thereof.
- each voxel brightness data within the vicinities of each voxel, for example, combinations of brightness values of voxels within a cube shaped region 5 pixels long in the X direction, 5 pixels long in the Y direction, and 5 pixels long in the Z direction, having the voxel at its center, are learned as the feature of each of the voxels.
- Evaluation functions are obtained that evaluate whether each voxel is a voxel that represents an outline of a tumor based on the features thereof.
- the obtained evaluation functions can be applied to cases in which each voxel of a three dimensional medical image is a voxel that represents an outline of a tumor.
- the point setting section 12 sets arbitrary points within tumor regions in input images.
- the point setting section 12 may set arbitrary points at positions within input images specified by operator input using a keyboard or a pointing device of the tumor region size measuring apparatus 1 .
- each point within tumor regions which are automatically detected by the target region detecting method disclosed in Japanese Unexamined Patent Publication No. 8(1996)-299341, has a uniform mass, and the arbitrary points may be set at the centers of gravity of the tumor regions.
- the arbitrary points maybe set at the approximate centers of tumor regions, or at positions shifted from the centers of the tumor regions.
- the region setting section 13 sets judgment regions, which are considered to include the entireties tumor regions, within the input images, with the points set by the point setting section 12 as references. For example, regions of sizes greater than possible sizes of tumor regions having the points set by the point setting section 12 at their centers may be set as the judgment regions. Thereby, the range of regions of interest within the input images can be limited, and the following processes can be accelerated.
- perimeters of the judgment regions may be various shapes, such as cubes, spheres, and ellipsoids.
- the feature obtaining section 14 obtains the features of each voxel within the judgment regions sets by the region setting section 13 .
- the aforementioned brightness data within the vicinities of each voxel are obtained as the feature of each of the voxels.
- the evaluation value calculating section 15 calculates evaluation values regarding whether each voxel within the judgment regions is a voxel representing the outline of a tumor, based on the features obtained by the feature obtaining section 14 and employing the evaluating function obtained by the evaluating function obtaining section 11 .
- the region dividing section 16 determines the outlines of tumor regions, employing the evaluation values for each voxel calculated by the evaluation value calculating section 15 .
- the Graph Cut method described in “Interactive Graph Cuts for Optimal Boundary and Region Segmentation of Objects in N-D Images”, Y. Y. Boykov and M. P. Jolly, Proceedings of Internation Conference on Computer Vision, Vol. I, pp. 105-112, Vancouver, Canada, July 2001, and U.S. Pat. No. 6,973,212 may be employed to divide tumor regions from background regions.
- the probabilities that adjacent voxels are voxels included in the same region are calculated using the evaluation value for each voxel.
- the judgment regions are divided into tumor regions and background regions, based on the calculated probabilities.
- tumor regions V can be determined within the three dimensional medical images, and the boundaries between the tumor regions V and the background regions are the outlines of the tumor regions V.
- the long axis/short axis designating section 20 designate the long axes and the short axes of the tumor regions V, which have been determined by the tumor region determining section 10 .
- an arbitrary point Q is set within a tumor region V.
- each point within the tumor region V has a uniform mass
- the arbitrary point Q may be set at the center of gravity of the tumor region V.
- the arbitrary point Q may be a position within the input image specified by operator input using a keyboard or a pointing device of the tumor region size measuring apparatus 1 .
- a line segment having the greatest length from among a plurality of line segments that pass through the set arbitrary point Q and the ends of which are positioned on the outline of the tumor region V is designated as the long axis.
- a three dimensional coordinate system is set having the arbitrary point Q as its origin, as illustrated in FIG. 3 , and directions ( ⁇ , ⁇ ) (0° ⁇ 360°, 0° ⁇ 180°) emanating from the point Q are set.
- ⁇ represents angles in a direction parallel to the X axis
- ⁇ represents angles in a direction perpendicular to the Z axis.
- the directions ( ⁇ , ⁇ ) are varied, by maintaining ⁇ at 0° and sequentially changing ⁇ from 0°, 15°, 30° . .
- the line segment which has the greatest length from among line segments that extend from the arbitrary point Q in directions parallel to the X axis and perpendicular to the Z axis which have been thinned out in 15 degree increments is designated as the long axis La.
- the increments by which the directions are thinned out may be any amount that does not practically deteriorate the evaluation results of the tumor, and may be set arbitrarily as necessary.
- line segments that extend through the arbitrary point Q in all directions may be searched thoroughly to detect a line segment having the greatest length, and the detected line segment may be designated as the long axis La.
- a cross section of the tumor region having the greatest area from among cross sections of the tumor region V that include the long axis La is detected.
- angles ⁇ (0° ⁇ 180°) are set in the counterclockwise direction from a reference point (an angular position designated as a reference) with the long axis La as the central axis, as illustrated in FIG. 4 .
- the angle ⁇ is sequentially varied at increments that do not practically deteriorate the evaluation results of the tumor, for example, increments of 15 degrees.
- the areas of cross sections R( ⁇ ) of the tumor region V at each of the angles ⁇ that include the long axis La are obtained, and a cross section Ra having the greatest area from among the cross sections R( ⁇ ) is detected.
- the angle ⁇ may be varied in extremely fine increments within a range from 0° to 180°, to thoroughly search for the cross section Ra having the greatest area.
- a line segment having the greatest length from among a plurality of line segments within the detected cross section Ra that perpendicularly intersect the designated long axis La and the ends of which are positioned on the outline of the tumor region V is designated as a short axis.
- lines within the cross section Ra that extend perpendicular to the long axis La are sequentially set at increments d that do not practically deteriorate the evaluation results of the tumor, for example, 1 voxel increments, as illustrated in FIG. 5 .
- an arbitrary point Q is set within a tumor region V which has been determined by the tumor region determining section 10 in the same manner as in the first embodiment.
- a cross section of the tumor region V having the greatest area from among cross sections of the tumor region V that pass through the set arbitrary point Q is detected.
- a three dimensional coordinate system is set having the arbitrary point Q as its origin, as illustrated in FIG. 6 , and directions ( ⁇ , ⁇ ) (0° ⁇ 360°, 0° ⁇ 180°) emanating from the point Q are set.
- ⁇ represents angles in a direction parallel to the X axis
- ⁇ represents angles in a direction perpendicular to the Z axis.
- the directions ( ⁇ , ⁇ ) are varied, by maintaining ⁇ at 0° and sequentially changing ⁇ from 0°, 15°, 30° . . .
- the cross section Ra 2 having the greatest area is searched for while thinning out the directions that extend from the arbitrary point Q in directions parallel to the X axis and perpendicular to the Z axis in 15 degree increments.
- the increments by which the directions are thinned out may be any amount that does not practically deteriorate the evaluation results of the tumor, and may be set arbitrarily as necessary.
- the cross section Ra having the greatest area may be searched for thoroughly in all directions from the arbitrary point Q.
- a line segment having the greatest length from among a plurality of line segments within the detected cross section Ra and the ends of which are positioned on the outline of the tumor region V is designated as a long axis La.
- different combinations of two points are selected sequentially, from among a plurality of points which are set on the outline of the cross section Ra 2 at predetermined intervals. The distances between the combinations of the two points are obtained, and the line segment having the combination of points having the greatest distance therebetween as its ends is designated as the long axis La.
- a line segment having the greatest length from among a plurality of line segments that perpendicularly intersect the designated long axis La and the ends of which are positioned on the outline of the tumor region V, is designated as a short axis Sa, in the same manner as in the first embodiment described above.
- a line segment that connects two points on the outline of a tumor region V having the greatest distance therebetween is designated as a long axis La.
- different combinations of two points are selected sequentially, from among a plurality of points P 1 (x 1 , y 1 ); P 2 (x 2 , y 2 ); . . . Pi (xi, yi), which are set on the outline of the tumor region V, as illustrated in FIG. 7 .
- the distances between the combinations of the two points are obtained, and the line segment having the combination of points having the greatest distance therebetween as its ends is designated as the long axis La.
- a cross section Ra 3 of the tumor region V having the greatest area from among cross sections of the tumor region V that include the long axis La is detected in the same manner as in the first embodiment.
