WO2017179256A1 - 核医学画像からの生理的集積の自動除去及びct画像の自動セグメンテーション - Google Patents
核医学画像からの生理的集積の自動除去及びct画像の自動セグメンテーション Download PDFInfo
- Publication number
- WO2017179256A1 WO2017179256A1 PCT/JP2017/001460 JP2017001460W WO2017179256A1 WO 2017179256 A1 WO2017179256 A1 WO 2017179256A1 JP 2017001460 W JP2017001460 W JP 2017001460W WO 2017179256 A1 WO2017179256 A1 WO 2017179256A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- value
- slice
- region
- image
- class
- Prior art date
Links
- 238000009206 nuclear medicine Methods 0.000 title claims abstract description 66
- 238000009825 accumulation Methods 0.000 title claims abstract description 26
- 230000011218 segmentation Effects 0.000 title claims abstract description 11
- 230000035508 accumulation Effects 0.000 title abstract description 24
- 210000004872 soft tissue Anatomy 0.000 claims abstract description 23
- 210000000988 bone and bone Anatomy 0.000 claims abstract description 20
- 230000003187 abdominal effect Effects 0.000 claims abstract description 19
- 238000000034 method Methods 0.000 claims description 148
- 230000008569 process Effects 0.000 claims description 99
- 210000001015 abdomen Anatomy 0.000 claims description 53
- 210000003141 lower extremity Anatomy 0.000 claims description 43
- 238000003860 storage Methods 0.000 claims description 27
- 230000008859 change Effects 0.000 claims description 24
- 210000000689 upper leg Anatomy 0.000 claims description 15
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 13
- 238000002372 labelling Methods 0.000 claims description 12
- 210000004197 pelvis Anatomy 0.000 claims description 12
- 230000001936 parietal effect Effects 0.000 claims description 10
- 238000001514 detection method Methods 0.000 claims description 7
- 238000009499 grossing Methods 0.000 claims description 7
- 230000000873 masking effect Effects 0.000 claims description 5
- 210000004556 brain Anatomy 0.000 claims description 3
- 210000002414 leg Anatomy 0.000 claims description 3
- 238000004590 computer program Methods 0.000 claims description 2
- 230000001419 dependent effect Effects 0.000 claims 3
- 230000002349 favourable effect Effects 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 6
- 230000002093 peripheral effect Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 3
- 230000017531 blood circulation Effects 0.000 description 2
- 210000000746 body region Anatomy 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 210000004072 lung Anatomy 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 230000004060 metabolic process Effects 0.000 description 2
- 229940121896 radiopharmaceutical Drugs 0.000 description 2
- 239000012217 radiopharmaceutical Substances 0.000 description 2
- 230000002799 radiopharmaceutical effect Effects 0.000 description 2
- 101100129500 Caenorhabditis elegans max-2 gene Proteins 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000001816 cooling Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000005429 filling process Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 230000003902 lesion Effects 0.000 description 1
- 235000004213 low-fat Nutrition 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000004393 prognosis Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000002603 single-photon emission computed tomography Methods 0.000 description 1
- 230000001225 therapeutic effect Effects 0.000 description 1
- 210000001519 tissue Anatomy 0.000 description 1
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
- A61B6/5205—Devices using data or image processing specially adapted for radiation diagnosis involving processing of raw data to produce diagnostic data
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/02—Arrangements for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
- A61B6/03—Computed tomography [CT]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/02—Arrangements for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
- A61B6/03—Computed tomography [CT]
- A61B6/032—Transmission computed tomography [CT]
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/02—Arrangements for diagnosis sequentially in different planes; Stereoscopic radiation diagnosis
- A61B6/03—Computed tomography [CT]
- A61B6/037—Emission tomography
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
- A61B6/5211—Devices using data or image processing specially adapted for radiation diagnosis involving processing of medical diagnostic data
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
- A61B6/52—Devices using data or image processing specially adapted for radiation diagnosis
- A61B6/5211—Devices using data or image processing specially adapted for radiation diagnosis involving processing of medical diagnostic data
- A61B6/5229—Devices using data or image processing specially adapted for radiation diagnosis involving processing of medical diagnostic data combining image data of a patient, e.g. combining a functional image with an anatomical image
- A61B6/5247—Devices using data or image processing specially adapted for radiation diagnosis involving processing of medical diagnostic data combining image data of a patient, e.g. combining a functional image with an anatomical image combining images from an ionising-radiation diagnostic technique and a non-ionising radiation diagnostic technique, e.g. X-ray and ultrasound
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01T—MEASUREMENT OF NUCLEAR OR X-RADIATION
- G01T1/00—Measuring X-radiation, gamma radiation, corpuscular radiation, or cosmic radiation
- G01T1/16—Measuring radiation intensity
- G01T1/161—Applications in the field of nuclear medicine, e.g. in vivo counting
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01T—MEASUREMENT OF NUCLEAR OR X-RADIATION
- G01T1/00—Measuring X-radiation, gamma radiation, corpuscular radiation, or cosmic radiation
- G01T1/29—Measurement performed on radiation beams, e.g. position or section of the beam; Measurement of spatial distribution of radiation
- G01T1/2914—Measurement of spatial distribution of radiation
- G01T1/2985—In depth localisation, e.g. using positron emitters; Tomographic imaging (longitudinal and transverse section imaging; apparatus for radiation diagnosis sequentially in different planes, steroscopic radiation diagnosis)
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H50/00—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics
- G16H50/20—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for computer-aided diagnosis, e.g. based on medical expert systems
-
- 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
- G06T2207/10081—Computed x-ray tomography [CT]
-
- 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
- G06T2207/10104—Positron emission tomography [PET]
-
- 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
Definitions
- This application relates to the automatic removal of physiological accumulation from nuclear medicine images and the automatic segmentation of CT images.
- Nuclear medicine technology is a technology that visualizes how radiopharmaceuticals administered to the body gather in organs and tissues. Functional changes such as blood flow and metabolism can be obtained as image information, which can provide useful information for diagnosing diseases, confirming stage and prognosis, and determining therapeutic effects.
- Paragraphs 0027 and 0029 of Japanese Patent Laid-Open No. 2013-088386 describe that only the portion corresponding to the femur is extracted from the PET image using the position information of the femur extracted from the CT image.
- radiopharmaceuticals accumulate in areas with high blood flow and metabolism, such as the brain and bladder, regardless of the disease. Since these physiological accumulations are displayed brightly on nuclear medicine images, they may hinder image diagnosis.
- the present application discloses an invention for automatically removing physiological accumulation from a nuclear medicine image. Also disclosed is an invention that performs automatic segmentation of CT images and can be used for automatic removal of physiological accumulation from nuclear medicine images as a preferred application example.
- the first invention is an invention for automatically classifying pixel values of CT images.
- This invention Creating a histogram of pixel values based on the CT image; Determining a fat region peak that is the peak of the fat region in the histogram; Determining a soft region peak that is the peak of the soft tissue region in the histogram; Determining a first threshold that is a threshold representing an upper limit of the class value of the air region based on the frequency value of the fat region peak in a region having a class value smaller than the class value of the fat region peak; Determining a second threshold that is a threshold representing a class value of the boundary between the fat region and the soft tissue region; In a region where the class value is larger than the class value corresponding to the soft part region peak, a third threshold value that is a threshold value representing the lower limit of the bone region class value is determined based on the frequency value of the soft part region peak.
- a third threshold value that is a threshold value representing the lower limit of the bone region class value is determined
- the pixel values of the CT image can be classified into the pixel value of the air region, the pixel value of the fat region, the pixel value of the soft tissue region, and the pixel value of the bone region.
- the CT image can be displayed flexibly.
- only the soft tissue region can be highlighted and displayed, or only the bone region can be highlighted and displayed.
- the first invention is also very useful for automatically performing CT image region segmentation. For example, a region where there are many pixels whose pixel values are air regions is highly likely to be a lung field. In addition, one of the regions where many pixels having the pixel value of the bone region exist is highly likely to be the pelvis, for example.
- creating the histogram includes Creating a reference image, which is a binarized image in which only the human body part is left by removing the bed part from the CT image; Creating the histogram using only the pixels of the CT image that have data in the corresponding pixels of the reference image; May be included.
- the histogram may be smoothed before the fat region peak and the soft region peak are determined.
- the determination of the fat region peak is to obtain a maximum frequency value in a range from a lower class value set so as to include the fat region peak to a reference class value that is a class value corresponding to water. If the maximum frequency value is not the frequency value of the reference class value, it may include determining that the maximum frequency value and the corresponding class value are the frequency value and the class value of the fat region peak. The process may further include determining the second threshold as a class value that minimizes a frequency value in a range from the class value of the fat region peak to the reference class value.
- determining the fat region peak may be determined to be the frequency value and class value of the fat region peak.
- the series of processes makes it possible to determine the fat region peak in most cases.
- determining the soft region peak may include performing a peak detection process within a predetermined class value range.
- the first threshold is A class value whose frequency value is less than or less than a predetermined percentage of the frequency value of the fat region peak in the range from the lower limit class value to the second threshold value; A class value in which the amount of change is maximum in a range from the lower limit class value to the class value of the fat region peak; It may be determined as either of these.
- the third threshold value is a class value that is larger than the class value corresponding to the soft part region peak
- the frequency value is a predetermined value of the frequency value of the soft part region peak. May be determined.
- a second invention is an invention for automatically performing region division of the CT image in which the pixel values are classified according to the first invention and the first to third threshold values are determined.
- This invention The slice number of the body axis cross-sectional slice of the CT image is taken on one axis, and the pixel values in the slice corresponding to each slice number on the other axis Creating an air region volume graph, which is a graph that takes at least a portion of the volume; A body axis cross-section slice on the parietal side of the body axis cross-section slice in which the air region volume graph exhibits the maximum value, the volume value of which is a predetermined ratio of the maximum value of the soft tissue volume graph To be the chest start slice located at the top of the chest; It is characterized by performing the process including.
- the second invention it is possible to automatically determine the chest start point slice located at the upper end of the chest in the body axis cross-sectional slice of the CT image. Therefore, for example, it is possible to easily grasp the position of the body in the CT image. In addition, when there is a nuclear medicine image registered with the CT image, it is possible to easily grasp the position of the start slice of the chest in the nuclear medicine image.
