WO2008035446A1 - Processeur d'images médicales, procédé et programme de traitement d'images associés - Google Patents

Processeur d'images médicales, procédé et programme de traitement d'images associés Download PDF

Info

Publication number
WO2008035446A1
WO2008035446A1 PCT/JP2006/319304 JP2006319304W WO2008035446A1 WO 2008035446 A1 WO2008035446 A1 WO 2008035446A1 JP 2006319304 W JP2006319304 W JP 2006319304W WO 2008035446 A1 WO2008035446 A1 WO 2008035446A1
Authority
WO
WIPO (PCT)
Prior art keywords
blood vessel
image
abnormal
length
relative
Prior art date
Application number
PCT/JP2006/319304
Other languages
English (en)
Japanese (ja)
Inventor
Hiroshi Fujita
Yoshikazu Uchiyama
Hitoshi Futamura
Original Assignee
Konica Minolta Medical & Graphic, Inc.
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Konica Minolta Medical & Graphic, Inc. filed Critical Konica Minolta Medical & Graphic, Inc.
Publication of WO2008035446A1 publication Critical patent/WO2008035446A1/fr

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/54—Signal processing systems, e.g. using pulse sequences ; Generation or control of pulse sequences; Operator console
    • G01R33/56—Image enhancement or correction, e.g. subtraction or averaging techniques, e.g. improvement of signal-to-noise ratio and resolution
    • G01R33/563—Image enhancement or correction, e.g. subtraction or averaging techniques, e.g. improvement of signal-to-noise ratio and resolution of moving material, e.g. flow contrast angiography
    • G01R33/5635—Angiography, e.g. contrast-enhanced angiography [CE-MRA] or time-of-flight angiography [TOF-MRA]
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/54—Signal processing systems, e.g. using pulse sequences ; Generation or control of pulse sequences; Operator console
    • G01R33/56—Image enhancement or correction, e.g. subtraction or averaging techniques, e.g. improvement of signal-to-noise ratio and resolution
    • G01R33/5608—Data processing and visualization specially adapted for MR, e.g. for feature analysis and pattern recognition on the basis of measured MR data, segmentation of measured MR data, edge contour detection on the basis of measured MR data, for enhancing measured MR data in terms of signal-to-noise ratio by means of noise filtering or apodization, for enhancing measured MR data in terms of resolution by means for deblurring, windowing, zero filling, or generation of gray-scaled images, colour-coded images or images displaying vectors instead of pixels
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0012—Biomedical image inspection
    • 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
    • G06T2207/10088—Magnetic resonance imaging [MRI]
    • 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
    • G06T2207/30016—Brain
    • 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
    • G06T2207/30101—Blood vessel; Artery; Vein; Vascular