- a line segment having the greatest length from among a plurality of line segments within the detected cross section Ra 3 that perpendicularly intersect the designated long axis La and the ends of which are positioned on the outline of the tumor region V is designated as a short axis Sa.
- a line segment that connects two points on the outline of a tumor region V having the greatest distance therebetween is designated as a long axis La, in the same manner as in the third embodiment.
- a line segment having the greatest length from among a plurality of line segments that perpendicularly intersect the designated long axis La and the ends of which are positioned on the outline of the tumor region V is designated as a short axis Sa.
- angles ⁇ (0° ⁇ 360°) are set in the counterclockwise direction from a reference point (an angular position designated as a reference) with the long axis La as the central axis, as illustrated in FIG. 8 .
- the angle ⁇ is sequentially varied at increments of 15 degrees, and line segments Si( ⁇ ) that extend in the directions of the angles ⁇ through positions set along the long axis La at intervals d (one voxel intervals, for example) and have two points that intersect with the outline of the tumor region V are set.
- the length of each line segment Si( ⁇ ) is obtained, and the line segment Si( ⁇ ) having the greatest length is designated as the short axis Sa.
- the increments by which the angles ⁇ are thinned out and the interval d may be any amount that does not practically deteriorate the evaluation results of the tumor, and may be set arbitrarily as necessary.
- the length obtaining section 30 obtains the lengths of the long axes La and the short axes Sa of tumor regions V, which have been designated by the long axis/short axis designating section 20 , as the sizes of the tumor regions.
- the lengths of the long axes La and the short axes Sa may be obtained by measurement. If the lengths of the long axes La and/or the short axes Sa measured during the designating process for the long axes La and the short axes Sa performed by the long axis/short axis designating section 20 , these known values may be obtained.
- the display section 40 displays the designated long axes La and the designated short axes Sa on the screen of a display device, such as a monitor.
- the display section 40 displays the lengths of the designated long axes La and the designated short axes Sa.
- the display section 40 displays both the designated long axes La and the designated short axes Sa and the lengths thereof.
- the positions and lengths of a long axis La and a short axis Sa may be displayed overlapped with an image of a cross sectional image 31 of a tumor region that includes the long axis La and the short axis Sa, as illustrated in FIG. 9 .
- a tumor region V is determined from within a three dimensional medical image, a long axis La and a short axis Sa of the determined tumor region V are designated, and the lengths of the designated long axis La and the designated short axis Sa are obtained as the size of the tumor region V.
- the lengths of a long axis La and a short axis Sa that appropriately reflect the three dimensional size of a tumor can be provided.
- the designation of the long axis La and the short axis Sa comprises the steps of: setting an arbitrary point Q within the determined tumor region V; designating a line segment having the greatest length from among a plurality of line segments that pass through the set arbitrary point Q and the ends of which are positioned on the outline of the tumor region V, as the long axis La; detecting a cross section Ra of the tumor region V having the greatest area from among cross sections of the tumor region V that include the long axis La; and designating a line segment having the greatest length from among a plurality of line segments within the detected cross section Ra that perpendicularly intersect the designated long axis La and the ends of which are positioned on the outline of the tumor region V, as the short axis Sa, a long axis La and a short axis Sa which are appropriate as indices for evaluating a tumor can be designated.
- the designation of the long axis La. and the short axis Sa comprises the steps of: setting an arbitrary point Q within the determined tumor region V; detecting a cross section Ra 2 of the tumor region V having the greatest area from among cross sections of the tumor region V that pass through the point Q; designating a line segment having the greatest length from among a plurality of line segments within the detected cross section Ra 2 and the ends of which are positioned on the outline of the tumor region V, as the long axis La; and designating a line segment having the greatest length from among a plurality of line segments that perpendicularly intersect the designated long axis La and the ends of which are positioned on the outline of the tumor region V, as the short axis Sa, a long axis La and a short axis Sa which are appropriate as indices for evaluating a tumor can be designated.
- the designation of the long axis La and the short axis Sa comprises the steps of: designating a line segment that connects two points on the outline of the tumor region V having the greatest distance therebetween as the long axis La; detecting a cross section Ra 3 of the tumor region having the greatest area from among cross sections of the tumor region V that include the long axis La; and designating a line segment having the greatest length from among a plurality of line segments within the detected cross section Ra 3 that perpendicularly intersect the designated long axis La and the ends of which are positioned on the outline of the tumor region V, as the short axis Sa, a long axis La and a short axis Sa which are appropriate as indices for evaluating a tumor can be designated.
- the designation of the long axis La and the short axis Sa comprises the steps of: designating a line segment that connects two points on the outline of the tumor region V having the greatest distance therebetween as the long axis La; and designating a line segment having the greatest length from among a plurality of line segments that perpendicularly intersect the designated long axis La and the ends of which are positioned on the outline of the tumor region V, as the short axis Sa, a long axis La and a short axis Sa which are appropriate as indices for evaluating a tumor can be designated.
- the long axis/short axis designating section 20 designated the long axis and the short axis based on a single arbitrary point which was set within the tumor region V.
- a plurality of arbitrary points may be set at the centers of gravity of a plurality of cross sections of the tumor region.
- the designation of the long axis is performed by obtaining a line segment which has the greatest length by the process for designating a long axis according to the first or second embodiments, for each set arbitrary point. Then, the line segment having the greatest length from among the obtained line segments is designated as the long axis La.
- a line segment having the greatest length from among a plurality of line segments that pass through the set arbitrary point and the ends of which are positioned on the outline of the tumor region V may be obtained for each set arbitrary point. Then, the line segment having the greatest length from among the obtained line segments may be designated as the long axis La.
- a cross section of the tumor region V having the greatest area from among cross sections of the tumor region V that pass through the set arbitrary point may be detected for each set arbitrary point. Then, the cross section having the greatest area from among the detected cross sections that pass through the set arbitrary point may be detected. Thereafter, a line segment having the greatest length from among a plurality of line segments within the detected cross section and the ends of which are positioned on the outline of the tumor region V, may be designated as the long axis La.
- the plurality of cross sections of the tumor region V may be any set of two or more cross sections that pass through the tumor region V, such as: cross sections, which are perpendicular to the body axis of the tumor region V, extracted at predetermined intervals along the body axis; cross sections, of which the directions are randomly set; and cross sections, of which the positions are randomly set.
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Geometry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Magnetic Resonance Imaging Apparatus (AREA)
- Ultra Sonic Daignosis Equipment (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
A tumor region is determined within a three dimensional medical image. A long axis and a short axis of the determined tumor region are designated. The lengths of the designated long axis and the designated short axis are measured. The measured lengths of the long axis and the short axis are displayed.
Description
- This application is a Continuation of co-pending application Ser. No. 12/354,529, filed on Jan. 15, 2009, which claims the priority of Japanese Application No. 2008-006424, filed on Jan. 16, 2008, the entire contents of which are hereby incorporated by reference and for which priority is claimed under 35 U.S.C. §120.
- The present invention is related to a method, an apparatus, and a program for measuring the sizes of tumor regions that appear within three dimensional medical images.
- Conventionally, the malignancy of tumors have been evaluated by measuring the greatest diameter within a horizontal cross section (a plane that perpendicularly intersects the body axis of the tumor) of a tumor and using the measured value as an index, as disclosed in “New Guidelines to Evaluate the Response to Treatment in Solid Tumors”, P. Therasse et al., RECIST Guidelines, Journal of the National Cancer Institute, Vol. 92, No. 3, pp. 205-216, 2000.
- However, in these two dimensional evaluation methods that employ the long axes (greatest diameters) and short axes of ovals that approximate the horizontal cross sections of tumors, the indices of evaluation do not appropriately reflect the three dimensional sizes of the tumors, and there are cases in which accurate evaluations cannot be performed.
- On the other hand, techniques for generating three dimensional images, which are groups of great numbers of two dimensional images obtained by CT apparatuses, MRI apparatuses, ultrasound apparatuses and the like, of subjects have emerged recently. Methods for obtaining the volumes of tumors from three dimensional medical images have been proposed. Japanese Unexamined Patent Publication No. 8 (1996)-299341, U.S. Patent Application Publication No. 20050129297, and Japanese Unexamined Patent Publication No. 2003-265475 disclose methods in which the volumes of tumors as indices are employed to judge the sizes of the tumors.