- the process may include performing a smoothing process on the air region volume graph before determining the chest start point slice.
- the process is The slice number of the body axis slice of the CT image is taken as one axis, and the pixel value in the slice corresponding to each slice number is changed from the second threshold value to the third threshold value as the other axis.
- Creating a soft region volume graph which is a graph that takes the volume of a group of pixels in between; -The body axis cross-sectional slice on the lower limb side of the body axis cross-sectional slice where the air region volume graph exhibits the maximum value, and the body axis cross-sectional slice where the volume value of the soft part region volume graph is the largest at the upper end of the upper abdomen Determining that it is a located upper abdominal starting slice; May be included.
- the process may include performing a smoothing process on the soft region volume graph before determining the upper abdomen start slice.
- the processing is a body-axis cross-sectional slice on the parietal side of the chest start point slice, and the body-axis cross-sectional slice with the largest volume value in the air region volume graph is positioned at the upper end of the neck. Determining to be a cervical starting point slice.
- the process is Performing 3D labeling on the body axis cross-sectional slice on the parietal side of the neck start point slice;
- a head label which is a centrally located label in the body axis slice; A pixel group in a region corresponding to the head label, the volume of the pixel group having a pixel value between the second threshold value and the third threshold value, from the neck start point slice to the top of the head. Calculating for the axial slice and determining that the axial slice whose volume is initially zero is the head start slice located at the top of the head; May be included.
- the above process determines that the lower abdominal start slice is located at the upper end of the lower abdomen based on the change in the amount of bone for the lower limb side axial slice of the upper abdomen start slice. It may further include.
- determining the lower abdominal starting slice is: Performing 3D labeling on the body axis cross-sectional slice on the lower limb side with respect to the upper abdomen starting point slice to determine the trunk label that is the largest label; A pixel group in a region corresponding to the body label with respect to a body axis cross-sectional slice on the lower limb side with respect to the upper abdomen start point slice, the pixel value being greater than or equal to the third threshold value or greater than the third threshold value Extracting a pixel group and performing 3D labeling on the extracted pixel group to determine the largest label of the spine, pelvis, and femur; A pixel group corresponding to the spine, pelvis, and femur label, and a circumscribed rectangle of a pixel group having a pixel value greater than or equal to the third threshold value or greater than the third threshold value is lower than the upper abdominal start point slice. Creating for each of the lateral axial slices; Determining the body axis cross-sectional slice with the largest amount of change in
- the above processing further includes: A pixel group in a region corresponding to the trunk label in the lower limb side axial slice from the lower abdomen starting point slice, the pixel value of which is greater than or equal to the third threshold value or greater than the third threshold value; Calculating the maximum volume; Of the body axis cross-sectional slices on the lower limb side from the lower abdomen starting point slice, the femoral volume value is a predetermined ratio of the maximum value in the body axis cross-sectional slices for which the femoral volume can be calculated Determining that the first slice is the leg starting slice located at the upper end of the leg; May be included.
- a body axis slice of a CT image for example, a slice located at the upper end of the chest, a slice located at the upper end of the upper abdomen, a slice located at the upper end of the neck, the upper end of the head It is possible to determine where the slice located at the upper end of the lower abdomen, the slice located at the upper end of the lower limb is located. Therefore, for example, if these slice positions are displayed, it is possible to easily grasp the position of the body in the CT image. In addition, when there is a nuclear medicine image that has been registered with a CT image, it is also easy to grasp these body positions in the nuclear medicine image.
- the third invention is an invention for automatically removing physiological accumulation from a nuclear medicine image.
- the present invention uses the information on the slice position of the head start point slice and the information on the slice position of the neck start point slice determined by the embodiment of the second invention.
- the present invention also provides In the nuclear medicine image, setting a maximum pixel value search region using the slice position of the head start slice and the slice position of the neck start slice; Determining a pixel having the maximum pixel value in the maximum pixel value search area; Determining a highly integrated region of the head from the pixel having the maximum pixel value by a region growing method; It is characterized by performing the process including.
- the highly integrated region of the head is automatically determined from the nuclear medicine image, it is easy to observe other lesions in the nuclear medicine image. For example, if this region is masked to display this nuclear medicine image, the nuclear medicine image can be observed without being obstructed by unnecessary signals from the head.
- the fourth invention is another invention for automatically removing physiological accumulation from a nuclear medicine image.
- This invention uses the information on the slice position of the lower abdomen starting point slice and the information on the slice position of the lower limb starting point slice, which are determined by the embodiment of the second invention.
- the present invention also provides -In the nuclear medicine image, using the slice position of the lower abdomen starting point slice and the slice position of the lower limb starting point slice, setting a maximum pixel value search region; Determining a pixel having the maximum pixel value in the maximum pixel value search area; Determining a highly integrated region of the bladder from a pixel having the maximum pixel value by a region growing method; It is characterized by performing the process including.
- the highly integrated region of the bladder is automatically determined from the nuclear medicine image, it is easy to observe the nuclear medicine image. For example, if this nuclear medicine image is displayed by masking this area, the nuclear medicine image can be observed without being obstructed by unnecessary signals from the bladder.
- a computer program comprising program instructions configured to cause the apparatus to perform the above processing when executed by the processing means of the apparatus.
- program instructions configured to cause the apparatus to perform the above processing when executed by the processing means of the apparatus.
- method performed by the apparatus by executing a program command by the processing means of the apparatus, which includes performing the above processing.
- an apparatus comprising processing means and storage means for storing a program instruction.
- the apparatus When the program instruction is executed by the processing means, the apparatus performs the above processing.
- the apparatus There is an apparatus configured to let
- FIG. 3 is a flowchart for explaining a specific example of processing in step 212 in FIG. 2. It is a figure for demonstrating the bed part removal process of step 306. FIG. It is an example of a histogram created in Step 308. It is a flowchart for demonstrating the process 600 which is a specific example of the process of step 310 of FIG. It is a figure for demonstrating the investigation class value determined by step 614.
- FIG. FIG. 1 is a flowchart for demonstrating the flow of the process 200 which performs the automatic segmentation of CT image and the automatic removal of the physiological accumulation
- 3 is a flowchart for explaining a specific example of processing in step 212 in FIG. 2. It is a figure for demonstrating the bed part removal process of step 306. FIG. It is an example of a histogram created in Step 308. It is a flowchart for demonstrating the process 600 which is a specific example of the process of step 310
- FIG. 6 is a diagram schematically representing a relationship between a histogram, a fat region peak, a soft region region peak, and first to third threshold values.
- 4 is a flowchart for explaining a process 800, which is a specific example of the process in step 214 of FIG. It is a display example of a graph created in Step 806.
- FIG. 10 is a diagram showing the positions of the chest start point slice, upper abdominal start point slice, and neck start slice determined in the example of Steps 808-812 on the graph of FIG.
- FIG. 10 is a diagram for explaining the processing of step 814.
- 10 is a flowchart for explaining a process 1200, which is a specific example of step 816 in FIG. It is a figure for demonstrating the process of step 1204.
- FIG. 14B is an example in which circumscribed rectangles are superimposed and displayed on the spine / pelvis / femur labels of FIG. 14A.
- FIG. 10 is a flowchart for explaining a process 1500, which is a specific example of step 818 in FIG.
- FIG. 10 is a diagram for explaining the processing in step 1508. It is a flowchart for demonstrating the process 1700 which is one of the specific examples of the process of step 216 of FIG. A display example of the result is shown.
- FIG. 1 is a diagram for explaining a hardware configuration of a system 100 that can implement the present invention.
- the system 100 is similar to a general computer in hardware, and includes a CPU 102, a main storage device 104, a mass storage device 106, a display interface 107, a peripheral device interface 108, a network.
- An interface 109 or the like can be provided.
- a high-speed RAM Random Access Memory
- an inexpensive and large-capacity hard disk or SSD can be used as the large-capacity storage device 106. it can.
- a display for displaying information can be connected to the system 100, which is connected via a display interface 107.
- a user interface such as a keyboard, a mouse, and a touch panel can be connected to the system 100, and this is connected via the peripheral device interface 108.
- the network interface 109 can be used to connect to another computer or the Internet via a network.
- the mass storage device 106 stores an operating system (OS) 110, a registration program 120, a CT image segmentation program 122, and a physiological accumulation removal program 124.
- the most basic functions of the system 100 are provided by the OS 110 being executed by the CPU 102.
- the registration program 120 is a program for performing registration (registration) between a CT image and a nuclear medicine image.
- the CT image segmentation program 122 includes program instructions relating to the novel processing disclosed by the present application. When at least a part of the instructions are executed by the CPU 102, the system 100 can perform segmentation of CT images, that is, automatic segmentation of body regions.
- the physiological accumulation removal program 124 also includes program instructions relating to the novel processing disclosed by the present application. When at least a part of the instructions are executed by the CPU 102, the system 100 can automatically remove the physiological accumulation from the nuclear medicine image.
- the mass storage device 106 can further store a CT image 130 and a nuclear medicine image 132.
- the CT image 130 is three-dimensional image data in which each pixel value corresponds to a CT value, and is image data to be analyzed or operated by the CT image segmentation program 122.
- the nuclear medicine image 132 is a PET image in this example, and is three-dimensional image data in which each pixel value corresponds to a radiation count value.
- the nuclear medicine image 132 is subjected to analysis or manipulation by the physiological accumulation removal program 124.
- the data 140, 142, 144, etc. may be further stored in the mass storage device 106. These data will be described later when these data are generated.
- the system 100 can have the same configuration as an apparatus included in a normal computer system, such as a power supply or a cooling device.
- the implementation form of the computer system includes the distribution / redundancy and virtualization of storage devices, the use of multiple CPUs, CPU virtualization, the use of a processor specialized for decision processing such as DSP, and the decision processing in hardware.
- Various forms using various techniques, such as combining, are known.
- the invention disclosed in the present application may be mounted on any form of computer system, and the scope thereof is not limited by the form of the computer system.