Definitions

  • the present invention relates to a medical image processing apparatus that automatically detects a vascular site where occlusion occurs in an MRA image.
  • the number of examinations of brain docks is rapidly increasing as the performance and spread of magnetic resonance tomography (MRI) devices increase.
  • MRI magnetic resonance tomography
  • the purpose of the brain dog is to prevent the onset or progression of the discovered risk factors by early detection of risk factors related to cerebrovascular diseases such as vascular closure or stenosis and unruptured aneurysms. .
  • the MRI device can generate MRA (Magnetic Resonance Angiography) images that visualize the blood flow in the blood vessels. Abnormalities such as occlusion and stenosis of blood vessels and unruptured aneurysms can be generated from these MRA images. Can be found.
  • MRA Magnetic Resonance Angiography
  • Patent Document 1 A technique related to a display method of MRA images is disclosed (see Patent Document 1).
  • Patent Document 2 a technique related to a method and apparatus for automatically detecting stenosis by comparing the diameter of a blood vessel measured from a blood vessel image obtained by MRI with the diameter of a blood vessel serving as a reference is disclosed (Patent Document 2). reference).
  • Patent Document 1 Japanese Patent Laid-Open No. 5-277091
  • Patent Document 2 Japanese Patent Application Laid-Open No. 2004-329929
  • the blood vessel image is removed when another blood vessel image crosses further in front of the blood vessel image observed by the doctor. I can't. Therefore, it is not always possible to extract and observe only a desired blood vessel image, and it is difficult to observe in detail at the intersection.
  • the method described in Patent Document 2 is a technique for automatically detecting a stenosis or an aneurysm of a blood vessel, but is not related to a technique for automatically detecting “occlusion” of a blood vessel.
  • the method described in Patent Document 2 even if the presence or absence of abnormal blood vessels can be detected, as described above in Patent Document 1! /, When blood vessel images intersect, In some cases, it is unclear which vascular site is abnormal.
  • An object of the present invention is to provide a medical image processing apparatus having a function of automatically detecting an abnormal blood vessel site with a high possibility of occurrence of occlusion.
  • the first aspect of the present invention is:
  • the blood vessel image is compared with a reference image in which one or more blood vessel portions included in the blood vessel image are specified in advance, and one or more blood vessel portions included in the blood vessel image are determined based on the comparison result.
  • Abnormal blood vessel detecting means for calculating relative blood vessel lengths for each of the blood vessel portions discriminated by the blood vessel discriminating means, and detecting abnormal blood vessel portions based on the result of discriminant analysis using each relative blood vessel length as a feature quantity; ,
  • Display control means for identifying and displaying the detected abnormal blood vessel site in the displayed examination image
  • a third aspect of the present invention is the second aspect of the present invention.
  • the blood vessel name is determined for each blood vessel site,
  • the abnormal blood vessel detecting means determines a blood vessel name of the detected abnormal blood vessel site based on the reference image;
  • the display control means displays the identified abnormal blood vessel name in association with the abnormal blood vessel identified and displayed.
  • the abnormal blood vessel detection means detects the presence or absence of an abnormal blood vessel portion by comparing the calculated relative blood vessel length of each blood vessel portion with the relative blood vessel length of each blood vessel portion in a normal case. .
  • the abnormal blood vessel detection means performs thinning processing on the blood vessel image, and calculates the relative blood vessel length of each blood vessel site using the processed image.
  • a sixth aspect of the present invention provides any one of the first to fifth aspects of the present invention.
  • the inspection target image is an MRA image
  • the abnormal blood vessel detecting means detects an abnormal blood vessel site where an occlusion has occurred.
  • a seventh aspect of the present invention is the process according to any one of the first to sixth aspects of the present invention.
  • Storage means for storing information relating to abnormal blood vessel sites detected by the abnormal blood vessel detection means in association with information on the relative blood vessel length used for the detection;
  • the display control means distributes the relative blood vessel length information stored in association with the storage means in a two-dimensional space, displays the information on the display means, and further distributes the information in the two-dimensional space.
  • the relative blood vessel length information is designated, information on the abnormal blood vessel part associated with the designated relative blood vessel length information is acquired from the storage means and displayed on the display means. It is characterized by that.
  • Storage means for storing information relating to abnormal blood vessel sites and feature amounts in association with each other, and a distance between the feature amount of the examination target image in the coordinate space when the feature amount is plotted in the coordinate space is within a predetermined range
  • a search means for searching for information related to the abnormal blood vessel site associated with the feature quantity stored in the storage means. It is a sign.
  • a ninth aspect of the present invention provides:
  • the blood vessel image is compared with a reference image in which one or more blood vessel portions included in the blood vessel image are specified in advance, and one or more blood vessel portions included in the blood vessel image are determined based on the comparison result. Discriminating step to perform,
  • a tenth aspect of the present invention is the ninth aspect of the present invention.
  • Blood vessel image force in the examination target image The blood vessel length of each blood vessel site is calculated, and the ratio of the blood vessel length of each blood vessel site to the blood vessel length of all blood vessels (relative blood vessel length) is further calculated.
  • the relative vascular length found in relation to multiple vascular sites as described above may cause a large change in the relative vascular length of other vascular sites if the vascular length changes in one vascular site. It becomes. Therefore, by discriminating and analyzing each calculated relative blood vessel length as a feature amount, it is detected whether or not the force includes an abnormal blood vessel portion that is likely to be abnormal in each blood vessel portion of the target blood vessel image. In addition, it is possible to specify which vascular part is the abnormal vascular part.
  • the name of each blood vessel part is attached to the target blood vessel image and displayed.
  • the doctor can easily grasp the name of each blood vessel site.
  • it is possible to prevent erroneous recognition of blood vessel sites on a two-dimensional image in which blood vessel images overlap in a complex manner.
  • Target vessel image force By comparing the calculated relative blood vessel length of each blood vessel with the reference relative vessel length calculated separately, whether or not each vessel in the target blood vessel image contains an abnormal blood vessel is determined. Detected. When an abnormal blood vessel is detected, a notice to that effect is sent to a reader such as a doctor or engineer. As a result, it is possible to prevent an abnormal blood vessel site from being overlooked due to an examination target image that is difficult to interpret or due to a doctor's fatigue or the like.
  • Thinning is performed on the blood vessel image extracted from the examination target image.
  • the calculation of the length of blood vessels is greatly affected by the thickness of the blood vessels, and the thickness of the blood vessels varies greatly depending on the subject. Therefore, the blood vessel image can be calculated using only the central portion of the blood vessel image by performing thin line processing, and the influence of individual differences on the calculation of the blood vessel length is minimized. Can be. Therefore, the relative blood vessel length of each blood vessel part in the blood vessel image can be calculated with high accuracy.
  • the vascular site where the occlusion occurs can be detected.
  • the storage means can store the relative blood vessel length data calculated in the past by the abnormal blood vessel detection means in association with the blood vessel image or the like. Among the stored data, one or a plurality of data is displayed by the display control means. The operator can easily search for past cases similar to the currently detected case based on the displayed past data. Based on the retrieved cases, the doctor can take appropriate measures for the abnormal blood vessel site that is currently detected.
  • FIG. 1 is a diagram showing an internal configuration of a medical image processing apparatus 10 according to the present invention.
  • FIG. 2 is a flowchart explaining an outline of processing of the medical image processing apparatus 10 according to the present invention.
  • FIG. 3A is a diagram showing a reference image.
  • [3B] A diagram showing a blood vessel image.
  • FIG. 4 is a flowchart for explaining blood vessel part discrimination processing.