- However, it is a fact that the lengths of the long axes and short axes of horizontal cross sections which are perpendicular to the body axes of tumors are used as indices for evaluations thereof, according to guidelines such as those disclosed in the aforementioned RECIST Guidelines, in clinical practice. There is demand for long axes and short axes that more appropriately reflect the three dimensional sizes of tumors to be provided, not only the volumes provided by the methods disclosed in the aforementioned Japanese Unexamined Patent Publications.
- The present invention has been developed in view of the foregoing circumstances. It is an object of the present invention to provide a method, an apparatus, and a computer program for measuring the sizes of tumor regions, which are capable of providing long axes and short axes that accurately reflect the three dimensional sizes of tumors.
- A tumor region size measuring method of the present invention comprises the steps of:
- determining a tumor region from a three dimensional medical image;
- designating a long axis and a short axis of the determined tumor region; and
- obtaining the lengths of the designated long axis and the designated short axis as the size of the tumor region.
- Here, the long axis and the short axis refer to a long axis and a short axis that perpendicularly intersect each other within the tumor region. The long axis and the short axis appropriately reflect the three dimensional size of a tumor, to be employed in accurate evaluation thereof.
- In the above method, a configuration may be adopted wherein the designation of the long axis and the short axis comprises the steps of:
- setting an arbitrary point within the determined tumor region;
- designating a line segment having the greatest length from among a plurality of line segments that pass through the set arbitrary point and the ends of which are positioned on the outline of the tumor region, as the long axis;
- detecting a cross section of the tumor region having the greatest area from among cross sections of the tumor region that include the long axis; and
- designating a line segment having the greatest length from among a plurality of line segments within the detected cross section that perpendicularly intersect the designated long axis and the ends of which are positioned on the outline of the tumor region, as the short axis.
- Here, a configuration may be adopted, wherein:
- the setting of the arbitrary point is performed by setting a plurality of arbitrary points at the centers of gravity of a plurality of cross sections of the tumor region;
- the designation of the long axis is performed by obtaining a line segment having the greatest length from among a plurality of line segments that pass through each of the set arbitrary points and the ends of which are positioned on the outline of the tumor region, and designating the line segment having the greatest length from among the obtained line segments.
- The “line segment having the greatest length from among a plurality of line segments that pass through each of the set arbitrary points and the ends of which are positioned on the outline of the tumor region” does not necessarily refer to a line segment having the absolute greatest length, which is searched for from among line segments that pass through the arbitrary point is all directions. The “line segment having the greatest length from among a plurality of line segments that pass through each of the set arbitrary points and the ends of which are positioned on the outline of the tumor region” may be that which has the greatest length from among line segments that extend in a number of directions which have been thinned out to a degree that does not practically deteriorate the evaluation results of the tumor. Similarly, the “cross section having the greatest area”, the “line segment . . . which has the greatest length”, and the phrase “two points . . . having the greatest distance therebetween” do not necessarily refer to those which have been obtained by thoroughly searching the search ranges for each object in the present specification. The “cross section having the greatest area”, the “line segment . . . which has the greatest length”, and the phrase “two points . . . having the greatest distance therebetween” may be those which have been searched for among search ranges which have been thinned out to a degree that does not practically deteriorate the evaluation results of the tumor.
- Alternatively, a configuration may be adopted wherein the designation of the long axis and the short axis comprises the steps of:
- setting an arbitrary point within the determined tumor region;
- detecting a cross section of the tumor region having the greatest area from among cross sections of the tumor region that pass through the point;
- designating a line segment having the greatest length from among a plurality of line segments within the detected cross section and the ends of which are positioned on the outline of the tumor region, as the long axis; and
- designating a line segment having the greatest length from among a plurality of line segments that perpendicularly intersect the designated long axis and the ends of which are positioned on the outline of the tumor region, as the short axis.
- Here, a configuration may be adopted, wherein:
- the setting of the arbitrary point is performed by setting a plurality of arbitrary points at the centers of gravity of a plurality of cross sections of the tumor region; and
- the designation of the long axis is performed by obtaining a line segment having the greatest length from among a plurality of line segments that pass through each of the set arbitrary points and the ends of which are positioned on the outline of the tumor region, and designating the line segment having the greatest length from among the obtained line segments.
- As a further alternative, a configuration may be adopted wherein the designation of the long axis and the short axis comprises the steps of:
- designating a line segment that connects two points on the outline of the tumor region having the greatest distance therebetween as the long axis;
- detecting a cross section of the tumor region having the greatest area from among cross sections of the tumor region that include the long axis; and
- designating a line segment having the greatest length from among a plurality of line segments within the detected cross section that perpendicularly intersect the designated long axis and the ends of which are positioned on the outline of the tumor region, as the short axis.
- As a still further alternative, a configuration may be adopted wherein the designation of the long axis and the short axis comprises the steps of:
- designating a line segment that connects two points on the outline of the tumor region having the greatest distance therebetween as the long axis; and
- designating a line segment having the greatest length from among a plurality of line segments that perpendicularly intersect the designated long axis and the ends of which are positioned on the outline of the tumor region, as the short axis.
- In all of the above methods, the designated long axis and the designated short axis may be displayed by a display means. Alternatively, the lengths of the designated long axis and the designated short axis may be displayed by the display means. As a further alternative, both the designated long axis and the designated short axis and the lengths thereof may be displayed by the display means.
- In addition, the setting of the arbitrary point may be performed by setting the arbitrary point at the center of gravity of the tumor region.
- A tumor region size measuring apparatus of the present invention comprises:
- tumor region determining means for determining a tumor region from a three dimensional medical image;
- axis designating means for designating a long axis and a short axis of the determined tumor region; and
- obtaining means for obtaining the lengths of the designated long axis and the designated short axis as the size of the tumor region.
- A tumor region size measuring program of the present invention causes a computer to execute the procedures of:
- determining a tumor region from a three dimensional medical image;
- designating a long axis and a short axis of the determined tumor region; and
- obtaining the lengths of the designated long axis and the designated short axis as the size of the tumor region.
- Note that the program of the present invention may be provided being recorded on a computer readable medium. Those who are skilled in the art would know that computer readable media are not limited to any specific type of device, and include, but are not limited to: floppy disks, CD's, RAM's, ROM's, hard disks, magnetic tapes, and Internet downloads, in which computer instructions can be stored and/or transmitted. Transmission of the computer instructions through a network or through wireless transmission means is also within the scope of the present invention. Additionally, computer instructions include, but are not limited to: source, object, and executable code, and can be in any language, including higher level languages, assembly language, and machine language.
- The method, apparatus, and program for measuring the sizes of tumor regions of the present invention determine a tumor region from within a three dimensional medical image, designate a long axis and a short axis of the determined tumor region, and obtain the lengths of the designated long axis and the designated short axis as the size of the tumor region. Thereby, the lengths of a long axis and a short axis that appropriately reflect the three dimensional size of a tumor can be provided.
- A configuration may be adopted wherein the designation of the long axis and the short axis comprises the steps of: setting an arbitrary point within the determined tumor region; designating a line segment having the greatest length from among a plurality of line segments that pass through the set arbitrary point and the ends of which are positioned on the outline of the tumor region, as the long axis; detecting a cross section of the tumor region having the greatest area from among cross sections of the tumor region that include the long axis; and designating a line segment having the greatest length from among a plurality of line segments within the detected cross section that perpendicularly intersect the designated long axis and the ends of which are positioned on the outline of the tumor region, as the short axis. In this case, a long axis and a short axis which are appropriate as indices for evaluating a tumor can be designated.
- Alternatively, a configuration may be adopted wherein the designation of the long axis and the short axis comprises the steps of: setting an arbitrary point within the determined tumor region;
- detecting a cross section of the tumor region having the greatest area from among cross sections of the tumor region that pass through the point; designating a line segment having the greatest length from among a plurality of line segments within the detected cross section and the ends of which are positioned on the outline of the tumor region, as the long axis; and designating a line segment having the greatest length from among a plurality of line segments that perpendicularly intersect the designated long axis and the ends of which are positioned on the outline of the tumor region, as the short axis. In this case as well, a long axis and a short axis which are appropriate as indices for evaluating a tumor can be designated.