- the technical idea disclosed in this specification is generally (1) executed by a processing unit to cause an apparatus or a system including the processing unit to perform various processes described in this specification. (2) an operation method of an apparatus or a system realized by the processing means executing the program, and (3) the program and the program.
- the present invention can be embodied as an apparatus or system provided with processing means. As described above, some software processing may be implemented as hardware.
- the CT image 130, the PET image 132, and various data 140, 142, 144, etc. are often not stored in the mass storage device 106 at the time of manufacture and sale of the system 100 or at the time of activation.
- the CT image 130 and the PET image 132 may be data transferred from an external device to the system 100 via the peripheral device interface 108 or the network interface 109, for example.
- the data 140, 142, and 144 may be data stored in the mass storage device 106 as a result of at least a part of the CT image segmentation program 122 and the physiological accumulation removal program 124 being executed by the CPU 102. It should be noted that the scope of the invention disclosed in the present application is not limited by whether image data or the like is stored in the storage device.
- the process 200 can be a process performed by the system 100 when the CPU 102 executes a CT image segmentation program 122, a physiological accumulation removal program 124, or the like, for example.
- Step 204 indicates the start of processing.
- step 208 registration (registration) between the nuclear medicine image 132 and the CT image 130 is performed. That is, the direction, size, and position of the body are three-dimensionally matched between the CT image 130 and the nuclear medicine image 132. Through this process, the CT image 130 and the nuclear medicine image 132 can be compared with each other.
- the registration process may be a process performed by the system 100 when the registration program 120 is executed by the CPU 102.
- Registration may be by a manual method. That is, the CT image and the nuclear medicine image may be displayed together as an image, and one image may be translated or rotated by an operation with a mouse or the like to match the other image.
- the image data 130 and 132 have been registered.
- the registration between the image data 130 and 132 may have already been completed.
- the registration of the CT image 130 and the nuclear medicine image 132 is output from the apparatus. May have already been completed.
- the process of step 208 is unnecessary.
- registration program 120 may also be unnecessary.
- step 212 as pre-processing in step 214, the pixel value of the CT image 130 is divided into an air region, a fat region, a soft tissue region, and a bone region.
- An example of the process of step 212 will be described in detail later with reference to FIG.
- step 214 the body part in the CT image 130 is specified using the result of step 212. Specifically, a slice located at a specific part is determined from the body axis cross-sectional slices of the CT image 130.
- An example of the process of step 214 will be described in detail later with reference to FIGS.
- step 216 the physiological accumulation in the nuclear medicine image 132 is specified based on the result of step 212.
- An example of the processing in step 216 will be described in detail later with reference to FIG.
- step 220 based on the result of step 216, a nuclear medicine image 132 is displayed by masking a portion with physiological accumulation.
- a display example is shown in FIG.
- Step 224 indicates the end of the process.
- the process 300 may be a process performed by the system 100 when at least a part of program instructions included in the CT image segmentation program 122 is executed by the CPU 102.
- Step 302 indicates the start of processing.
- step 304 data to be processed by the CT image segmentation program 122 is read (loaded). That is, all or part of the CT image 130 is read from the mass storage device 106 and stored in the main storage device 104. In some embodiments, the CT image 130 may be copied directly from an external device to the main storage device 104 via the peripheral device interface 108 or the network interface 109.
- the CT image 130 that is the processing target of the processing 300 is corrected in advance so that the pixel value representing water becomes zero. For this reason, some pixels have negative pixel values.
- Step 306 represents a step for creating a reference image used for removing the bed portion. Step 306 is performed as follows.
- Step 306A First, a binary image obtained by binarizing the CT image 130 is created.
- the threshold value for binarization may be a threshold value at which a human body is extracted as much as possible.
- a binary image may be created with -190 as a threshold value.
- An example of the created binarized image is shown in FIG. 4A.
- Step 306B Next, search for a place where data exists in the upper and lower slices in the body axis direction from the center of the created binarized image, and create a binarized image from which the bed portion has been removed by performing region growing processing therefrom. To do.
- FIG. 4B shows a binarized image created by processing the binarized image of FIG. 4A in this way.
- Step 306C After removing the bed, a filling process is performed to put a value into the hole in the body, and the bed that could not be removed by opening the morphology operation is removed.
- An image created by processing the binarized image of FIG. 4B in this way is shown in FIG. 4C.
- the image obtained by the processing in step 306C is set as a reference image.
- a histogram of the pixel values of the CT image 130 is created.
- a histogram is created using only pixels in which data exists in the corresponding pixels of the reference image created in step 306 among the pixels of the CT image 130.
- the entire CT image 130 may be used to create a histogram.
- An example of the created histogram is shown in FIG.
- the CT image 130 is corrected in advance so that the pixel value representing water becomes zero.
- the CT image 130 is standardized in advance so that the range of pixel values is distributed in the range of -1024 to +1024. For this reason, the range of the horizontal axis (class value) of the illustrated histogram is also from -1024 to +1024.
- the histogram of a CT image has two peaks near the class value corresponding to water (class value 0 in this example).
- the peak of the class value lower than the class value corresponding to water is a peak created by pixels corresponding to the fat region, and is referred to as a fat region peak in this specification.
- “Max1” is displayed.
- the peak of the class value higher than the class value corresponding to water is a peak formed by pixels corresponding to soft tissue (muscle, brain, etc.), and is referred to as a soft part region peak in this specification.
- “Max2” is displayed. In the case of a subject with low fat, the peak of the fat region may not be clear.
- the class value around the fat region peak is considered to correspond to the fat region.
- this area is displayed as Fat.
- the class value around the soft part region peak is considered to correspond to soft tissue.
- this area is displayed as Soft.
- a lower class value than the fat region is considered to correspond to the air region.
- this area is displayed as Air.
- An area where there are many pixels whose CT values are class values of the air area may be a lung field.
- a higher class value than the soft tissue region is considered to correspond to bone.
- this area is displayed as Bone.
- the boundary between the air region and the fat region, the boundary between the fat region and the soft tissue region, and the boundary between the soft tissue region and the bone region are indicated as th1, th2, and th3 in FIG. In subsequent steps, the fat region peak, soft region region peak, th1, th2, and th3 are automatically determined.
- smoothing processing may be performed on the histogram before proceeding from step 308 to step 310.
- step 310 the above-described fat region peak and a second threshold value (th2 depicted in FIG. 5) that is a threshold value representing the boundary between the fat region and the soft tissue region are determined.
- a second threshold value depicted in FIG. 5
- An example of the processing in step 310 will be described with reference to FIG. 6A.
- FIG. 6A is a flowchart for explaining a process 600, which is a specific example of the process of step 310.
- the process 600 can be a process performed by the system 100 when at least a part of program instructions included in the CT image segmentation program 122 is executed by the CPU 102.
- Step 602 indicates the start of processing.
- a first attempt is made to determine a fat region peak in the histogram created in step 308.
- the peak detection process does not need to be performed over the entire range of class values, and may be performed in a class value range that is empirically known to be a fat region peak. For example, in the case of a CT image in which the pixel value corresponding to water is 0 and the pixel value range is corrected / standardized from ⁇ 1024 to +1024, the fat region peak of the histogram is almost always in the range of the class value ⁇ 256 to 0.
- the lower limit class value is set to, for example, ⁇ 256
- the upper limit class value is set to, for example, 0 (that is, the class value corresponding to water)
- the fat region peak is searched in this range. May be.
- step 604 the maximum frequency value in this search range is searched.
- this maximum frequency value is not the frequency value in the class value corresponding to water (class value 0 in this example)
- the maximum frequency value and the corresponding class value are the frequency value and class value of the fat region peak. decide.
- the maximum frequency value of the search range may be a frequency value at the class value 0.
- the fat region peak is determined using another method.
- step 606 it is determined whether a fat region peak has been determined. If it has been determined, the process proceeds to step 608 to determine a second threshold that is a threshold representing the boundary between the fat region and the soft tissue region. In step 608, the second threshold value is determined as the class value that minimizes the frequency value in the range from the class value of the fat region peak to the reference class value (class value 0 in this example). Thereafter, the process 600 proceeds to step 622 and ends the process.
- step 606 If it is determined in step 606 that the fat region peak has not been determined, the process proceeds to step 610.
- the second threshold value which is a threshold value that represents a boundary between the fat region and the soft tissue region, is determined as a class value that minimizes the amount of change in the range from the lower limit class value to the reference class value. To do.
- a survey class value that is a lower class value for searching for a fat region peak is determined.
- This survey class value is determined as one of the following: (A) A class value that maximizes the amount of change calculated in step 610 in the range from the lower limit class value set in step 604 to the second threshold value determined in step 612. (B) A class value corresponding to a frequency value that first checks a frequency value change from the second threshold value determined in step 612 toward the lower limit class value and first falls below the frequency value in the second threshold value.
- the state of the survey class value defined in (b) is shown in FIG. 6B.
- step 616 a second attempt is made to determine the fat region peak.
- the class value and the corresponding frequency value at which the “previous difference” value calculated in step 610 is positive and minimum are obtained as fat region peak values. Determine class and frequency values.
- step 618 it is determined whether a fat region peak has been determined. If it is determined, the process proceeds to step 622 and the process is terminated. However, in step 616, if a fat region peak cannot be determined, such as when a class value having a positive “previous difference” is not found, the process proceeds to step 620.
- step 620 a third attempt is made to determine the fat region peak.
- the maximum frequency value and the corresponding class value in the range from the lower limit class value set in step 604 to the second threshold value determined in step 612 are determined as the fat region peak frequency value and class value. Determine that there is. If the third trial is performed, the fat region peak and the second threshold can be determined without fail.
- the process 600 proceeds to step 622 and ends the process.
- the soft part region peak is determined.
- the soft region peak detection process need not be performed over the entire range of class values, and may be performed within the class value range that is empirically known to exist as a soft region peak.
- the soft region peak detection process in step 312 may be performed in the range of the class value 0 to +256.
- the soft region peak can be determined by this detection process.
- a first threshold value (th1 depicted in FIG. 5) that is a threshold value that represents the boundary between the air region and the fat region is determined.