  • FIG. 5A is a diagram showing an example of normalization processing for normal cases.
  • FIG. 5B is a diagram showing an example of normalization processing for abnormal cases.
  • FIG. 6A is a diagram showing an example of a binary key image.
  • FIG. 6B is a diagram showing a target blood vessel image after normal ⁇ .
  • FIG. 7A is a view in which a target blood vessel image and a reference image before alignment are superimposed.
  • FIG. 7B is a diagram in which the centroid positions of the target blood vessel image and the reference image are matched.
  • FIG. 8A is a diagram showing landmarks in a reference image.
  • FIG. 8B is a diagram showing corresponding points in a blood vessel extraction image.
  • FIG. 9A is a diagram showing a blood vessel extraction image of an abnormal case and a discrimination result of the blood vessel part.
  • ⁇ 9B] is a diagram showing a blood vessel extraction image of a normal case and a discrimination result of the blood vessel part.
  • FIG. 10 is a diagram showing an example of identification display of each blood vessel part discriminated in the target image.
  • FIG. 11 is a flowchart for explaining abnormal blood vessel detection processing.
  • FIG. 12A is a diagram showing an example of a thinning process for normal cases.
  • FIG. 12B is a diagram showing an example of a thinning process for abnormal cases.
  • FIG. 13 is a diagram showing an example of relative blood vessel lengths calculated for a plurality of subjects.
  • FIG. 14A is a diagram showing a distribution plotted in a three-dimensional space as a three-dimensional feature.
  • ⁇ 14B] shows the ratio of normal cases and occluded cases with three relative vessel lengths.
  • FIG. 15A is a diagram showing a distribution after conversion to a two-dimensional space.
  • FIG. 15B is a diagram showing an example of search and display of past cases.
  • FIG. 15C is a diagram showing the results of discriminant analysis.
  • FIG. 16 is a diagram showing an example in which each blood vessel part name is associated with each relative blood vessel length. Explanation of symbols [0025] 10 Medical image processing apparatus
  • FIG. 1 shows an internal configuration of the medical image processing apparatus 10 in the present embodiment.
  • the medical image processing apparatus 10 includes a control unit 11, an operation unit 12, a display unit 13, a storage unit 14, and a communication unit 15.
  • the control unit 11 functions as each means of an extraction means, a blood vessel discrimination means, an abnormal blood vessel discrimination means, a display control means, and a search means.
  • the display unit 13 functions as a display unit
  • the storage unit 14 functions as a storage unit.
  • the operation unit 12 and the communication unit 15 operate according to instructions from the above-described units.
  • the control unit 11 includes a CPU (Central Processing Unit), a RAM (Random Access Memory), and the like.
  • the control unit 11 expands a predetermined program stored in the storage unit 14 into the RAM and cooperates with the program. Controls the overall processing operation.
  • the operation unit 12 includes a keyboard having numeric keys, character keys, function keys, and the like, a mouse, and the like, and outputs an operation signal corresponding to the operated key to the control unit 11.
  • the display unit 13 includes an LCD (Liquid Crystal Display) or the like, and displays various display information such as medical images and operation screens in accordance with display control of the control unit 11. Note that the display unit 13 is configured as a touch panel integrated with the operation unit 12.
  • LCD Liquid Crystal Display
  • the storage unit 14 includes various programs such as a control program executed by the control unit 11 and an image processing program in addition to the system program, and stores various types of meters necessary for the execution of each program. The data processed later is stored. In addition, information on abnormal blood vessel sites in the past case and a relative blood vessel length that is a feature amount calculated in the case are stored in association with each other. The relative blood vessel length will be described later.
  • the communication unit 15 includes a communication interface such as a network interface card or a modem. A medical image generated by the medical image processing apparatus 10 is transmitted to an external apparatus such as a printer server, an image server, or a view lamp via the communication unit 15.
  • blood vessel part discrimination process (steps S1 to S4) for naming each blood vessel part for short and “abnormal blood vessel detection process” (steps S5 to S8) for detecting abnormal blood vessel candidates such as occlusions. ing.
  • step S1 corresponds to an extraction process
  • steps S2 to S4 correspond to a discrimination process
  • steps SI1 to S17 correspond to a discrimination process
  • steps S5 to S8 and steps S21 to S30 correspond to a detection process.
  • a three-dimensional MRA image (hereinafter referred to as “target image”) to be examined is input by the control unit 11 (step S1).
  • the control unit 11 extracts a blood vessel image (hereinafter referred to as “target blood vessel image”) from the input target image, and compares the target blood vessel image with a reference image (see FIG. 3A) in which a blood vessel part is set in advance. Then, each image processing involving translation and rotation is performed on the target image (step S2).
  • the target blood vessel image and the reference image are compared (step S3), and a plurality of blood vessel parts included in the target blood vessel image are determined based on the comparison result.
  • information on the names of the respective blood vessel parts set in the reference image is written in the header area of the target image, which is the original image, and attached (step S4).
  • the ratio of the length of each blood vessel portion to the length of all blood vessel portions (hereinafter referred to as "relative blood vessel length") is calculated individually by the control unit 11 (step S). Five). Each calculated relative blood vessel length is compared with the ratio of the length of each blood vessel site in a normal case (hereinafter “normal value”) (step S6).
  • the normal value here is normal and normal. 3D MRA image power of a subject who is recognized by a doctor as being in an average health state.
  • control unit 11 when each relative blood vessel length for which the target blood vessel image force is also calculated as a result of the comparison is different from the normal value by a predetermined value or more, the control unit 11 is occluded at a blood vessel part that is the predetermined value or more. It is determined that there is a high possibility that the blood vessel portion is present, and the blood vessel site is detected as an abnormal blood vessel candidate (step S7).
  • MRA images are images of blood flow. Since blood flow does not flow in the blood vessel where the occlusion occurs, the image is not displayed as a blood vessel image in the MRA image. Therefore, in the case where occlusion occurs, it is considered that the balance of the relative blood vessel lengths is greatly lost as compared with the normal case.
  • abnormal blood vessel candidates with a high possibility of occlusion are detected by obtaining each relative blood vessel length and comparing it with each relative blood vessel length (normal value) of a normal case.
  • step S8 when the name of the abnormal blood vessel candidate is determined on the basis of the incidental information of the 3D MRA image in the control unit 11 (step S8), this process ends.
  • the reference image is obtained by presetting positions and names of one or a plurality of blood vessel sites in a blood vessel image on the 3D MRA image.
  • the vascular part refers to the anatomical classification of blood vessels, and the position of the vascular part belongs to the vascular part. Refers to the position of the botacel.
  • FIG. 3A eight vascular sites included in the blood vessel image (anterior cerebral artery AC, right middle cerebral artery RC, left middle cerebral artery LC, right internal carotid artery RI, left internal carotid artery LI, right posterior cerebral artery)
  • the positions of RP, left posterior cerebral artery LP, and basilar artery BA are set.
  • the three vascular sites (the right middle cerebral artery RC, the anterior cerebral artery AC, and the basilar artery BA) have names indicating all eight vascular sites. The name is set.
  • the reference image g2 is generated as a three-dimensional data force of the head MRA image gl selected for the reference image as shown in FIG. 3B.
  • an axial image (a two-dimensional tomographic image obtained by cutting out a botasel in a plane perpendicular to the body axis) is created at intervals of this three-dimensional data force.
  • the botacell belonging to each vascular site is designated by manual operation, and the name of the vascular site is further designated.
  • landmark botasels are set at characteristic points such as a blood vessel bending point, an end point, and an intersection of blood vessel parts.
  • the landmark is used for alignment between the target image and the reference image, and will be described in detail later.
  • Landmarks are also set according to manual operations based on doctors' indications.
  • the reference image g2 may be created by the control unit 11 of the medical image processing apparatus 10, or an externally created image may be stored in the storage unit 14.
  • each blood vessel part is identified and displayed in FIG. 3A, but the actual reference image g2 has a black background (low signal value) and a white blood vessel image (high signal value). ) And binarized image. Then, the position information, name information, and position information of the button cell that belongs to each blood vessel site are attached to the reference image or stored in the storage unit 14 as a separate file in association with the reference image. Is done.