- As a further alternative, a configuration may be adopted wherein the designation of the long axis and the short axis comprises the steps of: designating a line segment that connects two points on the outline of the tumor region having the greatest distance therebetween as the long axis; detecting a cross section of the tumor region having the greatest area from among cross sections of the tumor region that include the long axis; and designating a line segment having the greatest length from among a plurality of line segments within the detected cross section that perpendicularly intersect the designated long axis and the ends of which are positioned on the outline of the tumor region, as the short axis. In this case as well, a long axis and a short axis which are appropriate as indices for evaluating a tumor can be designated.
- As a still further alternative, a configuration may be adopted wherein the designation of the long axis and the short axis comprises the steps of: designating a line segment that connects two points on the outline of the tumor region having the greatest distance therebetween as the long axis; and designating a line segment having the greatest length from among a plurality of line segments that perpendicularly intersect the designated long axis and the ends of which are positioned on the outline of the tumor region, as the short axis. In this case as well, a long axis and a short axis which are appropriate as indices for evaluating a tumor can be designated.
-
FIG. 1 is a block diagram that illustrates the configuration of a tumor region size measuring apparatus of the present invention. -
FIG. 2 is a diagram that illustrates the configuration of a tumor region determining section. -
FIG. 3 is a diagram for explaining the method by which a long axis is designated by a long axis/short axis designating section in a process according to a first embodiment of the present invention. -
FIG. 4 is a diagram for explaining the method by which a desired cross section is detected by a long axis/short axis designating section in a process according to the first embodiment of the present invention. -
FIG. 5 is a diagram for explaining the method by which a short axis is designated by a long axis/short axis designating section in a process according to the first embodiment of the present invention. -
FIG. 6 is a diagram for explaining the method by which a desired cross section is detected by a long axis/short axis designating section in a process according to a second embodiment of the present invention. -
FIG. 7 is a diagram for explaining the method by which a long axis is designated by a long axis/short axis designating section in a process according to a third embodiment of the present invention. -
FIG. 8 is a diagram for explaining the method by which a short axis is designated by a long axis/short axis designating section in a process according to a fourth embodiment of the present invention. -
FIG. 9 is a diagram that illustrates an example of a display screen which is displayed by a display section. - Hereinafter, an embodiment of the tumor region size measuring apparatus of the present invention will be described with reference to the attached drawings. Note that the configuration of the tumor region
size measuring apparatus 1 illustrated inFIG. 1 is realized by executing an image processing program, which is read out form an auxiliary memory device of a computer (such as a personal computer). At this time, the image processing program may be distributed being recorded in recording media such as CD-ROM's, or via a network such as the Internet, and installed in the computer. - As illustrated in
FIG. 1 , the tumor region size measuring apparatus is equipped with: a tumorregion determining section 10; a long axis/shortaxis designating section 20; alength obtaining section 30; and adisplay section 40. The tumorregion determining section 10 determines tumor regions within three dimensional medical images. The long axis/shortaxis designating section 20 designates the long axis and the short axis of the determined tumor region. Thelength obtaining section 30 obtains the lengths of the designated long axis and the designated short axis as the size of the tumor region. Thedisplay section 40 displays the obtained lengths of the long axis and the short axis. - The tumor
region determining section 10 determines tumor regions from within three dimensional medical images (hereinafter, referred to as “input images”) which are groups of great numbers of two dimensional images obtained by CT apparatuses, MRI apparatuses, ultrasound apparatuses and the like. As illustrated inFIG. 2 , the tumorregion determining section 10 is equipped with: an evaluatingfunction obtaining section 11; apoint setting section 12; aregion setting section 13; afeature obtaining section 14; an evaluationvalue calculating section 15; aregion dividing section 16; and the like. - The evaluating
function obtaining section 11 performs machine learning to learn the features of voxels of a plurality of three dimensional medical images (hereinafter, referred to as “sample images”) that include tumor regions of which the outlines are known, employing machine learning techniques such as the Adaboosting algorithm technique, the neural network technique, and the SVM (Support Vector Machine) technique. Thereby, the evaluatingfunction obtaining section 11 obtains evaluation functions that evaluate whether a voxel is a voxel that represents an outline of a tumor, based on the features thereof. - Specifically, brightness data within the vicinities of each voxel, for example, combinations of brightness values of voxels within a cube shaped region 5 pixels long in the X direction, 5 pixels long in the Y direction, and 5 pixels long in the Z direction, having the voxel at its center, are learned as the feature of each of the voxels. Evaluation functions are obtained that evaluate whether each voxel is a voxel that represents an outline of a tumor based on the features thereof. The obtained evaluation functions can be applied to cases in which each voxel of a three dimensional medical image is a voxel that represents an outline of a tumor.
- The
point setting section 12 sets arbitrary points within tumor regions in input images. For example, thepoint setting section 12 may set arbitrary points at positions within input images specified by operator input using a keyboard or a pointing device of the tumor regionsize measuring apparatus 1. Alternatively, it can be assumed that each point within tumor regions, which are automatically detected by the target region detecting method disclosed in Japanese Unexamined Patent Publication No. 8(1996)-299341, has a uniform mass, and the arbitrary points may be set at the centers of gravity of the tumor regions. - Note that the arbitrary points maybe set at the approximate centers of tumor regions, or at positions shifted from the centers of the tumor regions.
- The
region setting section 13 sets judgment regions, which are considered to include the entireties tumor regions, within the input images, with the points set by thepoint setting section 12 as references. For example, regions of sizes greater than possible sizes of tumor regions having the points set by thepoint setting section 12 at their centers may be set as the judgment regions. Thereby, the range of regions of interest within the input images can be limited, and the following processes can be accelerated. - Note the perimeters of the judgment regions may be various shapes, such as cubes, spheres, and ellipsoids.
- The
feature obtaining section 14 obtains the features of each voxel within the judgment regions sets by theregion setting section 13. The aforementioned brightness data within the vicinities of each voxel are obtained as the feature of each of the voxels. - The evaluation
value calculating section 15 calculates evaluation values regarding whether each voxel within the judgment regions is a voxel representing the outline of a tumor, based on the features obtained by thefeature obtaining section 14 and employing the evaluating function obtained by the evaluatingfunction obtaining section 11. - The
region dividing section 16 determines the outlines of tumor regions, employing the evaluation values for each voxel calculated by the evaluationvalue calculating section 15. For example, the Graph Cut method described in “Interactive Graph Cuts for Optimal Boundary and Region Segmentation of Objects in N-D Images”, Y. Y. Boykov and M. P. Jolly, Proceedings of Internation Conference on Computer Vision, Vol. I, pp. 105-112, Vancouver, Canada, July 2001, and U.S. Pat. No. 6,973,212 may be employed to divide tumor regions from background regions. At this time, the probabilities that adjacent voxels are voxels included in the same region are calculated using the evaluation value for each voxel. The judgment regions are divided into tumor regions and background regions, based on the calculated probabilities. Thereby, tumor regions V can be determined within the three dimensional medical images, and the boundaries between the tumor regions V and the background regions are the outlines of the tumor regions V. - The long axis/short
axis designating section 20 designate the long axes and the short axes of the tumor regions V, which have been determined by the tumorregion determining section 10. - A first embodiment of the process for designating a long axis and a short axis will be described with reference to
FIGS. 3 through 5 . First, an arbitrary point Q is set within a tumor region V. For example, it can be assumed that each point within the tumor region V has a uniform mass, and the arbitrary point Q may be set at the center of gravity of the tumor region V. Alternatively, the arbitrary point Q may be a position within the input image specified by operator input using a keyboard or a pointing device of the tumor regionsize measuring apparatus 1. - Next, a line segment having the greatest length from among a plurality of line segments that pass through the set arbitrary point Q and the ends of which are positioned on the outline of the tumor region V, is designated as the long axis. Specifically, a three dimensional coordinate system is set having the arbitrary point Q as its origin, as illustrated in
FIG. 3 , and directions (θ, φ) (0°≦θ≦360°, 0°≦φ≦180°) emanating from the point Q are set. Here, θ represents angles in a direction parallel to the X axis, and φ represents angles in a direction perpendicular to the Z axis. The directions (θ, φ) are varied, by maintaining θ at 0° and sequentially changing φ from 0°, 15°, 30° . . . to 180°; maintaining θ at 15° and sequentially changing φ from (0°), 15°, 30° . . . 165°, (180°); . . . and maintaining θ at 345° and sequentially changing φ from (0°), 15°, 30° . . . 165°, (180°). Then, the lengths of line segments L (θ, φ) that extend in the directions (θ, φ) and which have two points C1 and C2 that intersect with the outline of the tumor region V as its ends are measured. The line segment L (θ, φ) having the greatest length from among the line segments L (θ, φ) is designated as a long axis La. - Note that here, a case has been described in which the line segment which has the greatest length from among line segments that extend from the arbitrary point Q in directions parallel to the X axis and perpendicular to the Z axis which have been thinned out in 15 degree increments is designated as the long axis La. The increments by which the directions are thinned out may be any amount that does not practically deteriorate the evaluation results of the tumor, and may be set arbitrarily as necessary. Alternatively, line segments that extend through the arbitrary point Q in all directions may be searched thoroughly to detect a line segment having the greatest length, and the detected line segment may be designated as the long axis La.