- a frequency value in a range from the lower limit class value set in step 604 to the second threshold value determined in step 310 is a predetermined value of the fat region peak frequency value.
- Use class values that are less than or less than the percentage.
- the predetermined ratio the inventor has confirmed that 10% is one of appropriate values, for example. However, other values such as 5% may be used. You may determine for every CT apparatus using the data of a some test subject.
- a condition that the absolute value of the frequency value difference from the immediately preceding class value is equal to or less than a predetermined value may be included in the first threshold determination condition.
- a predetermined value the inventor has confirmed that, for example, 50 is one of appropriate values. However, this value is merely an example, and various values can be taken depending on the embodiment.
- the first threshold value is the same as that calculated in step 610 in the range from the lower limit class value set in step 604 to the second threshold value determined in step 310. A class value that maximizes the “change amount” may be used.
- the first threshold may be determined using the condition 314c.
- a third threshold value (th3 depicted in FIG. 5) that is a threshold value representing the boundary between the fat region and the bone region is determined.
- this threshold is determined to be a class value that is larger than the class value corresponding to the soft part region peak
- the frequency value is a class value that is a predetermined ratio of the frequency value of the soft part region peak. May be.
- the inventor has confirmed that, for example, 5% is an appropriate value as the predetermined ratio. However, other values such as 10% may be used. You may determine for every CT apparatus using the data of a some test subject.
- the first to third threshold values are stored in the storage device.
- the threshold data 140 may be stored in the mass storage device 106.
- step 320 represents the end of process 300.
- the process 800 can be a process performed by the system 100 when at least a part of program instructions included in the CT image segmentation program 122 is executed by the CPU 102.
- Step 802 indicates the start of processing.
- data to be processed by the CT image segmentation program 122 is read (loaded).
- the data loaded here is the CT image 130 (all or a part thereof) as in step 304.
- the threshold data 140 created and stored in step 212 is also read.
- the CT image 130 and / or the threshold data 140 are already stored in the main storage device 102 and are not loaded again. In this case, the process of step 804 will be unnecessary.
- step 806 the volume of pixels belonging to the air region, fat region, soft tissue, and bone region classified in the process 300 is calculated for each body axis slice of the CT image, and graph data is created.
- FIG. 9 shows an example in which an example of the created graph is displayed.
- the horizontal axis of the graph is the slice number of the body axis slice of the CT image 130. In this example, the smaller number is the head side, and the larger number is the lower limb side.
- the vertical axis of the graph is the volume (cm 3 ). The vertical axis may simply be the number of pixels.
- the graph displayed as Air in FIG. 9 represents the volume (or the number of pixels) of the pixels belonging to the air region.
- the pixel belonging to the air region is a pixel whose pixel value is less than or less than the first threshold value. See the description of step 314 for the first threshold. In the present specification, this graph is referred to as an air region volume graph.
- the graph displayed as Fat in FIG. 9 represents the volume (or the number of pixels) of the pixels belonging to the fat region.
- the pixel belonging to the fat region is a pixel whose pixel value is included in the range from the first threshold value to the second threshold value.
- This range is a boundary (that is, a first threshold value or a second threshold value). It depends on the embodiment.
- this graph is referred to as a fat region volume graph.
- the graph displayed as Soft in FIG. 9 represents the volume (or the number of pixels) of the pixels belonging to the soft tissue.
- the pixel belonging to the soft tissue is a pixel whose pixel value is included in the range from the second threshold value to the third threshold value. Whether this range includes a boundary (that is, the second threshold or the third threshold) depends on the embodiment. Refer to the description of steps 314 and 316 for the second threshold and the third threshold. In this specification, this graph is referred to as a soft region volume graph.
- the graph displayed as Bone represents the volume (or the number of pixels) of the pixels belonging to the bone region.
- the pixel belonging to the bone region is a pixel whose pixel value is greater than or equal to the third threshold value or greater than the third threshold value. See the description of step 316 for the third threshold.
- this graph is referred to as a bone region volume graph.
- a smoothing process may be performed on the graph depicted in FIG. 9 and subsequent steps may be performed.
- a chest start point slice which is a body axis cross-sectional slice located at the upper end of the chest in the CT image 130 is determined.
- the chest start point slice is a body axis cross-sectional slice on the parietal side of the body axis cross-sectional slice where the air region volume graph exhibits the maximum value
- the volume value is a predetermined ratio of the maximum value of the air region volume graph. It is determined that the slice is a body axis slice.
- the predetermined ratio for example, the inventor has confirmed that 10% is a suitable value, but is not limited to this value and can be changed according to the embodiment.
- a condition for determining the chest start point slice a condition may be added in which a frequency difference with an adjacent slice is a predetermined ratio of the maximum value of the air region volume graph.
- the predetermined ratio for example, the inventor has confirmed that 1% is a suitable value, but is not limited to this value, and can be changed according to the embodiment.
- an upper abdomen start point slice located at the upper end of the upper abdomen in the CT image 130 is determined.
- the upper abdomen starting point slice is a lower-limb body axis slice than the body axis section slice in which the air region volume graph exhibits the maximum value, and the body axis section slice in which the volume value of the soft region volume graph is the largest It is determined that
- a neck start point slice located at the upper end of the neck in the CT image 130 is determined.
- the neck start point slice is determined to be the body axis cross-sectional slice on the parietal side of the chest start point slice and the body axis cross-sectional slice having the largest volume value in the air region volume graph.
- the positions of the chest start slice determined in step 808, the upper abdominal start slice determined in step 810, and the neck start slice determined in step 812 are shown on the graph of FIG. Put it on.
- a head start point slice located at the upper end of the head in the CT image 130 is determined. This determination process is performed as follows. (814a) 3D labeling is performed on the body axis cross-sectional slice on the parietal side of the neck start point slice. (814b) A head label, which is a label located at the center of the body axis slice, is determined. (814c) The volume of the pixel group in the region corresponding to the head label, the pixel value of which is between the second threshold value and the third threshold value, is transferred from the cervical starting point slice to the parietal side. The calculation is performed on the cross-sectional slice, and the body-axis cross-sectional slice whose volume is initially zero is determined to be the head start point slice.
- FIG. 11 shows an example of a head label located in a certain body axis slice.
- step 816 in the CT image 130, a lower abdomen start point slice located at the upper end of the lower abdomen is determined. An example of this determination process will be described with reference to FIGS. 12-14.
- the process 1200 can also be a process performed by the system 100 by causing the CPU 102 to execute at least a part of program instructions included in the CT image segmentation program 122.
- Step 1202 indicates the start of processing.
- step 1204 3D labeling is performed on the body axis cross-sectional slice on the lower limb side from the upper abdominal starting point slice determined in step 810 among the body axis cross-sectional slices of the CT image 130, and the body label which is the largest label is obtained. decide.
- the state of the determined body label is shown in FIG. 13 by taking a slice of a body axis as an example.
- a pixel group in a region corresponding to the body label with respect to the body axis cross-sectional slice on the lower limb side from the upper abdomen starting point slice the pixel value of which is greater than or equal to the third threshold value or greater than the third threshold value Extract groups. See the description of step 316 for the third threshold.
- the extracted pixel group indicates a bone region on the lower limb side of the upper abdomen starting point slice.
- 3D labeling is performed on the extracted pixel group to determine the largest label, the spine / pelvis / femur label.
- the state of the determined spine / pelvis / femur label is shown in FIG. 14A, taking a slice of a body axis as an example.
- a circumscribed rectangle of a pixel group corresponding to the spine, pelvis, and femur label, the pixel value of which is greater than or equal to the third threshold value or greater than the third threshold value is displayed on the lower limb side from the upper abdominal start point slice.
- FIG. 14B An example in which circumscribed rectangles are superimposed and displayed on the spine, pelvis, and femur labels in FIG. 14A is shown in FIG. 14B.
- the amount of change in the area of the created circumscribed rectangle is examined.
- the body axis slice having the largest change amount is determined to be the lower abdomen starting point slice.
- the amount of bone change may be examined without depending on the circumscribed rectangle, and the lower abdomen starting point slice may be determined according to the result.
- the size (number of pixels) of the spine, pelvis, and femur labels is calculated for each slice, and the slice having the largest difference in size from the adjacent slice is determined to be the lower abdominal starting point slice.
- change_quantity For example, the lower abdomen start point slice may be determined in response to the above-described area or label size being below a predetermined threshold.
- This predetermined threshold value may be determined from data (for example, an average value) of a plurality of subjects, or may be determined based on the above-described area or the maximum value of the label. For example, a slice in which the above-described area or label has a predetermined ratio of the maximum value may be determined as the lower abdomen start point slice.
- Step 1216 indicates the end of the process.
- a lower limb starting point slice located at the upper end of the lower limb is determined in the CT image 130. An example of this determination process will be described with reference to FIGS.
- the process 1500 is a specific example of Step 818 in FIG.
- the process 1500 can also be a process performed by the system 100 by causing the CPU 102 to execute at least a part of program instructions included in the CT image segmentation program 122.
- Step 1502 indicates the start of processing.
- the processing indicated by reference numerals 1506 to 1512 is performed for each of the body axis cross-sectional slices from the lower abdominal starting point slice determined in step 816 among the body axis cross-sectional slices of the CT image 130. Do.
- step 1506 for the current body axis cross-sectional slice of the loop, a pixel group in a region corresponding to the body label determined in step 1204, the pixel group having a pixel value greater than or equal to the third threshold value or greater than the third threshold value.
- Label See the description of step 316 for the third threshold. Then, the top two labels having the largest sizes are extracted, and if these two labels do not overlap, it is determined that these two labels are femoral labels (step 1508). The state of the femoral label is shown in FIG.
- a hole is extracted for each femur label (step 1510). If the hole can be extracted, the pixel group of the region corresponding to the femoral label for the current body axis cross-sectional slice of the loop, the pixel value is equal to or greater than the third threshold value or the third threshold value
- a femur volume which is the volume (volume or number of pixels) of a larger pixel group, is calculated (step 1512).