  • a normalization process is first performed on the target image by the control unit 11 (step S11).
  • the botacell may be a rectangular parallelepiped, or the maximum and minimum values of the voxel may vary. It occurs. Therefore, normalization processing is performed to unify the preconditions regarding the target image.
  • the target image is converted by the linear interpolation method so that all the sides constituting the botacell have the same size.
  • a histogram is created for all the botacell values of the target image, and all the botacel values of the target image are 0, with the top 5% or more of the histograms having a value of 1024 for the top 5% or more and 0 for the minimum. Linear conversion to 1024 tones.
  • the density gradation range is not limited to 0 to 1024 and can be set as appropriate.
  • FIG. 5A and FIG. 5B show an example of normalization processing.
  • the target image g3 shown in FIG. 5A and the target image g4 shown in FIG. 5B are different patients.
  • the histogram hi (see Fig. 5A) obtained from the target image g3 and the histogram h3 (see Fig. 5B) obtained from the target image g4 share the common feature that there are two local maxima. It can be seen that there is a considerable difference in the range of the values, and that the histogram characteristics are different as a whole.
  • the histogram h2 shown in FIG. 5A and the histogram h4 shown in FIG. Histograms h2 and h4 force As shown by the component force, the histogram characteristics of the target images g3 and g4 are almost the same by normalization.
  • control unit 11 extracts a blood vessel image from the normalized target image (step S12).
  • threshold processing is performed on the target image, and binarization is performed.
  • the blood vessel image is white as shown in Fig. 6A. Therefore, in the binary image, the blood vessel image has a different value from other regions. Therefore, the region having the same signal value as the blood vessel image is extracted by the region expansion method.
  • a binary cell is used to determine the start point (the whitest white and high density value voxel) and the target image before the binarization process is determined as the start point.
  • 26 neighbors are examined, and a neighboring vessel that satisfies a certain determination condition (for example, a density value of 500 or more) is determined to be a blood vessel image. Then, the same processing as described above is repeated for the neighboring buttonacell determined to be the blood vessel image. In this way, the image region of the blood vessel image can be extracted by sequentially extracting the botacell satisfying the determination condition while expanding the region.
  • a certain determination condition for example, a density value of 500 or more
  • FIG. 6B shows a target blood vessel image g6 from which a blood vessel image is extracted.
  • a blood vessel image is extracted from the target image g5 after normal input shown in FIG. 6A, the region of the blood vessel image is white (density value 1024), and the other regions are black (density value 0). It is binarized.
  • step S13 In the control unit 11, in order to make the position of the target blood vessel image substantially coincide with the position of the blood vessel image of the reference image, alignment is performed based on the barycentric position of each image (step S13).
  • the center-of-gravity position is the position of the botasel that is the center of gravity of all the botasels belonging to the blood vessel image.
  • FIG. 7A is a diagram in which the target blood vessel image before alignment and the reference image are superimposed.
  • Fig. 7 A force It can be seen that the positions of the respective blood vessel images coincide with each other simply by combining the target blood vessel image and the reference image.
  • the control unit 11 obtains the positions of the center of gravity P (xl, yl, zl) of the target blood vessel image and the center of gravity Q (x2, y2, z2) of the reference image.
  • the target blood vessel image or the reference image is translated so that the barycentric positions P and Q coincide with each other.
  • FIG. 7B shows the result of matching the center positions P and Q by translation. From FIG. 7B, it can be seen that the positions of the target blood vessel image and the reference image roughly match.
  • control unit 11 performs rigid body deformation on the target blood vessel image (step S 14).
  • a corresponding point search using a cross-correlation coefficient is performed as preprocessing for rigid body deformation.
  • rigid deformation a plurality of corresponding points are set for each of two images to be aligned, and one image is converted so that the corresponding points set in the two images match each other.
  • a landmark botel cell that is determined in advance in the reference image and a target vascular image botel that has locally similar image characteristics are set as corresponding points.
  • the similarity in image characteristics is determined based on the cross-correlation coefficient obtained for the target blood vessel image and the reference image.
  • corresponding points corresponding to 12 landmarks set in advance in the blood vessel image of the reference image g7 are searched from the target blood vessel image.
  • the target blood vessel image g8 has a start point at the position corresponding to each landmark in the reference image g7.
  • Botacel force at the start point and landmarks Botacells within the range of -10 to +10 botacells in the X, Y and Z directions (21 X 21 X 21 Botacel cubic region) are searched for each
  • the cross-correlation coefficient C (hereinafter referred to as “correlation value C”) is calculated by equation (1).
  • Correlation value C has a range of ⁇ 1.0 to 1.0, and the closer to maximum value 1.0, the more similar the image characteristics of reference image g7 and target blood vessel image g8! / Indicates.
  • the position force of the boatel cell having the largest correlation value C is set as the corresponding point of the target blood vessel image g8 corresponding to the landmark of the reference image g7.
  • the control unit 11 aligns the target blood vessel image g8 and the reference image g7 by performing rigid body deformation on the target blood vessel image g8 based on the corresponding point.
  • the Rigid body deformation is one of the affine transformations, in which coordinate transformation is performed by rotation and translation.
  • the alignment is performed so that the corresponding point of the target blood vessel image g8 coincides with the landmark of the reference image g7 by an ICP (Iterative Closest Point) algorithm that repeats rigid body deformation using the least squares method multiple times.
  • each blood vessel part included in the target blood vessel image aligned by rigid body deformation is determined by the control unit 11 (step S 15).
  • all the botasels belonging to each blood vessel part of the reference image are targeted (hereinafter referred to as “target botasels
  • the square of the Euclidean distance from the target vessel cell in the target blood vessel image is obtained. Then, it is determined that the blood vessel part to which the target button cell having the shortest Euclidean distance belongs is the blood vessel part to which the target button cell belongs. At this time, the name of the blood vessel part of the target button cell is determined from the name of the blood vessel part that is set as the target button cell.
  • the button cell at the position (x3, y3, z3) is a blood vessel site of the anterior cerebral artery
  • the button cell at the position ((x3, y3, z3)) has the vessel name “anterior cerebral artery”.
  • the blood vessel part information indicating the blood vessel part is attached by being stored in the header area of the target image.
  • FIG. 9A and FIG. 9B show the results of determining the blood vessel site.
  • the target blood vessel image g9 shown in FIG. 9A and the target blood vessel image gl l shown in FIG. 9B are images obtained from different subjects, respectively, and the image glO shown in FIG. 9A and the image g 12 shown in FIG. 9B are respectively obtained.
  • This is an image in which each blood vessel part is discriminated from the target blood vessel images g9 and gl l and is identified and displayed by changing the color for each blood vessel part.
  • the blood vessels in the images glO and gl2 have different shapes (blood vessel position, size, extension direction, etc.) and the same It can be seen that the part can be identified.
  • the above is the flow from determining the blood vessel part in the target image to attaching the blood vessel part information.
  • the control unit 11 performs MIP (Maximum Intensity Projection) processing on the target image to generate a MIP image, and the display unit (The display of the MIP image is hereinafter referred to as the MIP display).
  • MIP Maximum Intensity Projection
  • the MIP method is a technique for displaying the maximum value in the projection path on the projection plane (2D plane) by performing projection processing on the image data constructed in 3D in an arbitrary viewpoint direction. This is an effective technique when it is difficult to grasp the continuity as a whole only by tomographic images.
  • the control unit 11 refers to the vascular part information attached to the target image, and the vascular part information is Based on this, display control is performed so that each blood vessel part can be identified in the blood vessel image in the MIP image (step S17).