- Next, a cross section of the tumor region having the greatest area from among cross sections of the tumor region V that include the long axis La is detected. Specifically, angles γ (0°≦γ≦180°) are set in the counterclockwise direction from a reference point (an angular position designated as a reference) with the long axis La as the central axis, as illustrated in
FIG. 4 . The angle γ is sequentially varied at increments that do not practically deteriorate the evaluation results of the tumor, for example, increments of 15 degrees. The areas of cross sections R(γ) of the tumor region V at each of the angles γ that include the long axis La are obtained, and a cross section Ra having the greatest area from among the cross sections R(γ) is detected. - Note that here, a case has been described in which the cross section Ra having the greatest area was detected while varying the angle γ at increments that do not practically deteriorate the evaluation results of the tumor. Alternatively, the angle γ may be varied in extremely fine increments within a range from 0° to 180°, to thoroughly search for the cross section Ra having the greatest area.
- Next, a line segment having the greatest length from among a plurality of line segments within the detected cross section Ra that perpendicularly intersect the designated long axis La and the ends of which are positioned on the outline of the tumor region V, is designated as a short axis. Specifically, lines within the cross section Ra that extend perpendicular to the long axis La are sequentially set at increments d that do not practically deteriorate the evaluation results of the tumor, for example, 1 voxel increments, as illustrated in
FIG. 5 . The lengths of line segments Si (i=1, 2, . . . ) having two points C3 and C4 that intersect with the outline of the tumor region V as its ends are measured, and the line segment having the greatest length among these is designated as a short axis Sa. - Note that here, a case has been described in which the line segment Si having the greatest length was detected while moving the lines along the long axis La at the increment d that does not practically deteriorate the evaluation results of the tumor. Alternatively, the lines may be moved extremely fine increments along the long axis La, to thoroughly search for line segment Si having the greatest length.
- Hereinafter, a second embodiment of the process for designating a long axis and a short axis will be described.
- First, an arbitrary point Q is set within a tumor region V which has been determined by the tumor
region determining section 10 in the same manner as in the first embodiment. - Next, a cross section of the tumor region V having the greatest area from among cross sections of the tumor region V that pass through the set arbitrary point Q is detected. Specifically, a three dimensional coordinate system is set having the arbitrary point Q as its origin, as illustrated in
FIG. 6 , and directions (θ, φ) (0°≦θ≦360°, 0°≦φ≦180°) emanating from the point Q are set. Here, θ represents angles in a direction parallel to the X axis, and φ represents angles in a direction perpendicular to the Z axis. The directions (θ, φ) are varied, by maintaining θ at 0° and sequentially changing φ from 0°, 15°, 30° . . . to 180°; maintaining θ at 15° and sequentially changing φ from (0°, 15°, 30° . . . 165°, (180°); . . . and maintaining θ at 345° and sequentially changing φ from (0°), 15°, 30° . . . 165°, (180°). Then, the areas of cross sections R(θ, φ) having the directions (θ, φ) as their normal lines and that pass through the origin Q are obtained, and a cross section Ra2 having the greatest area from among the cross sections R(θ, φ) is detected. - Note that here, a case has been described in which the cross section Ra2 having the greatest area is searched for while thinning out the directions that extend from the arbitrary point Q in directions parallel to the X axis and perpendicular to the Z axis in 15 degree increments. The increments by which the directions are thinned out may be any amount that does not practically deteriorate the evaluation results of the tumor, and may be set arbitrarily as necessary. Alternatively, the cross section Ra having the greatest area may be searched for thoroughly in all directions from the arbitrary point Q.
- Next, a line segment having the greatest length from among a plurality of line segments within the detected cross section Ra and the ends of which are positioned on the outline of the tumor region V, is designated as a long axis La. Specifically, different combinations of two points are selected sequentially, from among a plurality of points which are set on the outline of the cross section Ra2 at predetermined intervals. The distances between the combinations of the two points are obtained, and the line segment having the combination of points having the greatest distance therebetween as its ends is designated as the long axis La.
- Next, a line segment having the greatest length from among a plurality of line segments that perpendicularly intersect the designated long axis La and the ends of which are positioned on the outline of the tumor region V, is designated as a short axis Sa, in the same manner as in the first embodiment described above.
- Hereinafter, a third embodiment of the process for designating a long axis and a short axis will be described.
- First, a line segment that connects two points on the outline of a tumor region V having the greatest distance therebetween is designated as a long axis La. Specifically, different combinations of two points are selected sequentially, from among a plurality of points P1 (x1, y1); P2 (x2, y2); . . . Pi (xi, yi), which are set on the outline of the tumor region V, as illustrated in
FIG. 7 . The distances between the combinations of the two points are obtained, and the line segment having the combination of points having the greatest distance therebetween as its ends is designated as the long axis La. - Next, a cross section Ra3 of the tumor region V having the greatest area from among cross sections of the tumor region V that include the long axis La is detected in the same manner as in the first embodiment. Then, a line segment having the greatest length from among a plurality of line segments within the detected cross section Ra3 that perpendicularly intersect the designated long axis La and the ends of which are positioned on the outline of the tumor region V, is designated as a short axis Sa.
- Hereinafter, a fourth embodiment of the process for designating a long axis and a short axis will be described.
- First, a line segment that connects two points on the outline of a tumor region V having the greatest distance therebetween is designated as a long axis La, in the same manner as in the third embodiment.
- Next, a line segment having the greatest length from among a plurality of line segments that perpendicularly intersect the designated long axis La and the ends of which are positioned on the outline of the tumor region V, is designated as a short axis Sa. Specifically, angles γ (0°≦γ≦360°) are set in the counterclockwise direction from a reference point (an angular position designated as a reference) with the long axis La as the central axis, as illustrated in
FIG. 8 . The angle γ is sequentially varied at increments of 15 degrees, and line segments Si(γ) that extend in the directions of the angles γ through positions set along the long axis La at intervals d (one voxel intervals, for example) and have two points that intersect with the outline of the tumor region V are set. The length of each line segment Si(γ) is obtained, and the line segment Si(γ) having the greatest length is designated as the short axis Sa. Here, the increments by which the angles γ are thinned out and the interval d may be any amount that does not practically deteriorate the evaluation results of the tumor, and may be set arbitrarily as necessary. - The
length obtaining section 30 obtains the lengths of the long axes La and the short axes Sa of tumor regions V, which have been designated by the long axis/shortaxis designating section 20, as the sizes of the tumor regions. For example, the lengths of the long axes La and the short axes Sa may be obtained by measurement. If the lengths of the long axes La and/or the short axes Sa measured during the designating process for the long axes La and the short axes Sa performed by the long axis/shortaxis designating section 20, these known values may be obtained. - The
display section 40 displays the designated long axes La and the designated short axes Sa on the screen of a display device, such as a monitor. Alternatively, thedisplay section 40 displays the lengths of the designated long axes La and the designated short axes Sa. As a further alternative, thedisplay section 40 displays both the designated long axes La and the designated short axes Sa and the lengths thereof. For example, the positions and lengths of a long axis La and a short axis Sa may be displayed overlapped with an image of a crosssectional image 31 of a tumor region that includes the long axis La and the short axis Sa, as illustrated inFIG. 9 . - According to the embodiments described above, a tumor region V is determined from within a three dimensional medical image, a long axis La and a short axis Sa of the determined tumor region V are designated, and the lengths of the designated long axis La and the designated short axis Sa are obtained as the size of the tumor region V. Thereby, the lengths of a long axis La and a short axis Sa that appropriately reflect the three dimensional size of a tumor can be provided.