- step 1516 the maximum value of the pelvic volume in the lower limb side axial slice from the lower abdomen starting point slice is determined. This determination is a pixel group in a region corresponding to the body label determined in step 1204 for each of the body axis cross-sectional slices from the lower abdomen starting point slice to the lower limb side, and the pixel value is greater than or equal to the third threshold value or This may be done by calculating the volume (volume or number of pixels) of the pixel group that is greater than the threshold of 3, and specifying the maximum value of the calculated volume.
- the lower limb starting point slice is determined.
- the lower limb starting point slice is the pelvis where the femoral volume value is determined in step 1516 among the body axis sectional slices from which the femoral volume can be calculated among the body axis sectional slices on the lower limb side from the lower abdominal starting point slice. It is determined as the first slice that becomes a predetermined ratio of the volume maximum value. As the predetermined ratio, for example, it has been confirmed by the inventors that half or less than the maximum value of the pelvic volume is one of appropriate values. However, other values may be adopted depending on the embodiment.
- Step 1520 indicates the end of the process.
- the slice position information of the chest start point slice, upper abdominal start point slice, cervical start point slice, head start point slice, lower abdominal start point slice, and lower limb start point slice determined in steps 808 to 818 is stored in the system 100.
- the slice position information 142 may be stored in the mass storage device 106. Depending on the embodiment, it may be stored in the main storage device 104.
- the slice position information may be, for example, the slice number of the body axis slice of the CT image 130.
- Step 822 indicates the end of process 800.
- step 216 in FIG. 2 a specific example of the processing in step 216 in FIG. 2 will be described with reference to FIG.
- FIG. 17 is a diagram for explaining a process 1700, which is one specific example of the process of step 216 in FIG.
- the process 1700 can be a process performed by the system 100 when at least a part of the program instructions included in the physiological accumulation removal program 124 is executed by the CPU 102.
- Step 1702 indicates the start of processing.
- data to be processed by the physiological accumulation removal program 124 is read (loaded).
- the data loaded here is a nuclear medicine image (PET image in this example) 132 (all or a part thereof).
- the slice position information 142 created and stored by the process 800 is also read.
- the nuclear medicine image 132 and / or the slice position information 142 is already stored in the main storage device 102 and is not loaded again. In this case, the processing in step 1704 will not be necessary.
- the slice position information 142 includes slice position information (slice number) of at least one of the chest start slice, upper abdomen start slice, neck start slice, head start slice, lower abdomen start slice, and lower limb start slice. )It is included.
- the slice position information is created based on the nuclear medicine image 132 and the registered CT image 130. Therefore, by using this slice position information, the chest start point slice, the upper abdomen start point slice, and the like can be specified also in the nuclear medicine image 132. That is, for example, in the nuclear medicine image 132, the body axis slice having the same slice number as the chest start point slice number included in the slice position information 142 is considered to be a slice located at the upper end of the chest in the nuclear medicine image 132. it can.
- the body axis slice having the same slice number as the chest start point slice number included in the slice position information 142 can be considered to be a chest start point slice in the nuclear medicine image 132 as well.
- the phrases “chest starting slice, upper abdominal starting slice, neck starting slice, head starting slice, lower abdominal starting slice, lower limb starting slice” It should be understood that it represents a slice.
- a process for determining a highly integrated portion of the head in the CT image 130 and the registered nuclear medicine image 132 is performed.
- a search region for a pixel having the maximum pixel value is set in the nuclear medicine image 132 using the slice position of the head start point slice and the slice position of the neck start point slice.
- the search region may be narrowed down to improve the processing speed and memory consumption. The inventor has confirmed that favorable results can be obtained in the search area defined as follows.
- ⁇ The axial direction is from the middle slice between the head start slice and the neck start slice to the head start slice. -The entire range for the sagittal direction. That is, from 0 to the matrix size in the sagittal slice. ⁇ For coronal direction, only left and right center.
- the definition of the search area may vary depending on the embodiment.
- step 1708 the pixel having the maximum pixel value in the nuclear medicine image 132 is determined in the search region set in step 1706.
- step 1710 region growing is performed from the pixel having the maximum pixel value determined in step 1708. It is determined that the region obtained by the region growing is a highly integrated portion of the head.
- a process for determining a highly integrated portion of the bladder in the CT image 130 and the registered nuclear medicine image 132 is performed.
- a search area for a pixel having the maximum pixel value in the nuclear medicine image 132 is set using the slice position of the lower abdomen start point slice and the slice position of the lower limb start point slice.
- the search area may be narrowed down to improve the processing speed and memory consumption. The inventor has confirmed that favorable results can be obtained in the search area defined as follows.
- the axial direction is from the middle slice between the lower abdomen start slice and lower limb start slice to the lower limb start slice. -The entire range for the sagittal direction. That is, from 0 to the matrix size in the sagittal slice. ⁇ For coronal direction, only left and right center.
- the definition of the search area may vary depending on the embodiment.
- step 1714 the pixel having the maximum pixel value in the nuclear medicine image 132 is determined in the search region set in step 1712.
- step 1716 region growing is performed from the pixel having the maximum pixel value determined in step 1714.
- the region obtained by this region growing is determined to be a highly integrated portion of the bladder.
- step 1718 the data of the head highly integrated region and bladder highly integrated region obtained in steps 1710 and 1718 are stored in the storage means of the system 100.