  • the control unit 11 uses the MIP image to identify the button cells belonging to the blood vessel part of the anterior cerebral artery AC.
  • the color of each blood vessel part is set to the button cell located at the position determined for each blood vessel part, such as blue, green for the botacel belonging to the blood vessel part of the basilar artery BA. Then, the set color is reflected in the MIP image of the target image.
  • an annotation image indicating the name of the blood vessel part is created and combined with the corresponding blood vessel part of the MIP image.
  • FIG. 10 shows an example of identification display.
  • MIP image gl3 is the target image with the head upward force displayed in MIP.
  • the identification display image gl4 is displayed.
  • the identification display image gl4 is obtained by identifying and displaying each blood vessel part by giving different colors to the eight kinds of blood vessel parts in the target blood vessel image.
  • an annotation m indicating the name of the blood vessel part such as “basal artery” is displayed in association with the selected blood vessel part by the display control of the control unit 11. Is displayed.
  • the control unit 11 When the MIP display from the side surface direction is instructed for the MIP image gl3 in the head upward direction, the control unit 11 creates and displays the MIP image gl5 corresponding to the side surface direction.
  • the MIP image gl5 In the MIP image gl5, the blood vessel image on the front side overlaps the blood vessel image on the rear side, making it difficult to observe. Even in such a MIP image gl5, it is possible to identify and display blood vessel sites.
  • the blood vessel part information it is possible to determine which botacell corresponds to which blood vessel part, and therefore, it is possible to identify the botacell to be identified and displayed regardless of the change in the observation direction of the MIP display. is there.
  • the identification display image corresponding to the MIP image g15 is the image g16.
  • this identification display image g 16 it is possible to extract and display only one of the blood vessel sites.
  • the MIP image in which only the luminance of the selected vessel vessel is projected that is, the MIP image of the target image.
  • a blood vessel selection image gl7 in which only the blood vessel site selected from gl5 is extracted is displayed.
  • the blood vessel selection image gl7 only the selected blood vessel part is displayed in MIP and the other blood vessel parts are not displayed, so that the doctor can observe only the blood vessel part of interest.
  • the image g 13 ⁇ Gl7 that is they appear may be displayed side by side on the same screen, it is also possible to switch the display as one screen first image.
  • the MIP image gl3 the identification display image gl4, the blood vessel selection image gl7, etc.
  • each image gl3 to gl7 can be observed in full screen display, Observe details.
  • the control unit 11 After the above-described blood vessel part discrimination processing, the control unit 11 performs thinning processing on the target blood vessel image (step S21).
  • Thin line ⁇ is a process that converts a figure with a certain thickness into a line figure with a thickness of 1, excluding characteristic points such as inflection points, end points, and intersections between blood vessel sites.
  • the length of the blood vessel is calculated. This length is calculated by the number of botasels constituting the blood vessel region. Even if the blood vessel region is the same blood vessel region, individual differences occur depending on the subject.
  • the thin line process is performed to minimize individual differences in the blood vessel region.
  • the Euclidean distance-converted image is an image defined by the square of the minimum value of the distance from the button cell in which all input images have a value of 0 (hereinafter “0 button cell”) to the target button cell. Images with the smallest distance are extracted from the boundary bocells (one or more 0 bocells exist in the vicinity of 26 of the target ones), and those botacells are classified into 9 types according to the arrangement of the 0 bocells in the vicinity.
  • FIG. 12A and FIG. 12B show an example of the result of the thinning process.
  • Fig. 12A shows the blood vessel image of the subject in normal and average health condition (hereinafter referred to as "normal blood vessel image”) gl8 and gl9
  • Fig. 12B shows the blood vessel image of the subject with obstruction (hereinafter referred to as “occluded blood vessel image”).
  • Image ) Shows gl9 and g20.
  • the blood vessel images gl8 and g20 are those before the thin line processing
  • the blood vessel images gl9 and g21 are those after the thin line processing. Comparing the normal blood vessel image gl9 and the occluded blood vessel image g21, for example, it can be seen that the difference in the length of the blood vessel in the right middle cerebral artery RC is clearly displayed!
  • the length of the blood vessel in each blood vessel part is calculated from the occluded blood vessel image g21 that has been subjected to the thin line processing by the control unit 11 (step S22).
  • a blood vessel core image image where the central part (core) of the blood vessel obtained by thinning the blood vessel is 1 votacel and the other botasels are 0 botacels
  • the length of the blood vessel is calculated by counting the number of the botacells per one.
  • the relative blood vessel length is calculated by the control unit 11 (step S23).
  • Relative vessel length is the ratio of the length of each vessel site to the length of all vessel sites.
  • the length of all blood vessel parts is calculated by obtaining the sum of the lengths of the respective blood vessel parts calculated in step S22.
  • the eight blood vessels described above in the blood vessel discrimination process anterior cerebral artery AC, right middle cerebral artery RC, left middle cerebral artery LC, right internal carotid artery RI, left internal carotid artery LI, right posterior cerebral artery RP, left
  • An example of calculating the relative vascular length of each of the eight vascular regions of the posterior cerebral artery LP and the basilar artery BA will be described.
  • FIG. 13 shows an example of relative blood vessel lengths calculated for a plurality of subjects.
  • the relative blood vessel lengths (normal values) calculated from the normal blood vessel images of subjects in an average health state are the same ratio even for different subjects (for example, as shown in FIG. 13).
  • the proportion of the anterior cerebral artery AC of subjects A to C is similar).
  • each relative blood vessel length calculated from the blood vessel image of the subject with occlusion has a significantly reduced proportion of vascular sites where the occlusion has occurred, and a larger proportion of other normal vascular sites.
  • the balance of relative vessel length is greatly broken.
  • each blood vessel site is normal or not. Judgment can be made easily. Furthermore, when the obstruction values are distributed at close positions, the obstruction values within the distribution range can be easily retrieved as past similar cases.
  • step S24 the eight relative blood vessel lengths calculated by the processing in step 23 described above are plotted as eight feature amounts in an eight-dimensional space.
  • FIG. 14A shows a distribution diagram plotted in a three-dimensional space with three relative vessel lengths as features. For convenience of explanation, an example of plotting in a three-dimensional space based on three relative blood vessel lengths has been shown here, but in reality, as described above, eight relative blood vessel lengths are used as elements and eight dimensions. It shall be plotted in space.
  • FIG. 14B shows normal subjects A to C at three relative blood vessel lengths and subjects D and E with obstruction.
  • PA to PC in FIG. 14A are distributions of normal amounts of the normal values calculated from the normal blood vessel image gl9. At normal values, the relative blood vessel lengths maintain a certain balance, so PA to PC are distributed close to each other and located at a certain position, and constitute a certain range N (hereinafter “normal range N”).
  • PD and PE are distributions of occlusion values, and the normal range N force is also distributed at a certain distance. Therefore, Euclidean distance from PD or PE to normal range N is For the distribution calculated by the control unit 11 and the calculated distance exceeding a predetermined value, it is determined that the possibility of blockage is high.
  • the control unit 11 converts the 8-dimensional space, which is the feature space, into a 2-dimensional space by principal component analysis (step S25).
  • Principal component analysis is a method for determining the characteristics of a plurality of components by combining and compressing a plurality of correlated elements into a single component.
  • the principal component analysis performed here compresses eight relative vessel lengths, calculates one component, and plots its characteristics in a two-dimensional space. Plotted positions will not be plotted at the same location unless all eight features are the same.
  • Fig. 15A is a distribution diagram obtained by transforming an 8-dimensional space into a 2-dimensional space by principal component analysis.
  • Several normal values PA to PC, PAa to PCa
  • occlusion values PD, PE, PDa, PEa
  • the normal value and the occlusion value plotted in the coordinate space are based on the past cases and the examination target images stored in the storage unit 14.