- In the case that the designation of the long axis La and the short axis Sa comprises the steps of: setting an arbitrary point Q within the determined tumor region V; designating a line segment having the greatest length from among a plurality of line segments that pass through the set arbitrary point Q and the ends of which are positioned on the outline of the tumor region V, as the long axis La; detecting a cross section Ra of the tumor region V having the greatest area from among cross sections of the tumor region V that include the long axis La; and designating a line segment having the greatest length from among a plurality of line segments within the detected cross section Ra that perpendicularly intersect the designated long axis La and the ends of which are positioned on the outline of the tumor region V, as the short axis Sa, a long axis La and a short axis Sa which are appropriate as indices for evaluating a tumor can be designated.
- In the case that the designation of the long axis La. and the short axis Sa comprises the steps of: setting an arbitrary point Q within the determined tumor region V; detecting a cross section Ra2 of the tumor region V having the greatest area from among cross sections of the tumor region V that pass through the point Q; designating a line segment having the greatest length from among a plurality of line segments within the detected cross section Ra2 and the ends of which are positioned on the outline of the tumor region V, as the long axis La; and designating a line segment having the greatest length from among a plurality of line segments that perpendicularly intersect the designated long axis La and the ends of which are positioned on the outline of the tumor region V, as the short axis Sa, a long axis La and a short axis Sa which are appropriate as indices for evaluating a tumor can be designated.
- In the case that the designation of the long axis La and the short axis Sa comprises the steps of: designating a line segment that connects two points on the outline of the tumor region V having the greatest distance therebetween as the long axis La; detecting a cross section Ra3 of the tumor region having the greatest area from among cross sections of the tumor region V that include the long axis La; and designating a line segment having the greatest length from among a plurality of line segments within the detected cross section Ra3 that perpendicularly intersect the designated long axis La and the ends of which are positioned on the outline of the tumor region V, as the short axis Sa, a long axis La and a short axis Sa which are appropriate as indices for evaluating a tumor can be designated.
- In the case that the designation of the long axis La and the short axis Sa comprises the steps of: designating a line segment that connects two points on the outline of the tumor region V having the greatest distance therebetween as the long axis La; and designating a line segment having the greatest length from among a plurality of line segments that perpendicularly intersect the designated long axis La and the ends of which are positioned on the outline of the tumor region V, as the short axis Sa, a long axis La and a short axis Sa which are appropriate as indices for evaluating a tumor can be designated.
- Note that in the first and second embodiments of the process for designating the long axis and the short axis described above, the long axis/short
axis designating section 20 designated the long axis and the short axis based on a single arbitrary point which was set within the tumor region V. Alternatively, a plurality of arbitrary points may be set at the centers of gravity of a plurality of cross sections of the tumor region. In this case, the designation of the long axis is performed by obtaining a line segment which has the greatest length by the process for designating a long axis according to the first or second embodiments, for each set arbitrary point. Then, the line segment having the greatest length from among the obtained line segments is designated as the long axis La. - For example, a line segment having the greatest length from among a plurality of line segments that pass through the set arbitrary point and the ends of which are positioned on the outline of the tumor region V, may be obtained for each set arbitrary point. Then, the line segment having the greatest length from among the obtained line segments may be designated as the long axis La. Alternatively, a cross section of the tumor region V having the greatest area from among cross sections of the tumor region V that pass through the set arbitrary point may be detected for each set arbitrary point. Then, the cross section having the greatest area from among the detected cross sections that pass through the set arbitrary point may be detected. Thereafter, a line segment having the greatest length from among a plurality of line segments within the detected cross section and the ends of which are positioned on the outline of the tumor region V, may be designated as the long axis La.
- Note that the plurality of cross sections of the tumor region V may be any set of two or more cross sections that pass through the tumor region V, such as: cross sections, which are perpendicular to the body axis of the tumor region V, extracted at predetermined intervals along the body axis; cross sections, of which the directions are randomly set; and cross sections, of which the positions are randomly set.
Claims (21)
1. A tumor region size measuring method comprising the steps of :
determining a tumor region, which is a three dimensional region that represents a tumor, from a three dimensional medical image;
designating a long axis and a short axis of the determined tumor region; and
obtaining the lengths of the designated long axis and the designated short axis as the size of the tumor region, wherein:
the designation of the long axis and the short axis comprises the steps of:
setting an arbitrary point within the determined tumor region;
designating a line segment having the greatest length from among a plurality of line segments that pass through the set arbitrary point and the ends of which are positioned on the outline of the tumor region, as the long axis;
detecting a cross section of the tumor region having the greatest area from among cross sections of the tumor region that include the long axis; and
designating a line segment having the greatest length from among a plurality of line segments within the detected cross section that perpendicularly intersect the designated long axis and the ends of which are positioned on the outline of the tumor region, as the short axis.
2. A tumor region size measuring method comprising the steps of:
determining a tumor region, which is a three dimensional region that represents a tumor, from a three dimensional medical image;
designating a long axis and a short axis of the determined tumor region; and
obtaining the lengths of the designated long axis and the designated short axis as the size of the tumor region, wherein:
the designation of the long axis and the short axis comprises the steps of:
setting an arbitrary point within the determined tumor region;
detecting a cross section of the tumor region having the greatest area from among cross sections of the tumor region that pass through the point;
designating a line segment having the greatest length from among a plurality of line segments within the detected cross section and the ends of which are positioned on the outline of the tumor region, as the long axis; and
designating a line segment having the greatest length from among a plurality of line segments that perpendicularly intersect the designated long axis and the ends of which are positioned on the outline of the tumor region, as the short axis.
3. A tumor region size measuring method comprising the steps of :
determining a tumor region, which is a three dimensional region that represents a tumor, from a three dimensional medical image;
designating a long axis and a short axis of the determined tumor region; and
obtaining the lengths of the designated long axis and the designated short axis as the size of the tumor region, wherein:
the designation of the long axis and the short axis comprises the steps of:
designating a line segment that connects two points on the outline of the tumor region having the greatest distance therebetween as the long axis;
detecting a cross section of the tumor region having the greatest area from among cross sections of the tumor region that include the long axis; and
designating a line segment having the greatest length from among a plurality of line segments within the detected cross section that perpendicularly intersect the designated long axis and the ends of which are positioned on the outline of the tumor region, as the short axis.
4. A tumor region size measuring method comprising the steps of:
determining a tumor region, which is a three dimensional region that represents a tumor, from a three dimensional medical image;
designating a long axis and a short axis of the determined tumor region; and
obtaining the lengths of the designated long axis and the designated short axis as the size of the tumor region, wherein:
the designation of the long axis and the short axis comprises the steps of:
designating a line segment that connects two points on the outline of the tumor region having the greatest distance therebetween as the long axis; and
designating a line segment having the greatest length from among a plurality of line segments that perpendicularly intersect the designated long axis and the ends of which are positioned on the outline of the tumor region, as the short axis.
5. A tumor region size measuring method as defined in claim 1 , wherein:
the designated long axis and the designated short axis are displayed on a display means.
6. A tumor region size measuring method as defined in claim 1 , wherein:
the lengths of the designated long axis and the designated short axis are displayed on a display means.
7. A tumor region size measuring method as defined in claim 1 , wherein:
the setting of the arbitrary point is performed by setting the arbitrary point at the center of gravity of the tumor region.