- the area data 144 may be stored in the mass storage device 106.
- step 220 based on the result of step 216, the nuclear medicine image 132 is displayed by masking the highly integrated region of the head and the highly integrated region of the bladder.
- a display example is shown in FIG.
- FIG. 18 shows one of the screens of the implementation example of the system 100.
- the image 1802 shows the boundary of the body region based on the slice position information 142.
- an image 1806 in which the head highly integrated region 1810 and the bladder highly integrated region 1812 are masked in the image 1804 appears based on the region data 144.
- the head highly integrated region 1810 and the bladder highly integrated region 1812 are masked, it becomes possible to observe nuclear medicine images without being obstructed by unnecessary signals from the head and bladder.
- the order of the processes introduced in the flowchart does not have to be executed in the order in which they were introduced. Depending on the practitioner's preference and necessity, the order may be changed or executed in parallel, These blocks may be implemented in an integral manner or may be implemented as an appropriate loop. All of these variations are included in the scope of the invention disclosed in the present application, and the scope of the invention is not limited by the implementation of processing.
- the description order of the processing determined in the claims does not necessarily determine the essential order of the processing. For example, an embodiment in which the processing order is different or an embodiment in which the processing is executed including a loop. Is also included in the scope of the claimed invention.
- embodiments of the CT image segmentation program 122 and the physiological accumulation removal program 124 include those that are a single program or programs that are composed of a plurality of independent programs. Can be. These two programs may be provided as a single program in some embodiments. In some embodiments, the registration program 120 may also be provided. As is well known, there are various implementation forms of the program, and all of these variations are included in the scope of the invention disclosed in the present application. Regardless of whether the present claims are claimed or not, the applicant alleges that he has the right to obtain a patent for all forms that do not depart from the spirit of the invention disclosed herein. Note that.
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- High Energy & Nuclear Physics (AREA)
- General Health & Medical Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Optics & Photonics (AREA)
- Public Health (AREA)
- Pathology (AREA)
- Biophysics (AREA)
- Heart & Thoracic Surgery (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Veterinary Medicine (AREA)
- Radiology & Medical Imaging (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Pulmonology (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Epidemiology (AREA)
- Primary Health Care (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Nuclear Medicine (AREA)
Abstract
Description
・ 上記CT画像に基づいて、画素値のヒストグラムを作成することと;
・ 上記ヒストグラムにおける脂肪領域のピークである脂肪領域ピークを決定することと;
・ 上記ヒストグラムにおける軟部組織領域のピークである軟部領域ピークを決定することと;
・ 上記脂肪領域ピークの階級値よりも階級値が小さい領域において、上記脂肪領域ピークの頻度値に基づいて、空気領域の階級値の上限を表す閾値である第1の閾値を決定することと;
・ 上記脂肪領域と上記軟部組織領域の境界の階級値を表す閾値である第2の閾値を決定することと;
・ 上記軟部領域ピークに対応する階級値よりも階級値が大きい領域において、上記軟部領域ピークの頻度値に基づいて、骨領域の階級値の下限を表す閾値である第3の閾値を決定することと;
を含む処理を遂行することを特徴とする。
・ 上記CT画像から、寝台部分を除去して人体部分のみを残した二値化画像である参照画像を作成することと;
・ 上記CT画像の画素のうち、上記参照画像の対応する画素にデータが存在する画素のみを用いて上記ヒストグラムを作成することと、
を含んでもよい。
・ 上記脂肪領域ピークを決定することは、上記脂肪領域ピークを含みうるように設定された下限の階級値から、水に対応する階級値である基準階級値までの範囲における最大頻度値を求めると共に、上記最大頻度値が上記基準階級値の頻度値ではない場合は、上記最大頻度値及び対応する階級値が、上記脂肪領域ピークの頻度値及び階級値であると決定することを含んでもよく。
・ 上記処理は、上記第2の閾値を、上記脂肪領域ピークの階級値から上記基準階級値までの範囲において、頻度値が最小になる階級値として決定することを更に含んでもよい。
・ 上記第2の閾値を、上記下限の階級値から上記基準階級値までの範囲において、頻度値の変化量が最小となる階級値として決定すること、ただし上記変化量は、iを整数として、
・ 階級値iの前の差分=(階級値i-1の頻度値)-(階級値iの頻度値);
・ 階級値iの後の差分=(階級値iの頻度値)-(階級値i+1の頻度値);
・ 階級値iの変化量=(階級値iの前の差分)2+(階級値iの後の差分)2
として定義される値である、上記決定することと;調査階級値を、
・ 上記下限の階級値から上記第2の閾値までの範囲で、上記変化量が最大となる階級値、または、
・ 上記第2の閾値から上記下限の階級値の方へと頻度値の変化を調べ、上記第2の閾値における頻度値を最初に下回る頻度値に対応する階級値、
のいずれかと定めることと;
・ 上記調査境界値から上記第2の閾値までの範囲で、上記前の差分の値が正で且つ最小となる階級値及び対応する頻度値を、上記脂肪領域ピークの階級値及び頻度値であると決定することと;
を含んでもよい。
・ 上記下限の階級値から上記第2の閾値までの範囲で、その頻度値が、上記脂肪領域ピークの頻度値の所定の割合以下または未満になる階級値;
・ 上記下限の階級値から上記脂肪領域ピークの階級値までの範囲で、上記変化量が最大となる階級値;
のいずれかとして決定されてもよい。
・ 一方の軸に、上記CT画像の体軸断面スライスのスライス番号を取り、もう一方の軸に、各スライス番号に対応するスライスにおける、画素値が上記第1の閾値以下または未満の画素群の少なくとも一部のボリュームを取ったグラフである空気領域ボリュームグラフを作成することと;
・ 上記空気領域ボリュームグラフが最大値を呈する体軸断面スライスより頭頂側の体軸断面スライスであって、そのボリューム値が、上記軟部組織ボリュームグラフの最大値の所定の割合となる体軸断面スライスを、胸部の上端に位置する胸部始点スライスであると決定することと;
を含む処理を遂行することを特徴とする。
・ 一方の軸に、上記CT画像の体軸断面スライスのスライス番号を取り、もう一方の軸に、各スライス番号に対応するスライスにおける、画素値が上記第2の閾値から上記第3の閾値の間にある画素群のボリュームを取ったグラフである軟部領域ボリュームグラフを作成することと;
・ 上記空気領域ボリュームグラフが最大値を呈する体軸断面スライスより下肢側の体軸断面スライスであって、上記軟部領域ボリュームグラフのボリューム値が最も大きくなる体軸断面スライスを、上腹部の上端に位置する上腹部始点スライスであると決定することと;
を含んでもよい。
・ 上記頸部始点スライスよりも頭頂側の体軸断面スライスに対して3Dラベリングを行うことと;
・ 上記頭部ラベルに対応する領域の画素群であって、画素値が上記第2の閾値から上記第3の閾値の間にある画素群のボリュームを上記頸部始点スライスから頭頂側へ各体軸断面スライスについて計算し、該ボリュームが最初にゼロになった体軸断面スライスを、頭部の上端に位置する頭部始点スライスであると決定することと;
を含んでもよい。
・ 上記上腹部始点スライスよりも下肢側の体軸断面スライスに対して3Dラベリングを行い、最も大きなラベルである胴体ラベルを決定することと;
・ 上記上腹部始点スライスよりも下肢側の体軸断面スライスに対して、上記胴体ラベルに対応する領域の画素群であって、画素値が上記第3の閾値以上又は上記第3の閾値より大きな画素群を抽出すると共に、該抽出した画素群に対して3Dラベリングを行い、最も大きなラベルである背骨・骨盤・大腿骨ラベルを決定することと;
・ 上記背骨・骨盤・大腿骨ラベルに対応する画素群であって、画素値が上記第3の閾値以上又は上記第3の閾値より大きな画素群の外接矩形を、上記上腹部始点スライスよりも下肢側の体軸断面スライスの各々について作成することと;
・ 上記外接矩形の面積の変化量が最も大きな体軸断面スライスを、下腹部の始点に位置する下腹部始点スライスであると決定することと;
を含んでもよい。
・ 上記胴体ラベルに対応する領域の画素群であって、画素値が上記第3の閾値以上又は上記第3の閾値より大きな画素群をラベリングすることと;
・ 同一のスライス内において、大きさが上位2つのラベルを抽出し、これら2つのラベルの左右の位置が重なっていなければ、これら2つのラベルを大腿骨ラベルであると決定することと;
・ 上記大腿骨ラベルが決定できた場合は、該大腿骨ラベルの各々について、穴を抽出することと;
・ 上記穴を抽出することができた場合は、上記大腿骨ラベルに対応する領域の画素群であって、画素値が上記第3の閾値以上又は上記第3の閾値より大きな画素群のボリュームである大腿骨ボリュームを計算することと;
を行うことを含み、さらに上記処理は、
・ 上記下腹部始点スライスから下肢側の体軸断面スライスにおける、上記胴体ラベルに対応する領域の画素群であって、画素値が上記第3の閾値以上又は上記第3の閾値より大きな画素群のボリュームの最大値を計算することと;
・ 上記下腹部始点スライスから下肢側の体軸断面スライスのうち、上記大腿骨ボリュームが計算できた体軸断面スライスの中で、該大腿骨ボリュームの値が、上記最大値の所定の割合になる最初のスライスを、下肢の上端に位置する下肢始点スライスであると決定することと;
を含んでもよい。
・ 上記核医学画像において、上記頭部始点スライスのスライス位置と、上記頸部始点スライスのスライス位置とを用いて、最大画素値探索領域を設定することと;
・ 上記最大画素値探索領域において、最大画素値を有する画素を決定することと;
・ 上記最大画素値を有する画素から、リージョングローイングの手法により、頭部の高集積領域を決定することと;
を含む処理を遂行することを特徴とする。
・ 上記核医学画像において、上記下腹部始点スライスのスライス位置と、上記下肢始点スライスのスライス位置とを用いて、最大画素値探索領域を設定することと;
・ 上記最大画素値探索領域において、最大画素値を有する画素を決定することと;
・ 上記最大画素値を有する画素から、リージョングローイングの手法により、膀胱部の高集積領域を決定することと;
を含む処理を遂行することを特徴とする。
・ 階級値iの前の差分=(階級値i-1の頻度値)-(階級値iの頻度値);
・ 階級値iの後の差分=(階級値iの頻度値)-(階級値i+1の頻度値);
・ 階級値iの変化量=(階級値iの前の差分)2+(階級値iの後の差分)2
ただしiは整数を表す。
(a)ステップ604で設定された下限の階級値から、ステップ612で決定された第2の閾値までの範囲で、ステップ610で計算された変化量が最大となる階級値。
(b)ステップ612で決定された第2の閾値から上記下限の階級値の方へと頻度値の変化を調べ、第2の閾値における頻度値を最初に下回る頻度値に対応する階級値。
このうち(b)で定められる調査階級値の様子を、図6Bに示す。
(314a)第1の閾値として、ステップ604で設定された下限の階級値から、ステップ310で決定された第2の閾値までの範囲で、その頻度値が、脂肪領域ピークの頻度値の所定の割合以下または未満になる階級値を採用する。この所定の割合として、例えば10%が適切な値の一つであることが発明者によって確かめられている。しかし、例えば5%等、他の値でもよい。複数の被験者のデータを用いて、CT装置毎に定めてもよい。