  • the normal values are distributed in the normal range N such that the normal values are close to each other, while the blockage values are distributed in the blockage range AN separated by the normal range N force. Even within the blockage range AN, the blockage values may be distributed at close distances (predetermined range). The fact that the occlusion values are distributed close to each other means that the blood vessel site where the occlusion occurs is the same, and that the degree of occlusion occurs! Means.
  • an advantage of displaying the feature amount distributed in the eight-dimensional space by the display unit 13 by distributing it in the two-dimensional space is that it can be easily visually recognized.
  • a desired portion plotted is clicked with a mouse or the like (not shown) provided in the operation unit 12.
  • the past similar cases corresponding to the clicked feature amount are displayed on the display unit 13 by the display control of the control unit 11 receiving the instruction from the operation unit 12.
  • Fig. 15B shows an example of the display screen when the desired location is clicked with the mouse.
  • the information I relating to the abnormal blood vessel site stored in the storage unit 14 in association with the relative blood vessel length information is displayed on the screen. Is displayed.
  • the information I regarding the abnormal blood vessel site is each piece of information such as the name of the abnormal blood vessel site, the relative blood vessel length, and the occluded blood vessel image. In this way, by setting the display unit 13 to display past similar cases, it is possible to easily search for past case data.
  • Fig. 15C is an example of discriminating normal values (PA to PC, PAa to PCa) and occlusion values (PD, PE, PDa, PEa) by discriminant analysis using QDA.
  • QDA is the extraction of a feature amount from normal values and occlusion values, and the distribution of normal values and occlusion values in the feature amount space is separated by a quadratic surface. This is a method for discriminating class Na and the other as blockage value class ANa.
  • the quadric surface can be expressed by the following equation (6).
  • QDA represents an average feature vector of occlusion values and normal values, and a covariance matrix of feature values.
  • QDA generates an identification boundary M that divides the feature space measured by the eight feature values into two classes: normal value class Na and occlusion value class ANa.
  • the discrimination boundary by QDA is a quadric surface given by the discriminant function.
  • the output value of the discriminant function M indicates the possibility of a blockage value. Therefore, by performing threshold processing on the output of the discriminant function M, the features can be classified into normal value class Na and occlusion value class ANa.
  • the threshold value of the output value of the discriminant function M the detection performance of the occlusion value can be arbitrarily determined.
  • the power of using QDA as a discriminator is not limited to this, and a neural network or a support vector machine may be used!
  • the blockage distribution can be grasped more visually, and the normal value and the blockage value can be easily distinguished.
  • the plot values displayed on the display unit 13 whether the plot value arbitrarily selected by the operator using the operation unit 12 (or the plot value from which the target image force has been acquired) is a normal value or an occlusion value depends on the identification boundary M. Is determined by the control unit 11 (step S27).
  • step S27; No When it is determined that the selected plot value is smaller than the identification boundary M (step S27; No), that is, when the normal value class is Na, the plot value is recognized as a normal value. Is done. In addition to such determination and recognition by the control unit 11 through such discriminant analysis, in the case of a mode in which the distribution charts shown in FIGS. Judgment and recognition are possible.
  • step S27 If it is determined that the selected plot value is larger than the identification boundary M (step S27; Yes), that is, if the selected block value is the occlusion value class ANa, the plot value is recognized as an occlusion value. (Step S28). For the plot value recognized as the occlusion value, the process proceeds to the blood vessel site detection process described below.
  • FIG. 16 shows the names of vascular sites associated by the vascular site discrimination process (anterior cerebral artery AC, right middle cerebral artery RC, left middle cerebral artery LC, right internal carotid artery RI, left internal carotid artery LI, The right posterior cerebral artery RP, the left posterior cerebral artery LP, the basilar artery BA), and each of the relative blood vessel lengths calculated by the abnormal blood vessel detection processing are shown as an example of inspection target values. Further, it is assumed that the same association is executed even if the normal value is reached (hereinafter, the associated normal value is “normal target value” ⁇ ⁇ ). The examination target value and the normal target value are obtained by associating each blood vessel part name with the rectification amount plotted in the special trace space. Note that the association and the process described below are executed by the control unit 11.
  • step S29 By comparing the test target value and the normal target value, it is detected which blood vessel site is likely to be occluded (step S29), and the possibility of occlusion occurs. If the blood vessel part name is detected, the detected blood vessel part name is notified to the operator. As notification means for the operator, the same processing as step S 17 in the above-described “blood vessel discrimination processing” is performed. Specifically, the display control is performed so that the annotation m is displayed only in the blood vessel part name where there is a high possibility of occlusion (see gl4 in FIG. 10). The blood vessel part name is displayed to the operator (step S30), and this process ends.
  • the control unit 11 calculates the relative blood vessel length of the target image, and uses the calculated relative blood vessel length to perform the discriminant analysis based on the normal value. It is. As a result of the discriminant analysis, the presence / absence of the occluded blood vessel and the name of the occluded blood vessel site are detected. By presenting the detected occluded blood vessel site name to an interpreting doctor or viewer, it is possible to shorten the interpretation time and prevent oversight of the occluded blood vessel when interpreting complex MRA images. .
  • the blood vessel discrimination processing by aligning the reference image and the target image in which the positions and names of the eight blood vessel parts included in the blood vessel image on the image are predetermined with respect to the blood vessel image, The position and name of each blood vessel part included in the target blood vessel image are determined. Then, this position and name information is attached to the target image as blood vessel part information.
  • each blood vessel part can be easily identified based on the blood vessel part information, and each blood vessel part can be identified and displayed. Therefore, the doctor can observe the target image while paying attention to a specific blood vessel site in the target image, and can improve the interpretation efficiency.
  • a target MIP image in which only the selected blood vessel part is extracted and displayed in accordance with the selection operation of an arbitrary blood vessel part among the identified blood vessel parts is generated and displayed.
  • the doctor can observe only the blood vessel site to be noticed. Therefore, duplication of a plurality of blood vessel sites can be eliminated, and observation can be performed by identifying a specific blood vessel site such as a site where multiple aneurysms occur.
  • the shape of the main blood vessel site (the length of the blood vessel, the extending direction, the thickness, etc.) is almost common even in different subjects (patients), but the shape of the non-main thin blood vessel site varies depending on the individual. Therefore, the shape of the blood vessel part varies depending on the subject.
  • each relative blood vessel length is calculated from the target blood vessel image, and each calculated relative blood vessel length is plotted in a two-dimensional feature space. It was decided to detect the presence or absence of occluded blood vessels and the name of the vascular site by comparing the plotted test target value with the normal target value, but this is not the only case, and the calculated test is not converted to a two-dimensional space.
  • a mode in which the occluded blood vessel site is detected by comparing the target value with a normal target value stored in advance in the storage unit 14 may be used.
  • an MRI image obtained by another imaging method such as a contrast MRA image obtained by imaging a blood vessel using a contrast agent may be used.
  • an image obtained by imaging a blood vessel with another imaging device such as CTA (Computed Tomography Angiography; DSA (Digital Subtraction Angiography)) may be used.
  • the present invention can be used in the medical field, and can be applied to a tomography apparatus that acquires a blood vessel image and a medical image analysis apparatus such as a view eyelid that analyzes the acquired blood vessel image.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Radiology & Medical Imaging (AREA)
  • Signal Processing (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Condensed Matter Physics & Semiconductors (AREA)
  • Artificial Intelligence (AREA)
  • Vascular Medicine (AREA)
  • Medical Informatics (AREA)
  • Quality & Reliability (AREA)
  • Theoretical Computer Science (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)

Abstract

L'invention concerne un processeur d'images médicales permettant de détecter automatiquement un candidat ayant un vaisseau sanguin anormal présentant une forte probabilité d'occlusion sur une image ARM. Une image de vaisseau sanguin présentant les noms de sites individuels est amincie, puis la longueur de chaque vaisseau sanguin est calculée. En fonction de la longueur de chaque vaisseau sanguin ainsi calculée, on calcule le rapport de la longueur du vaisseau sanguin par rapport à la longueur totale de vaisseaux sanguins. La comparaison du rapport ainsi calculé avec une valeur normale permet de détecter la présence/absence d'un vaisseau sanguin présentant une occlusion et d'un site de vaisseau sanguin présentant une occlusion.
PCT/JP2006/319304 2006-09-22 2006-09-28 Processeur d'images médicales, procédé et programme de traitement d'images associés WO2008035446A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2006-257617 2006-09-22
JP2006257617A JP3928977B1 (ja) 2006-09-22 2006-09-22 医用画像処理装置、画像処理方法及びプログラム

Publications (1)

Publication Number Publication Date
WO2008035446A1 true WO2008035446A1 (fr) 2008-03-27

Family

ID=38212161

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2006/319304 WO2008035446A1 (fr) 2006-09-22 2006-09-28 Processeur d'images médicales, procédé et programme de traitement d'images associés

Country Status (2)

Country Link
JP (1) JP3928977B1 (fr)
WO (1) WO2008035446A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2011027725A (ja) * 2009-06-25 2011-02-10 Olympus Corp 医療診断支援装置、バーチャル顕微鏡システムおよび標本支持部材
CN111820898A (zh) * 2019-04-10 2020-10-27 佳能医疗系统株式会社 医用图像处理装置以及医用图像处理系统

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4492645B2 (ja) * 2007-06-08 2010-06-30 富士フイルム株式会社 医用画像表示装置及びプログラム
JP4746068B2 (ja) * 2008-03-28 2011-08-10 一史 浦澤 画像解析装置及び画像解析方法
DE102010017630B4 (de) * 2010-06-29 2016-06-02 Leica Microsystems Cms Gmbh Verfahren und Einrichtung zur lichtmikroskopischen Abbildung einer Probenstruktur
JP2012115658A (ja) * 2010-11-12 2012-06-21 Toshiba Corp 画像診断装置
JP6433745B2 (ja) * 2014-10-07 2018-12-05 キヤノンメディカルシステムズ株式会社 医用画像処理装置およびｘ線診断装置
KR101793499B1 (ko) 2016-07-13 2017-11-06 연세대학교 산학협력단 Z-축 영상의 기하학적 정보를 이용한 대동맥 추출 방법
JP6871007B2 (ja) * 2017-02-13 2021-05-12 キヤノンメディカルシステムズ株式会社 医用画像処理装置及び医用画像診断システム
KR102223263B1 (ko) * 2018-09-04 2021-03-04 부산대학교 산학협력단 컴퓨터 보조 진단 시스템을 이용한 식도암 확대 내시경 영상 분석 방법 및 장치

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002325761A (ja) * 2000-06-30 2002-11-12 Hitachi Medical Corp 画像診断支援装置
JP2004133736A (ja) * 2002-10-11 2004-04-30 Hitachi Medical Corp 医用画像表示方法及びその装置

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002325761A (ja) * 2000-06-30 2002-11-12 Hitachi Medical Corp 画像診断支援装置
JP2004133736A (ja) * 2002-10-11 2004-04-30 Hitachi Medical Corp 医用画像表示方法及びその装置

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
TADA H. ET AL.: "Tobu 3D-MRA kara No Domyakuryo o Kenshutsu suru Algorithm no Kento", JAPANESE JOURNAL OF RADIOLOGICAL TECHNOLOGY SOKAI GAKUJUTSU TAIKAI YOKOSHU, vol. 62ND, 20 February 2006 (2006-02-20), pages 116, XP003021822 *
YAMAUCHI M. ET AL.: "No MRA Gazo ni Okeru Kekkan Chushutsu to Kekkanmei no Jido Taiozuke Shuho no Kaihatsu", JAPANESE JOURNAL OF RADIOLOGICAL TECHNOLOGY SOKAI GAKUJUTSU TAIKAI YOKOSHU, vol. 62ND, 20 February 2006 (2006-02-20), pages 116, XP003021821 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2011027725A (ja) * 2009-06-25 2011-02-10 Olympus Corp 医療診断支援装置、バーチャル顕微鏡システムおよび標本支持部材
CN111820898A (zh) * 2019-04-10 2020-10-27 佳能医疗系统株式会社 医用图像处理装置以及医用图像处理系统
CN111820898B (zh) * 2019-04-10 2024-04-05 佳能医疗系统株式会社 医用图像处理装置以及医用图像处理系统

Also Published As

Publication number Publication date
JP2008073332A (ja) 2008-04-03
JP3928977B1 (ja) 2007-06-13

Similar Documents

Publication Publication Date Title
WO2008035446A1 (fr) Processeur d'images médicales, procédé et programme de traitement d'images associés
Ali et al. A systematic review of automated melanoma detection in dermatoscopic images and its ground truth data
US8285013B2 (en) Method and apparatus for detecting abnormal patterns within diagnosis target image utilizing the past positions of abnormal patterns
US8958625B1 (en) Spiculated malignant mass detection and classification in a radiographic image
JP3928978B1 (ja) 医用画像処理装置、医用画像処理方法及びプログラム
US9383347B2 (en) Pathological diagnosis results assessment system, pathological diagnosis results assessment method, and pathological diagnosis results assessment device
CN102208105B (zh) 医学图像处理技术
EP2916737B1 (fr) Système et procédé de détection automatisée de nodules pulmonaires dans des images médicales
JP4026158B1 (ja) 読影支援装置、読影支援方法及びプログラム
US9373181B2 (en) System and method for enhanced viewing of rib metastasis
US20180140282A1 (en) Ultrasonic diagnostic apparatus and image processing method
US20080159604A1 (en) Method and system for imaging to identify vascularization
JP2008259622A (ja) レポート作成支援装置およびそのプログラム
Vos et al. Computational modeling for assessment of IBD: to be or not to be?
WO2011008906A1 (fr) Détection assistée par ordinateur (cad) d'anévrismes intracrâniens
WO2007026598A1 (fr) Processeur d’images médicales et procédé de traitement d’images
US20120027277A1 (en) Interactive iterative closest point algorithm for organ segmentation
US11200443B2 (en) Image processing apparatus, image processing method, and image processing system
Riaz et al. Impact of visual features on the segmentation of gastroenterology images using normalized cuts
JP2009226043A (ja) 医用画像処理装置及び異常陰影検出方法
CN115456990A (zh) 一种基于ct图像的肋骨计数方法、装置、设备及存储介质
JP2011250811A (ja) 医用画像処理装置及びプログラム
CN113658106A (zh) 一种基于腹部增强ct的肝脏病灶自动诊断系统
US7672496B2 (en) Forming three dimensional objects using a decision rule in medical image data
JP2006130049A (ja) 画像読影支援方法、およびシステム、ならびにプログラム

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 06810751

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: COMMUNICATION PURSUANT TO R112(1) EPC (EPOFORM 1205A) SENT 30.06.2009

122 Ep: pct application non-entry in european phase

Ref document number: 06810751

Country of ref document: EP

Kind code of ref document: A1