8. A tumor region size measuring method as defined in claim 1 , wherein:
the setting of the arbitrary point is performed by setting a plurality of arbitrary points at the centers of gravity of a plurality of cross sections of the tumor region; and
the designation of the long axis is performed by obtaining a line segment having the greatest length from among a plurality of line segments that pass through each of the set arbitrary points and the ends of which are positioned on the outline of the tumor region, and designating the line segment having the greatest length from among the obtained line segments.
9. A tumor region size measuring method as defined in claim 2 , wherein:
the designated long axis and the designated short axis are displayed on a display means.
10. A tumor region size measuring method as defined in claim 2 , wherein:
the lengths of the designated long axis and the designated short axis are displayed on a display means.
11. A tumor region size measuring method as defined in claim 2 , wherein:
the setting of the arbitrary point is performed by setting the arbitrary point at the center of gravity of the tumor region.
12. A tumor region size measuring method as defined in claim 2 , wherein:
the setting of the arbitrary point is performed by setting a plurality of arbitrary points at the centers of gravity of a plurality of cross sections of the tumor region;
the designation of the long axis is performed by obtaining a line segment having the greatest length from among a plurality of line segments that pass through each of the set arbitrary points and the ends of which are positioned on the outline of the tumor region, and designating the line segment having the greatest length from among the obtained line segments.
13. A tumor region size measuring method as defined in claim 3 , wherein:
the designated long axis and the designated short axis are displayed on a display means.
14. A tumor region size measuring method as defined in claim 3 , wherein:
the lengths of the designated long axis and the designated short axis are displayed on a display means.
15. A tumor region size measuring method as defined in claim 4 , wherein:
the designated long axis and the designated short axis are displayed on a display means.
16. A tumor region size measuring apparatus, comprising:
tumor region determining means for determining a tumor region from a three dimensional medical image;
axis designating means for designating a long axis and a short axis of the determined tumor region; and
obtaining means for obtaining the lengths of the designated long axis and the designated short axis as the size of the tumor region;
the axis designating means performing the steps of:
setting an arbitrary point within the determined tumor region;
designating a line segment having the greatest length from among a plurality of line segments that pass through the set arbitrary point, the ends of which are positioned on the outline of a tumor region, as the long axis of the tumor region;
detecting the cross section having the greatest area from among cross sections of the tumor region that include the designated long axis; and
designating the longest line segment from among line segments perpendicular to the designated long axis that connect two points along the outline of the tumor region as the short axis of the tumor region.
17. A non-transitory computer readable medium having a program recorded thereon that causes a computer to execute the procedures of:
determining a tumor region from a three dimensional medical image;
designating a long axis and a short axis of the determined tumor region; and
obtaining the lengths of the designated long axis and the designated short axis as the size of the tumor region;
the axis designating procedure including the steps of:
setting an arbitrary point within the determined tumor region;
designating a line segment having the greatest length from among a plurality of line segments that pass through the set arbitrary point, the ends of which are positioned on the outline of a tumor region, as the long axis of the tumor region;
detecting the cross section having the greatest area from among cross sections of the tumor region that include the designated long axis; and
designating the longest line segment from among line segments perpendicular to the designated long axis that connect two points along the outline of the tumor region as the short axis of the tumor region.
18. A tumor region size measuring method as defined in claim 1 , wherein:
the designation of the long axis is performed by obtaining the lengths of line segments that extend in a number of directions which have been thinned out from all directions from the arbitrary point, and designating the line segment having the greatest length from among these line segments as the long axis.
19. A tumor region size measuring method as defined in claim 2 , wherein:
the detection of the cross section is performed by obtaining the areas of cross sections of the tumor region that pass through the arbitrary point and have as their normal lines a number of directions which have been thinned out from all directions from the arbitrary point, and detecting the cross section having the greatest area from among these cross sections.
20. A tumor region size measuring method as defined in claim 3 , wherein:
the designation of the long axis is performed by sequentially selecting combinations of pairs of points from among a plurality of points which are set at predetermined intervals along the outline of the tumor region, obtaining the lengths between the selected combinations of points, and designating the line segment having a pair of points that have the greatest distance therebetween as the two ends thereof as the long axis.
21. A tumor region size measuring method as defined in claim 4 , wherein:
the designation of the long axis is performed by sequentially selecting combinations of pairs of points from among a plurality of points which are set at predetermined intervals along the outline of the tumor region, obtaining the lengths between the selected combinations of points, and designating the line segment having a pair of points that have the greatest distance therebetween as the two ends thereof as the long axis.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US13/591,067 US20120321155A1 (en) | 2008-01-16 | 2012-08-21 | Method, apparatus, and program for measuring sizes of tumor regions |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2008006424A JP5198883B2 (en) | 2008-01-16 | 2008-01-16 | Tumor area size measuring method, apparatus and program |
JP006424/2008 | 2008-01-16 | ||
US12/354,529 US8249320B2 (en) | 2008-01-16 | 2009-01-15 | Method, apparatus, and program for measuring sizes of tumor regions |
US13/591,067 US20120321155A1 (en) | 2008-01-16 | 2012-08-21 | Method, apparatus, and program for measuring sizes of tumor regions |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US12/354,529 Continuation US8249320B2 (en) | 2008-01-16 | 2009-01-15 | Method, apparatus, and program for measuring sizes of tumor regions |
Publications (1)
Publication Number | Publication Date |
---|---|
US20120321155A1 true US20120321155A1 (en) | 2012-12-20 |
Family
ID=40850669
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US12/354,529 Active 2030-12-28 US8249320B2 (en) | 2008-01-16 | 2009-01-15 | Method, apparatus, and program for measuring sizes of tumor regions |
US13/591,067 Abandoned US20120321155A1 (en) | 2008-01-16 | 2012-08-21 | Method, apparatus, and program for measuring sizes of tumor regions |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US12/354,529 Active 2030-12-28 US8249320B2 (en) | 2008-01-16 | 2009-01-15 | Method, apparatus, and program for measuring sizes of tumor regions |
Country Status (2)
Country | Link |
---|---|
US (2) | US8249320B2 (en) |
JP (1) | JP5198883B2 (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110243402A1 (en) * | 2009-11-30 | 2011-10-06 | Mirada Medical | Measurement system for medical images |
JP2017109074A (en) * | 2015-12-15 | 2017-06-22 | コニカミノルタ株式会社 | Ultrasonic diagnostic imaging apparatus |
CN110533638A (en) * | 2019-08-02 | 2019-12-03 | 杭州依图医疗技术有限公司 | A kind of method and device of measurement object size |
CN110865859A (en) * | 2019-10-29 | 2020-03-06 | 维沃移动通信有限公司 | Picture display method and device, electronic equipment and medium |
US10699401B2 (en) | 2014-09-17 | 2020-06-30 | Volume Graphics Gmbh | Method and system for determining the local quality of surface data extracted from volume date |
Families Citing this family (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4493679B2 (en) * | 2007-03-29 | 2010-06-30 | 富士フイルム株式会社 | Target region extraction method, apparatus, and program |