(314b)実施形態によっては、第1の閾値の決定条件に、直前の階級値との頻度値の差の絶対値が所定の値以下であること、という条件を入れてもよい。この所定の値として、例えば50が適切な値の一つであることが発明者によって確かめられている。しかしこの値は単なる例示に過ぎず、実施形態によって様々な値を取り得るものである。
(314c)実施形態によっては、第1の閾値として、ステップ604で設定された下限の階級値から、ステップ310で決定された第2の閾値までの範囲において、ステップ610で計算されるものと同じ「変化量」が最大となる階級値を採用してもよい。実施形態によっては、条件314a,314bを満たす階級値が存在しない場合に、条件314cを使って第1の閾値を決定することとしてもよい。
(814a)頸部始点スライスよりも頭頂側の体軸断面スライスに対して3Dラベリングを行う。
(814b)体軸断面スライスにおいて中央部に位置するラベルである頭部ラベルを決定する。
(814c)頭部ラベルに対応する領域の画素群であって、画素値が第2の閾値から第3の閾値の間にある画素群のボリュームを、頸部始点スライスから頭頂側へ各体軸断面スライスについて計算し、該ボリュームが最初にゼロになった体軸断面スライスを、頭部始点スライスであると決定する。
・ 体軸(axial)方向は、頭部始点スライスと頸部始点スライスとの中間のスライスから頭部始点スライスまで。
・ 矢状(sagittal)方向については全範囲。すなわち矢状断面スライスにおいて0からマトリクスサイズまで。
・ 冠状(coronal)方向については、左右中心のみ。
その他、実施形態によって、探索領域の定義は変わってよい。
・ 体軸(axial)方向は、下腹部始点スライスと下肢始点スライスとの中間のスライスから下肢始点スライスまで。
・ 矢状(sagittal)方向については全範囲。すなわち矢状断面スライスにおいて0からマトリクスサイズまで。
・ 冠状(coronal)方向については、左右中心のみ。
その他、実施形態によって、探索領域の定義は変わってよい。
102 CPU
104 主記憶装置
106 大容量記憶装置
107 ディスプレイ・インターフェース
108 周辺機器インタフェース
109 ネットワーク・インターフェース
120 レジストレーションプログラム
122 CT画像セグメンテーションプログラム
124 生理的集積除去プログラム
130 CT画像データ
132 核医学画像データ
Claims (24)
- CT画像の画素値の分類を自動で行う方法であって、
前記CT画像に基づいて、画素値のヒストグラムを作成することと;
前記ヒストグラムにおける脂肪領域のピークである脂肪領域ピークを決定することと;
前記ヒストグラムにおける軟部組織領域のピークである軟部領域ピークを決定することと;
前記脂肪領域ピークの階級値よりも階級値が小さい領域において、前記脂肪領域ピークの頻度値に基づいて、空気領域の階級値の上限を表す閾値である第1の閾値を決定することと;
前記脂肪領域と前記軟部組織領域の境界の階級値を表す閾値である第2の閾値を決定することと;
前記軟部領域ピークに対応する階級値よりも階級値が大きい領域において、前記軟部領域ピークの頻度値に基づいて、骨領域の階級値の下限を表す閾値である第3の閾値を決定することと;
を含む、方法。 - 前記ヒストグラムを作成することは、
前記CT画像から、寝台部分を除去して人体部分のみを残した二値化画像である参照画像を作成することと;
前記CT画像の画素のうち、前記参照画像の対応する画素にデータが存在する画素のみを用いて前記ヒストグラムを作成することと、
を含む、請求項1に記載の方法。 - 前記脂肪領域ピーク及び前記軟部領域ピークを決定する前に、前記ヒストグラムにスムージング処理を施す、請求項1または2に記載の方法。
- 請求項1から3のいずれかに記載の方法であって、
前記脂肪領域ピークを決定することは、前記脂肪領域ピークを含みうるように設定された下限の階級値から、水に対応する階級値である基準階級値までの範囲における最大頻度値を求めると共に、前記最大頻度値が前記基準階級値の頻度値ではない場合は、前記最大頻度値及び対応する階級値が、前記脂肪領域ピークの頻度値及び階級値であると決定することを含み、
前記方法は、前記第2の閾値を、前記脂肪領域ピークの階級値から前記基準階級値までの範囲において、頻度値が最小になる階級値として決定することを更に含む、
方法。 - 前記下限の階級値から前記基準階級値までの範囲における最大頻度値が、前記基準階級値の頻度値である場合、前記脂肪領域ピークを決定することは、
前記第2の閾値を、前記下限の階級値から前記基準階級値までの範囲において、頻度値の変化量が最小となる階級値として決定すること、ただし前記変化量は、iを整数として、
・ 階級値iの前の差分=(階級値i-1の頻度値)-(階級値iの頻度値);
・ 階級値iの後の差分=(階級値i1の頻度値)-(階級値i+1の頻度値);
・ 階級値iの変化量=(階級値iの前の差分)2+(階級値iの後の差分)2
として定義される値である、前記決定することと;
調査階級値を、
・ 前記下限の階級値から前記第2の閾値までの範囲で、前記変化量が最大となる階級値、または、
・ 前記第2の閾値から前記下限の階級値の方へと頻度値の変化を調べ、前記第2の閾値における頻度値を最初に下回る頻度値に対応する階級値、
のいずれかと定めることと;
前記調査境界値から前記第2の閾値までの範囲で、前記前の差分の値が正で且つ最小となる階級値及び対応する頻度値を、前記脂肪領域ピークの階級値及び頻度値であると決定することと;
を含む、請求項4に記載の方法。 - 前記調査境界値から前記第2の閾値までの範囲で、前記前の差分の値が正となる階級値が見つからない場合、前記脂肪領域ピークを決定することは、前記下限の階級値から前記第2の閾値までの範囲の最大頻度値及び対応する階級値を、前記脂肪領域ピークの頻度値及び階級値であると決定することを含む、請求項5に記載の方法。
- 前記軟部領域ピークを決定することは、所定の階級値範囲内でピーク検出処理を行うことを含む、請求項5又は6に記載の方法。
- 前記第1の閾値は、
・ 前記脂肪領域ピークの階級値よりも小さな階級値であって、前記下限の階級値から前記第2の閾値までの範囲で、その頻度値が、前記脂肪領域ピークの頻度値の所定の割合以下または未満になる階級値;
・ 前記下限の階級値から前記脂肪領域ピークの階級値までの範囲で、前記変化量が最大となる階級値;
のいずれかとして決定され、
前記第3の閾値は、前記軟部領域ピークに対応する階級値よりも大きな階級値であって、その頻度値が、前記軟部領域ピークの頻度値の所定の割合になる階級値として決定される、請求項7に記載の方法。 - 請求項1から8のいずれかに記載の方法により前記第1~第3の閾値が決定された前記CT画像の領域分割を自動で行うための方法であって、
一方の軸に、前記CT画像の体軸断面スライスのスライス番号を取り、もう一方の軸に、各スライス番号に対応するスライスにおける、画素値が前記第1の閾値以下または未満の画素群の少なくとも一部のボリュームを取ったグラフである空気領域ボリュームグラフを作成することと;
前記空気領域ボリュームグラフが最大値を呈する体軸断面スライスより頭頂側の体軸断面スライスであって、そのボリューム値が、前記空気領域ボリュームグラフの最大値の所定の割合となる体軸断面スライスを、胸部の上端に位置する胸部始点スライスであると決定することと;
を含む、方法。 - 前記胸部始点スライスを決定する前に、前記空気領域ボリュームグラフにスムージング処理を施すことを含む、請求項9に記載の方法。
- 一方の軸に、前記CT画像の体軸断面スライスのスライス番号を取り、もう一方の軸に、各スライス番号に対応するスライスにおける、画素値が前記第2の閾値から前記第3の閾値の間にある画素群のボリュームを取ったグラフである軟部領域ボリュームグラフを作成することと;
前記空気領域ボリュームグラフが最大値を呈する体軸断面スライスより下肢側の体軸断面スライスであって、前記軟部領域ボリュームグラフのボリューム値が最も大きくなる体軸断面スライスを、上腹部の上端に位置する上腹部始点スライスであると決定することと;
を更に含む、請求項9または10に記載の方法。 - 前記上腹部始点スライスを決定する前に、前記軟部領域ボリュームグラフにスムージング処理を施すことを含む、請求項11に記載の方法。
- 前記前記胸部始点スライスよりも頭頂側の体軸断面スライスであって、前記空気領域ボリュームグラフのボリューム値が最も大きくなる体軸断面スライスを、頸部の上端に位置する頸部始点スライスであると決定することを更に含む、請求項9から13の何れかに記載の方法。
- 前記頸部始点スライスよりも頭頂側の体軸断面スライスに対して3Dラベリングを行うことと;
体軸断面スライスにおいて中央部に位置するラベルである頭部ラベルを決定することと;
前記頭部ラベルに対応する領域の画素群であって、画素値が前記第2の閾値から前記第3の閾値の間にある画素群のボリュームを前記頸部始点スライスから頭頂側へ各体軸断面スライスについて計算し、該ボリュームが最初にゼロになった体軸断面スライスを、頭部の上端に位置する頭部始点スライスであると決定することと;
を更に含む、請求項13に記載の方法。 - 前記上腹部始点スライスよりも下肢側の体軸断面スライスについて、骨の量の変化に基づいて、下腹部の上端に位置する下腹部始点スライスであると決定することを更に含む、請求項11又は12、及び、請求項11若しくは12に従属する請求項13又は14のいずれか1項に記載の方法。
- 前記下腹部始点スライスを決定することは、
前記上腹部始点スライスよりも下肢側の体軸断面スライスに対して3Dラベリングを行い、最も大きなラベルである胴体ラベルを決定することと;
前記上腹部始点スライスよりも下肢側の体軸断面スライスに対して、前記胴体ラベルに対応する領域の画素群であって、画素値が前記第3の閾値以上又は前記第3の閾値より大きな画素群を抽出すると共に、該抽出した画素群に対して3Dラベリングを行い、最も大きなラベルである背骨・骨盤・大腿骨ラベルを決定することと;
前記背骨・骨盤・大腿骨ラベルに対応する画素群であって、画素値が前記第3の閾値以上又は前記第3の閾値より大きな画素群の外接矩形を、前記上腹部始点スライスよりも下肢側の体軸断面スライスの各々について作成することと;
前記外接矩形の変化量が最も大きな体軸断面スライスを、下腹部の上端に位置する下腹部始点スライスであると決定することと;
を含む、請求項15に記載の方法。 - 前記下腹部始点スライスから下肢側の体軸断面スライスの各々について、
前記胴体ラベルに対応する領域の画素群であって、画素値が前記第3の閾値以上又は前記第3の閾値より大きな画素群をラベリングすることと;
大きさが上位2つのラベルを抽出し、これら2つのラベルが重なっていなければ、これら2つのラベルを大腿骨ラベルであると決定することと;
前記大腿骨ラベルが決定できた場合は、該大腿骨ラベルの各々について、穴を抽出することと;
前記穴を抽出することができた場合は、前記大腿骨ラベルに対応する領域の画素群であって、画素値が前記第3の閾値以上又は前記第3の閾値より大きな画素群のボリュームである大腿骨ボリュームを計算することと;
を行うことと;
前記下腹部始点スライスから下肢側の体軸断面スライスにおける、前記胴体ラベルに対応する領域の画素群であって、画素値が前記第3の閾値以上又は前記第3の閾値より大きな画素群のボリュームの最大値を計算することと;
前記下腹部始点スライスから下肢側の体軸断面スライスのうち、前記大腿骨ボリュームが計算できた体軸断面スライスの中で、該大腿骨ボリュームの値が、前記最大値の所定の割合になる最初のスライスを、下肢の上端に位置する下肢始点スライスであると決定することと;
を更に含む、請求項16に記載の方法。 - 核医学画像とのレジストレーションがなされたCT画像に対して行われる、請求項1から17のいずれかに記載の方法。
- 核医学画像から生理的集積を自動で除去するための方法であって、
請求項14に従属する請求項17に記載の方法により決定される、頭部始点スライスのスライス位置の情報と、頸部始点スライスのスライス位置の情報とを読み込むことと;
前記核医学画像において、前記頭部始点スライスのスライス位置と、前記頸部始点スライスのスライス位置とを用いて、最大画素値探索領域を設定することと;
前記最大画素値探索領域において、最大画素値を有する画素を決定することと;
前記最大画素値を有する画素から、リージョングローイングの手法により、頭部の高集積領域を決定することと;
を含む、方法。 - 前記決定した脳領域にマスクをかけて前記核医学画像を表示することを更に含む、請求項19に記載の方法。
- 核医学画像から生理的集積を自動で除去するための方法であって、 請求項17から19のいずれか一項であって、請求項16に従属する請求項に記載の方法により決定される、下腹部始点スライスのスライス位置の情報と、下肢始点スライスのスライス位置の情報とを読み込むことと;
前記核医学画像において、前記下腹部始点スライスのスライス位置と、前記下肢始点スライスのスライス位置とを用いて、最大画素値探索領域を設定することと;
前記最大画素値探索領域において、最大画素値を有する画素を決定することと;
前記最大画素値を有する画素から、リージョングローイングの手法により、膀胱部の高集積領域を決定することと;
を含む、方法。 - 前記決定した膀胱領域にマスクをかけて前記核医学画像を表示することを含む、請求項21に記載の方法。
- 処理手段と記憶手段とを備える装置であって、前記記憶手段はプログラム命令を備え、該プログラム命令は、前記処理手段に実行されると、前記装置に、請求項1から22のいずれかに記載の方法を遂行させるように構成される、装置。
- 装置の手段で実行されると、前記装置に、請求項1から22のいずれかに記載の方法を遂行させるように構成されるプログラム命令を備える、コンピュータプログラム。
Priority Applications (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US16/093,423 US20190150860A1 (en) | 2016-04-13 | 2017-01-18 | Automatic Removal of Physiological Accumulation from Nuclear Medicine Image, and Automatic Segmentation of CT Image |
CN201780023599.2A CN109069094A (zh) | 2016-04-13 | 2017-01-18 | 从核医学图像对生理性蓄积的自动去除及ct图像的自动分割 |
KR1020187028700A KR20180128001A (ko) | 2016-04-13 | 2017-01-18 | 핵의학 화상으로부터의 생리적 집적의 자동 제거 및 ct 화상의 자동 세그멘테이션 |
EP17782075.0A EP3443907A4 (en) | 2016-04-13 | 2017-01-18 | AUTOMATIC REMOVAL OF PHYSIOLOGICAL ACCUMULATIONS FROM IMAGE OF NUCLEAR MEDICINE, AND AUTOMATIC IMAGE SEGMENTATION CT |
JP2018511886A JP6735340B2 (ja) | 2016-04-13 | 2017-01-18 | 核医学画像からの生理的集積の自動除去及びct画像の自動セグメンテーション |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2016-080087 | 2016-04-13 | ||
JP2016080087 | 2016-04-13 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2017179256A1 true WO2017179256A1 (ja) | 2017-10-19 |
Family
ID=60041644
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2017/001460 WO2017179256A1 (ja) | 2016-04-13 | 2017-01-18 | 核医学画像からの生理的集積の自動除去及びct画像の自動セグメンテーション |
Country Status (7)
Country | Link |
---|---|
US (1) | US20190150860A1 (ja) |
EP (1) | EP3443907A4 (ja) |
JP (1) | JP6735340B2 (ja) |
KR (1) | KR20180128001A (ja) |
CN (1) | CN109069094A (ja) |
TW (1) | TW201736865A (ja) |
WO (1) | WO2017179256A1 (ja) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105931251A (zh) * | 2016-05-12 | 2016-09-07 | 中国科学院深圳先进技术研究院 | 一种ct图像扫描床去除方法及装置 |
TWI697010B (zh) * | 2018-12-28 | 2020-06-21 | 國立成功大學 | 醫療矢面影像的取得方法、神經網路的訓練方法及計算機裝置 |
CN115131388B (zh) * | 2022-03-03 | 2023-09-01 | 中国人民解放军总医院第四医学中心 | 骨量定向叠加计算的提取方法、装置及设备 |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS6324478A (ja) * | 1986-04-14 | 1988-02-01 | ピクサ− | 三次元デ−タセツトを表す二次元デ−タ表示を発生する方法 |
JP2011062253A (ja) * | 2009-09-15 | 2011-03-31 | Hamamatsu Photonics Kk | 断層画像処理装置、筋断面積測定システム及び断層画像処理方法 |
Family Cites Families (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5068788A (en) * | 1988-11-29 | 1991-11-26 | Columbia Scientific Inc. | Quantitative computed tomography system |
JP3453085B2 (ja) * | 1998-07-23 | 2003-10-06 | ジーイー横河メディカルシステム株式会社 | X線ct装置 |
KR100283106B1 (ko) * | 1998-10-28 | 2001-03-02 | 정선종 | 전산화 단층촬영에서 가우시안 함수 근사에 의한 체지방 범위설정 방법 |
JP2005506140A (ja) * | 2001-10-16 | 2005-03-03 | ザ・ユニバーシティー・オブ・シカゴ | コンピュータ支援の3次元病変検出方法 |
US7177453B2 (en) * | 2002-11-26 | 2007-02-13 | General Electric Company | Method and apparatus for partitioning a volume |
JP4800127B2 (ja) * | 2006-06-29 | 2011-10-26 | 富士フイルム株式会社 | 医用画像分割装置、及び、医用画像分割プログラム |
FR2921177B1 (fr) * | 2007-09-17 | 2010-01-22 | Gen Electric | Procede de traitement d'images anatomiques en volume et systeme d'imagerie mettant en oeuvre ce procede |
JP4437333B2 (ja) * | 2007-09-28 | 2010-03-24 | ジーイー・メディカル・システムズ・グローバル・テクノロジー・カンパニー・エルエルシー | 画像処理方法および画像処理装置並びにプログラム |
US8218848B2 (en) * | 2008-07-23 | 2012-07-10 | Siemens Aktiengesellschaft | System and method for the generation of attenuation correction maps from MR images |
WO2012125867A1 (en) * | 2011-03-15 | 2012-09-20 | The Trustees Of Columbia University In The City Of New York | Method and system for quality assurance of cross sectional imaging scans |
JP6073616B2 (ja) * | 2011-09-28 | 2017-02-01 | 東芝メディカルシステムズ株式会社 | X線ct装置、画像処理装置及びプログラム |
US9204817B2 (en) * | 2012-04-19 | 2015-12-08 | General Electric Company | Attenuation correction in positron emission tomography using magnetic resonance imaging |
CN103106408B (zh) * | 2013-01-25 | 2016-02-10 | 西安电子科技大学 | 无监督分割的胃部ct图像淋巴结自动辅助检测系统 |
CN103400385B (zh) * | 2013-07-22 | 2016-05-25 | 西安电子科技大学 | 稀疏动态集成选择的胃部ct图像疑似淋巴结提取方法 |
US8995739B2 (en) * | 2013-08-21 | 2015-03-31 | Seiko Epson Corporation | Ultrasound image object boundary localization by intensity histogram classification using relationships among boundaries |
WO2017106645A1 (en) * | 2015-12-18 | 2017-06-22 | The Regents Of The University Of California | Interpretation and quantification of emergency features on head computed tomography |
-
2017
- 2017-01-17 TW TW106101522A patent/TW201736865A/zh unknown
- 2017-01-18 CN CN201780023599.2A patent/CN109069094A/zh active Pending
- 2017-01-18 JP JP2018511886A patent/JP6735340B2/ja active Active
- 2017-01-18 EP EP17782075.0A patent/EP3443907A4/en not_active Withdrawn
- 2017-01-18 WO PCT/JP2017/001460 patent/WO2017179256A1/ja active Application Filing
- 2017-01-18 US US16/093,423 patent/US20190150860A1/en not_active Abandoned
- 2017-01-18 KR KR1020187028700A patent/KR20180128001A/ko unknown
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPS6324478A (ja) * | 1986-04-14 | 1988-02-01 | ピクサ− | 三次元デ−タセツトを表す二次元デ−タ表示を発生する方法 |
JP2011062253A (ja) * | 2009-09-15 | 2011-03-31 | Hamamatsu Photonics Kk | 断層画像処理装置、筋断面積測定システム及び断層画像処理方法 |
Also Published As
Publication number | Publication date |
---|---|
TW201736865A (zh) | 2017-10-16 |
JP6735340B2 (ja) | 2020-08-05 |
KR20180128001A (ko) | 2018-11-30 |
US20190150860A1 (en) | 2019-05-23 |
EP3443907A4 (en) | 2019-08-14 |
CN109069094A (zh) | 2018-12-21 |
EP3443907A1 (en) | 2019-02-20 |
JPWO2017179256A1 (ja) | 2019-02-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Li et al. | Deep learning for hemorrhagic lesion detection and segmentation on brain CT images | |
JP6947759B2 (ja) | 解剖学的対象物を自動的に検出、位置特定、及びセマンティックセグメンテーションするシステム及び方法 | |
JP5209984B2 (ja) | 血管画像抽出及びラベル付けのシステム及び方法 | |
Karthik et al. | A multi-scale approach for detection of ischemic stroke from brain MR images using discrete curvelet transformation | |
JP2006208250A (ja) | 画像データの領域分類解析システム | |
WO2017179256A1 (ja) | 核医学画像からの生理的集積の自動除去及びct画像の自動セグメンテーション | |
Pandey et al. | A systematic review of the automatic kidney segmentation methods in abdominal images | |
Cenek et al. | Survey of image processing techniques for brain pathology diagnosis: Challenges and opportunities | |
Gao et al. | CCE-Net: a rib fracture diagnosis network based on contralateral, contextual, and edge enhanced modules | |
Varçin et al. | Diagnosis of lumbar spondylolisthesis via convolutional neural networks | |
US20140228667A1 (en) | Determining lesions in image data of an examination object | |
Guerroudji et al. | Automatic brain tumor segmentation, and 3d reconstruction and visualization using augmented reality | |
Xu et al. | RUnT: A network combining residual U-Net and transformer for vertebral edge feature fusion constrained spine CT image segmentation | |
CN105844687B (zh) | 用于处理医学影像的装置和方法 | |
Smith et al. | Automated torso contour extraction from clinical cardiac MR slices for 3D torso reconstruction | |
Sharma et al. | Importance of deep learning models to perform segmentation on medical imaging modalities | |
JP6837054B2 (ja) | Ct画像中の骨領域の自動推定方法、装置、およびコンピュータプログラム | |
Affane et al. | Robust deep 3-d architectures based on vascular patterns for liver vessel segmentation | |
WO2020246192A1 (ja) | 修正指示領域表示装置、方法およびプログラム | |
CN111127636B (zh) | 一种智能的复杂关节内骨折桌面级三维诊断系统 | |
Whitney et al. | 7 ARTIFICIAL INTELLIGENCE IN MEDICAL IMAGING | |
Musatian et al. | Medical images segmentation operations | |
Ray et al. | An automatic method for complete brain matter segmentation from multislice CT scan | |
WO2022270151A1 (ja) | 画像処理装置、方法およびプログラム | |
Ren et al. | Cheung AL-y, Ho Wy, Qin J and Cai J (2021) Investigation of a Novel Deep Learning-Based Computed Tomography Perfusion Mapping Framework for Functional Lung Avoidance Radiotherapy. Front |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 2018511886 Country of ref document: JP |
|
ENP | Entry into the national phase |
Ref document number: 20187028700 Country of ref document: KR Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2017782075 Country of ref document: EP |
|
ENP | Entry into the national phase |
Ref document number: 2017782075 Country of ref document: EP Effective date: 20181113 |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 17782075 Country of ref document: EP Kind code of ref document: A1 |