JP5198883B2 (en) * | 2008-01-16 | 2013-05-15 | 富士フイルム株式会社 | Tumor area size measuring method, apparatus and program |
CN102197413B (en) * | 2008-10-29 | 2017-03-22 | 皇家飞利浦电子股份有限公司 | Analyzing an at least three-dimensional medical image |
JP5618129B2 (en) * | 2010-02-26 | 2014-11-05 | 学校法人東京理科大学 | Medical image display control device and program |
CN102639063B (en) * | 2010-09-30 | 2015-03-18 | 柯尼卡美能达株式会社 | Ultrasound diagnostic equipment |
BR112013016409A2 (en) * | 2010-12-29 | 2016-10-04 | Koninkl Philips Electronics Nv | system (sys) for classifying a tumor of interest, workstation (ws), method (m) for classifying a tumor of interest, and computer program product |
EP2673738A4 (en) | 2011-02-11 | 2017-08-23 | E-4 Endeavors, Inc. | System and method for modeling a biopsy specimen |
ES2692520T3 (en) * | 2011-03-04 | 2018-12-04 | Lbt Innovations Limited | Method to improve classification results of a classifier |
JP5872323B2 (en) * | 2011-03-29 | 2016-03-01 | 株式会社東芝 | X-ray CT apparatus and image processing method |
GB2489709B (en) * | 2011-04-05 | 2013-07-31 | Mirada Medical Ltd | Measurement system for medical images |
JP5087694B2 (en) * | 2011-04-14 | 2012-12-05 | 日立アロカメディカル株式会社 | Ultrasonic diagnostic equipment |
CN104102805A (en) * | 2013-04-15 | 2014-10-15 | 上海联影医疗科技有限公司 | Medical image information processing method and medical image information processing device |
KR20160014470A (en) * | 2014-07-29 | 2016-02-11 | 삼성전자주식회사 | Apparatus and method for supporting computer aided diagnosis based on providing non-visual information, and computer aided diagnosis system |
JP6640839B2 (en) * | 2014-08-29 | 2020-02-05 | コーニンクレッカ フィリップス エヌ ヴェKoninklijke Philips N.V. | Handling unconfirmed quantifiable lesions |
US10127683B2 (en) | 2015-01-06 | 2018-11-13 | Agfa Healthcare Nv | Method for determining the short axis in a lesion region in a three dimensional medical image |
US10325412B2 (en) | 2015-11-05 | 2019-06-18 | Shenyang Neusoft Medical Systems Co., Ltd. | Cutting three-dimensional image |
JP6792340B2 (en) * | 2016-03-29 | 2020-11-25 | ザイオソフト株式会社 | Medical image processing equipment, medical image processing methods, and medical image processing programs |
US10762636B2 (en) * | 2017-06-29 | 2020-09-01 | HealthMyne, Inc. | Systems and methods for volumetric segmentation of structures in planar medical images |
KR102007656B1 (en) * | 2017-10-25 | 2019-08-06 | 재단법인 아산사회복지재단 | Method for measuring length using motion sensor and apparatus for the same |
EP3766430A4 (en) * | 2018-03-16 | 2021-04-21 | FUJIFILM Corporation | Acoustic wave diagnostic device and method for controlling acoustic wave diagnostic device |
EP3818943A4 (en) * | 2018-07-02 | 2021-08-25 | FUJIFILM Corporation | Acoustic wave diagnostic device and method for controlling acoustic wave diagnostic device |
WO2020097100A1 (en) | 2018-11-05 | 2020-05-14 | HealthMyne, Inc. | Systems and methods for semi-automatic tumor segmentation |
KR102236617B1 (en) * | 2019-01-04 | 2021-04-06 | 서울대학교병원 | Method for measuring lung cancer tumor size and apparatus for executint the method |
CN110823311B (en) * | 2019-11-26 | 2022-07-19 | 湖南农业大学 | Method for rapidly estimating volume of rape pod |
US11100640B2 (en) | 2019-11-30 | 2021-08-24 | Ai Metrics, Llc | Systems and methods for lesion analysis |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8249320B2 (en) * | 2008-01-16 | 2012-08-21 | Fujifilm Corporation | Method, apparatus, and program for measuring sizes of tumor regions |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2895414B2 (en) | 1995-05-08 | 1999-05-24 | アロカ株式会社 | Ultrasonic volume calculator |
US5984870A (en) * | 1997-07-25 | 1999-11-16 | Arch Development Corporation | Method and system for the automated analysis of lesions in ultrasound images |
US6973212B2 (en) | 2000-09-01 | 2005-12-06 | Siemens Corporate Research, Inc. | Graph cuts for binary segmentation of n-dimensional images from object and background seeds |
JP4282939B2 (en) | 2002-03-19 | 2009-06-24 | 東芝医用システムエンジニアリング株式会社 | Ultrasonic diagnostic apparatus, image processing apparatus, and image processing program |
JP2004041490A (en) * | 2002-07-12 | 2004-02-12 | Konica Minolta Holdings Inc | Diagnostic imaging support system |
US7894646B2 (en) * | 2003-08-01 | 2011-02-22 | Hitachi Medical Corporation | Medical image diagnosis support device and method for calculating degree of deformation from normal shapes of organ regions |
US20050129297A1 (en) | 2003-12-15 | 2005-06-16 | Kamath Vidya P. | Classification of breast lesion method and system |
JP4062737B2 (en) * | 2004-07-22 | 2008-03-19 | 株式会社日立メディコ | Medical image diagnosis support system |
JP2006223449A (en) * | 2005-02-16 | 2006-08-31 | Fuji Photo Film Co Ltd | Cross section display device and its program |
-
2008
- 2008-01-16 JP JP2008006424A patent/JP5198883B2/en active Active
-
2009
- 2009-01-15 US US12/354,529 patent/US8249320B2/en active Active
-
2012
- 2012-08-21 US US13/591,067 patent/US20120321155A1/en not_active Abandoned
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8249320B2 (en) * | 2008-01-16 | 2012-08-21 | Fujifilm Corporation | Method, apparatus, and program for measuring sizes of tumor regions |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20110243402A1 (en) * | 2009-11-30 | 2011-10-06 | Mirada Medical | Measurement system for medical images |
US10699401B2 (en) | 2014-09-17 | 2020-06-30 | Volume Graphics Gmbh | Method and system for determining the local quality of surface data extracted from volume date |
JP2017109074A (en) * | 2015-12-15 | 2017-06-22 | コニカミノルタ株式会社 | Ultrasonic diagnostic imaging apparatus |
CN110533638A (en) * | 2019-08-02 | 2019-12-03 | 杭州依图医疗技术有限公司 | A kind of method and device of measurement object size |
CN110865859A (en) * | 2019-10-29 | 2020-03-06 | 维沃移动通信有限公司 | Picture display method and device, electronic equipment and medium |
Also Published As
Publication number | Publication date |
---|---|
US8249320B2 (en) | 2012-08-21 |
JP5198883B2 (en) | 2013-05-15 |
US20090180677A1 (en) | 2009-07-16 |
JP2009165615A (en) | 2009-07-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US8249320B2 (en) | Method, apparatus, and program for measuring sizes of tumor regions | |
US8457435B2 (en) | Methods and systems for extended ultrasound imaging | |
US20090129656A1 (en) | Voting in mammography processing | |
US20060184327A1 (en) | Method and system for processing and analyzing digital terrain data | |
US20080219533A1 (en) | Apparatus and Method For Correlating First and Second 3D Images of Tubular Object | |
JP2005506140A5 (en) | ||
US7072515B2 (en) | Method and apparatus for automatically tracing line-structure images | |
US11386553B2 (en) | Medical image data | |
US9092867B2 (en) | Methods for segmenting images and detecting specific structures | |
JP2005198708A (en) | Vasoconstriction rate analyzer and vasoconstriction rate analyzing method | |
JP2010000133A (en) | Image display, image display method and program | |
Li | Detecting lesion bounding ellipses with gaussian proposal networks | |
US7397942B2 (en) | Method for branch selection for probe alignment | |
US20070201737A1 (en) | System And Method For Vascular Visualization Using Planar Reformation Of Vascular Central Axis Surface With Biconvex Slab | |
US20090310883A1 (en) | Image processing apparatus, method, and program | |
Shim et al. | Stereo-vision-based 3D concrete crack detection using adversarial learning with balanced ensemble discriminator networks | |
Gulyaev | Measurement of the length of objects on scanning probe microscope images using curvature detectors | |
CN115409858A (en) | Medical image segmentation method, equipment and device | |
US8422748B2 (en) | System and method for grouping airways and arteries for quantitative analysis | |
CN114240899A (en) | Guided wave frequency dispersion curve modal tracking method and system based on image registration | |
CN109242823B (en) | Reference image selection method and device for positioning calculation and automatic driving system | |
WO2021122098A1 (en) | A method and system for improved ultrasound plane acquisition | |
CN117553756B (en) | Off-target amount calculating method, device, equipment and storage medium based on target tracking | |
JP2853676B2 (en) | Radio emission source range estimation device | |
Tao et al. | Anatomical Structure-Aware Pulmonary Nodule Detection via Parallel Multi-task RoI Head |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
STCB | Information on status: application discontinuation |
Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION |