WO2020241110A1 - 画像処理装置、画像処理方法及びプログラム - Google Patents

画像処理装置、画像処理方法及びプログラム Download PDF

Info

Publication number
WO2020241110A1
WO2020241110A1 PCT/JP2020/016777 JP2020016777W WO2020241110A1 WO 2020241110 A1 WO2020241110 A1 WO 2020241110A1 JP 2020016777 W JP2020016777 W JP 2020016777W WO 2020241110 A1 WO2020241110 A1 WO 2020241110A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
substance
radiation
image processing
noise
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/JP2020/016777
Other languages
English (en)
French (fr)
Japanese (ja)
Inventor
聡太 鳥居
貴司 岩下
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Canon Inc
Original Assignee
Canon 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 Canon Inc filed Critical Canon Inc
Publication of WO2020241110A1 publication Critical patent/WO2020241110A1/ja
Priority to US17/529,327 priority Critical patent/US12136191B2/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

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
    • A61B6/48—Diagnostic techniques
    • A61B6/482—Diagnostic techniques involving multiple energy imaging
    • 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
    • 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
    • A61B6/5258—Devices using data or image processing specially adapted for radiation diagnosis involving detection or reduction of artifacts or noise
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T1/00—General purpose image data processing
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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 OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10116—X-ray image
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; 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
    • G06T2207/30008—Bone

Definitions

  • the present invention relates to an image processing device, an image processing method, and a program, and more specifically, to an image processing technique used for still image shooting such as general shooting in medical diagnosis and moving image shooting such as fluoroscopic shooting.
  • spectral imaging technology which is an imaging technology that uses radiation energy information, has been used to obtain thickness images of multiple substances from multiple different radiation energy images, or to obtain images of surface density and effective atomic number images. can do.
  • Patent Document 1 discloses a technique for improving the image quality of a bone image by smoothing an image of soft tissue and subtracting the image from the accumulated image.
  • an object of the present invention is to provide an image processing technique capable of acquiring a substance property image having excellent quantitativeness and reduced noise.
  • the image processing apparatus generates a material property image of a substance contained in a plurality of radiation images taken with different radiation energies using the radiation energy spectrum.
  • Image generation means to be A noise reduction processing means for reducing a noise component of the substance characteristic image is provided.
  • the image generation means uses a composite image acquired from the plurality of radiation images, a first substance characteristic image in which the noise component is reduced, and a composite spectrum obtained from spectra of different radiation energies to reduce noise. It is characterized by generating a second material property image obtained.
  • the image processing apparatus generates a material property image of a substance contained in a plurality of radiation images taken with different radiation energies using the radiation energy spectrum, and the plurality of shot radiation images taken.
  • An image generation means for generating a first table that outputs a material property image corresponding to A noise reduction processing means for reducing a noise component of the substance characteristic image is provided.
  • the image generation means A second substance whose noise is reduced by using a composite image acquired from the plurality of radiation images, a first substance characteristic image in which the noise component is reduced, and a composite spectrum obtained from spectra of different radiation energies.
  • a second table for outputting a characteristic image is generated, and a second material characteristic image corresponding to the composite image and the first substance characteristic image is output from the second table.
  • noise is reduced and it becomes possible to acquire a substance property image having excellent quantitative properties.
  • the accompanying drawings are included in the specification and are used to form a part thereof, show an embodiment of the present invention, and explain the principle of the present invention together with the description thereof.
  • radiation includes not only X-rays but also ⁇ -rays, ⁇ -rays, ⁇ -rays, and various particle beams.
  • FIG. 1 is a diagram showing a configuration example of a radiography system 100 according to a first embodiment of the present invention.
  • the radiography system 100 includes a radiation generator 104, a radiation tube 101, an FPD 102 (Flat Panel Detector), and an image processing device 120.
  • the configuration of the radiography system 100 is also simply referred to as a radiography apparatus.
  • the image processing device 120 processes image information based on a radiation image of a subject.
  • the image processing device 120 includes a control unit 105, a storage unit 108, an image processing unit 109, a display control unit 116, and the like.
  • the radiation generator 104 gives a high voltage pulse to the radiation tube 101 to generate radiation by pressing the exposure switch, and the radiation tube 101 irradiates the subject 103 with radiation.
  • the type of radiation is not particularly limited, but generally X-rays are used as used here.
  • the FPD 102 When the subject 103 is irradiated with radiation from the radiation tube 101, the FPD 102 accumulates an electric charge based on the image signal and acquires a radiation image. The FPD 102 transfers the radiographic image to the image processing apparatus 120.
  • the FPD 102 has a radiation detection unit (not shown) provided with a pixel array for generating a signal corresponding to radiation.
  • the radiation detection unit detects the radiation transmitted through the subject 103 as an image signal.
  • pixels that output a signal corresponding to the incident light are arranged in an array (two-dimensional region).
  • the photoelectric conversion element of each pixel converts the radiation converted into visible light by the phosphor into an electric signal and outputs it as an image signal.
  • the radiation detection unit is configured to detect the radiation transmitted through the subject 103 and acquire an image signal (radiation image).
  • the driving unit of the FPD 102 outputs an image signal (radiation image) read out according to an instruction from the control unit 105 to the control unit 105.
  • the control unit 105 has an image processing unit 109 that processes a radiation image acquired from the FPD 102, and a storage unit 108 that stores the result of image processing and various programs.
  • the storage unit 108 is composed of, for example, a ROM (Read Only Memory), a RAM (Random Access Memory), and the like.
  • the storage unit 108 can store the image output from the control unit 105, the image processed by the image processing unit 109, and the calculation result of the image processing unit 109.
  • the image processing unit 109 has an image generation unit 110, a noise reduction processing unit 111, and a compositing unit 112 as functional configurations.
  • the functions of each unit are configured by using a program read from one or a plurality of CPUs (central processing units) and a storage unit 108.
  • the configuration of each portion of the image processing unit 109 fulfills the same function, they may be configured by an integrated circuit or the like.
  • the internal configuration of the image processing device 120 may include a graphic control unit such as a GPU (Graphics Processing Unit), a communication unit such as a network card, an input / output control unit such as a keyboard, a display, or a touch panel.
  • the monitor 106 displays a radiation image (digital image) received by the control unit 105 from the FPD 102 and an image processed by the image processing unit 109.
  • the display control unit 116 can control the display of the monitor 106 (display unit).
  • the operation unit 107 can input instructions to the image processing unit 109 and the FPD 102, and receives the input of the instructions to the FPD 102 via the user interface.
  • the control unit 105 can perform imaging control using an energy subtraction method for obtaining a new image (for example, a bone image and a fat image) by processing a plurality of radiation images having different energies of radiation irradiating the subject. Is.
  • a new image for example, a bone image and a fat image
  • the FPD 102 performs a plurality of samplings for one irradiation. As a result, the FPD 102 can acquire an image of low-energy radiation (low-energy radiation image) and an image of high-energy radiation (high-energy radiation image) with a single irradiation.
  • the shooting by the FPD 102 may be a still image shooting or a moving image shooting.
  • FPD 102 temporarily stored radiation distribution information in the after sampling hold embodiment allows reading, control unit 105, at different timings from FPD 102, the radiation distribution information (X L) and the radiation distribution information (X L + X H) Read out.
  • Control unit 105 by subtracting the radiation distribution information (X L) from the radiation distribution information (X L + X H), can be obtained radiation distribution information (X H).
  • the low-energy radiation distribution information ( XL ) becomes an image by low-energy radiation (low-energy image)
  • the high-energy radiation distribution information (X H ) becomes an image by high-energy radiation (high-energy image). become.
  • the image processing unit 109 has an image generation unit 110, a noise reduction processing unit 111, and a compositing unit 112 as functional configurations.
  • the image generation unit 110 can generate a plurality of material characteristic images using a plurality of radiation images taken with different radiation energies, and the image generation unit 110 separates substances from the radiation images taken by the FPD 102. Generates substance property images such as images and substance identification images.
  • the image generation unit 110 generates a substance characteristic image of a substance contained in a plurality of radiation images taken with different radiation energies using a radiation energy spectrum.
  • the image generation unit 110 generates a substance property image using a plurality of radiation images acquired by one irradiation.
  • the noise reduction processing unit 111 reduces the noise component of the material characteristic image by the noise reduction processing.
  • the synthesis unit 112 generates a composite image from a plurality of radiation images (low energy image, high energy image).
  • the image generation unit 110 synthesizes a composite image acquired from a plurality of radiation images (low-energy image, high-energy image), a noise-reduced first material characteristic image, and a composite obtained from different radiation energy spectra. The spectrum is used to generate a noise-reduced second material property image.
  • the substance-separated image is an image in which a plurality of substances contained in a plurality of radiation images (low-energy image, high-energy image) are separated, and when the subject is represented by two or more specific substances, the substance.
  • a plurality of substances include fat as a soft substance constituting a subject and bone as a hard substance.
  • the substance identification image is an image showing the distribution of effective atomic numbers and surface densities of substances contained in a plurality of radiation images (low energy image, high energy image), and when the subject is represented by one specific substance. In addition, it refers to an image decomposed into the distribution of the effective atomic number and surface density of the substance.
  • the effective atomic number indicates the atomic number of a substance corresponding to an average view of the elements of an element, compound, or mixture, and the atomic slope number of a virtual element that attenuates photons at the same rate as that substance. It is a quantitative index showing.
  • the effective atomic number image is an image composed of atomic numbers corresponding to the case where the subject is represented by a specific substance in units of pixels.
  • the FPD102 performs sampling a plurality of times for one irradiation. As a result, the FPD 102 can acquire an image of low-energy radiation (low-energy image 201) and an image of high-energy radiation (high-energy image 202) with a single irradiation.
  • the control unit 105 stores the radiation image captured by the FPD 102 in the storage unit 108, and transfers the radiation image to the image processing unit 109.
  • step S210 the image generation unit 110 generates a substance separation image which is a substance property image.
  • the image generator 110, the low energy image 201 (X L) and high energy image 202 (X H) the following equation from equation (1, Number 2) material separated image based on taken with FPD102 To generate.
  • ⁇ is the linear attenuation coefficient
  • d is the thickness of the substance
  • E is the energy of radiation
  • N (E) is the spectrum of radiation
  • H and L are high energy and low energy, respectively.
  • a and B indicate fat and bone, respectively.
  • Number 1 in Equation 2 the unknown variables are the thickness d A, a d B.
  • fat and bone are used as examples of substances as an example of separation, but the substance to be separated is not limited to fat and bone, and any substance can be separated.
  • the simultaneous equations of Equation 1 and Equation 2 in optimization techniques, such as Newton-Raphson method or the bisection method, solving for the thickness of each substance (fat thickness d A, bone thickness d B), shown in FIG. 2 It is possible to generate a substance-separated image 203 separated into a fat image and a bone image.
  • the substance separation image 203 illustrates the fat image.
  • step S211 the compositing unit 112 synthesizes radiation based on an image of low-energy radiation (low-energy image 201 ( XL )) and an image of high-energy radiation (high-energy image 202 ( XH )).
  • An image (hereinafter referred to as a synthetic radiation image) is generated. That is, the combining unit 112, the low-energy image 201 (X L) and high energy image 202 (X H), based on the following equation (3), to generate a composite radiographic image 204 (XProc).
  • the synthesis unit 112 generates a composite image (composite radiation image 204 (Xproc)) by performing a weighted addition process on a plurality of radiation images.
  • the weighting coefficients a and b are arbitrary positive numbers, and the sum of a and b is set to be 1.
  • the synthesis unit 112 calculates the dose and the number of photons when capturing the low-energy image 201 ( XL ) and the high-energy image 202 ( XH ), and calculates the dose ratio or the photon number ratio in a plurality of radiation images. It is also possible to set the values of the weighting coefficients a and b in the addition process of the equation 3 based on (ratio).
  • the synthesis unit 112 acquires the synthetic radiation spectrum (simply also referred to as “composite spectrum”) constituting the synthetic radiation image 204 (Xproc) Xproc together with the synthesis of the radiation image.
  • FIG. 3 is a diagram showing an outline of the composite spectrum.
  • the radiation spectrum (also simply referred to as the "spectrum”) contains information on the number of photons corresponding to the radiation energy.
  • the horizontal axis represents radiation energy (KeV) and the vertical axis represents the number of photons.
  • the storage unit 108 stores the spectrum for each different radiation energy.
  • the synthetic spectrum N proc (E) is, aX L, radiation spectrum N L (E) constituting the bX H, is calculated by calculation N H (E), the radiation was calculated spectrum N L (
  • the composite spectrum N proc (E) can be obtained by adding and averaging E) and NH (E). If the values of the weighting coefficients a and b are known in advance, the low-energy and high-energy radiation spectra are measured with a spectrometer, and the composite spectrum is acquired based on the radiation spectra of each radiation energy measured in advance. It is also possible.
  • step S212 the noise reduction processing unit 111 performs noise reduction processing on the substance separation image 203 generated in step S210.
  • the noise reduction processing unit 111 can perform noise reduction processing on a separated fat image or bone image as a target of noise reduction processing. For example, when acquiring a clearer bone image, the noise reduction processing unit 111 performs noise reduction processing on the fat image, and when acquiring a clearer fat image, the noise reduction processing unit 111 performs bone reduction processing. Noise reduction processing may be performed on the image.
  • the noise reduction processing unit 111 performs noise reduction processing on the fat image and acquires the smoothed fat image 205.
  • the noise reduction processing unit 111 can apply a structure preservation type noise reduction processing that does not lose the edge structure.
  • noise reduction methods such as a bilateral filter, NonLocalMeans, and an epsilon filter.
  • the noise reduction processing unit 111 can also apply a noise reduction method such as a Gaussian filter that reduces high frequency components to an image having a small edge structure such as a fat image.
  • step S213 the image generation unit 110 generates a noise-reduced substance-separated image (bone image 206).
  • the image generator 110 the low energy image 201 taken by the FPD102 and (X L) higher energy image 202 (X H) synthesized composite radiation image 204: and (XProc S211), the noise reduction A noise-reduced bone image 206 is generated from the fat image 205 (S212) and the noise-reduced bone image 206 based on the following equation (Equation 4).
  • ⁇ is the linear attenuation coefficient
  • d is the thickness of the substance
  • E is the energy of radiation
  • N (E) is the spectrum of radiation
  • proc is synthetic
  • subscripts A and B are fat, respectively. And show the bone.
  • the image generation unit 110 generates, for example, a fat image as the first substance characteristic image (S210), and the noise reduction processing unit 111 performs noise reduction processing on the fat image (S212). ).
  • the image generation unit 110 has different radiation energies from the composite image (composite radiation image 204 (Xproc)) and (S211) acquired from a plurality of radiation images, the first material property image (fat image) subjected to noise reduction processing, and the radiation energy.
  • synthetic spectrum N proc (E) synthetic radiation spectrum obtained from the spectrum
  • a noise-reduced second material characteristic image (bone image) is generated (S213).
  • the image generation unit 110 generates an image showing the thickness (or density) of the first substance (for example, fat) constituting the plurality of substances as the first substance property image, and generates the second substance property.
  • an image an image showing the thickness (or density) of a second substance (for example, bone) constituting a plurality of substances is generated.
  • fat and bone are used as examples of substances as an example of separation, but the substance to be separated is not limited to fat and bone, and any substance can be separated.
  • the equation of the number 4 in optimization techniques, such as Newton-Raphson method or the bisection method, the thickness of the material (e.g., d B the thickness of the bone) is solved for, obtaining bone image 206 from which noise has been reduced is shown in FIG. 2 be able to.
  • This embodiment is an example, and is not limited to the processing flow described in FIG.
  • performs noise reduction processing with respect to the sum of separate bone and fat images image solving the equation of Equation 4 for the thickness of each substance (fat thickness d A, bone thickness d B)
  • Number 4 of the equation, solving for the thickness of the material (the thickness d B of the bone) it is possible to obtain the noise reduced bone image as described above, the thickness of the material (fat thickness d A ) Can be solved to obtain a noise-reduced fat image.
  • a plurality of images having different radiation energies (low energy image 201, high energy image 202) are prepared and separated into a plurality of substances (for example, a plurality of fat images and a plurality of bone images). It is possible. Further, the process of FIG. 2 may be alternately repeated for each substance characteristic image, and further noise reduction processing may be performed.
  • the noise reduction processing unit reduces the noise component of the second substance characteristic image (for example, bone image) generated in step S213 by the noise reduction processing (S212). Then, the image generation unit 110 uses the composite image (S211 and 204), the second material characteristic image in which the noise component is reduced, and the composite spectrum to reduce the noise in the first material characteristic image (for example, A fat image) is further generated (S213).
  • the image generation unit 110 uses the composite image (S211 and 204), the second material characteristic image in which the noise component is reduced, and the composite spectrum to reduce the noise in the first material characteristic image (for example, A fat image) is further generated (S213).
  • the image generation unit 110 uses a composite image (S211 and 204), a first substance characteristic image (S212, fat image 205) in which the noise component is reduced, and a second substance whose noise is reduced by using the composite spectrum.
  • a process (first process) for generating a characteristic image (S213, bone image 206) is executed. Further, the image generation unit 110 uses the composite image (S211 and 204), the second material characteristic image (S212, bone image 206) in which the noise component is reduced, and the composite spectrum to reduce the noise.
  • the process (second process) for generating the material property image (S213, fat image 205) of the above is executed. Then, the image generation unit 110 alternately and repeatedly executes the first process and the second process.
  • the spatial distribution of the effective atomic number and the surface density can be obtained by solving the following equations 5 and 6.
  • E is the energy
  • N (E) is the radiation spectrum
  • ⁇ (Z eff , E) is the effective atomic number Z eff and the mass attenuation coefficient at the energy E
  • D eff is the areal density. ..
  • the subscripts H and L indicate high energy and low energy, respectively.
  • the unknown variables in Eqs. 5 and 6 are the effective atomic number Z eff and the areal density D eff . Therefore, as in the case of obtaining the case of obtaining the spatial distribution of the thickness of each substance (substance separation image), the number of images (multiple radiation images (low energy image, high energy image)) taken by radiation of two different energies is counted. Since two independent equations can be generated by substituting into equations 5 and 6, it is possible to obtain the values of the effective atomic number Z eff and the surface density D eff by solving the two independent equations. it can.
  • the synthetic radiation image 204 (Xproc) can be generated by using the equation 3 described above, and the noise is reduced by solving the following equation 7 instead of the equation 4.
  • An identification image can be generated.
  • Equation 7 If the equation of Equation 7 is solved by an optimization method such as Newton's method or dichotomy, a noise-reduced noise-reduced substance identification image (an image showing the distribution of effective atomic number Z eff or area density D eff ) can be obtained. Can be obtained.
  • an optimization method such as Newton's method or dichotomy
  • E energy
  • N (E) is the spectrum of radiation
  • ⁇ (Z eff , E) is the effective atomic number Z eff and the mass attenuation coefficient at energy E
  • D eff is the areal density.
  • proc indicates composition.
  • the image generation unit 110 generates a surface density image as the first substance characteristic image (S210), and the noise reduction processing unit 111 performs noise reduction processing on the surface density image (S210). S212).
  • the image generation unit 110 has different radiation energies from the composite image (composite radiation image 204 (Xproc)) and (S211) acquired from the plurality of radiation images and the noise-reduced first material characteristic image (surface density image).
  • a second material characteristic image an image showing the distribution of effective atomic numbers
  • N proc (E) synthetic radiation spectrum obtained from the spectrum of (S213). That is, the image generation unit 110 generates a surface density image as the first substance characteristic image (S210) and an effective atomic number image as the second substance characteristic image (S213).
  • the image generation unit 110 generates an effective atomic number image as the first material characteristic image (S210), and the noise reduction processing unit 111 performs noise reduction processing on the effective atomic number image.
  • the image generation unit 110 includes a composite image (composite radiation image 204 (Xproc)) and (S211) acquired from a plurality of radiation images, and a noise-reduced first substance property image (effective atomic number image).
  • a second material characteristic image (plane density image) with reduced noise is generated using a synthetic spectrum (synthetic radiation spectrum N proc (E)) obtained from spectra of different radiation energies (S213). That is, the image generation unit 110 generates an image showing the effective atomic number as the first substance characteristic image (S210), and generates an area density image as the second substance characteristic image (S213).
  • the effective atomic number of iodine contained in a contrast medium or the like is 53
  • the effective atomic number of barium is 56
  • the effective atomic number of stainless steel as a member used for a catheter guide wire or the like is 26
  • the effective atomic number of titanium is 22.
  • the image generation unit 110 can generate an image that identifies a substance inside the human body (subject) according to the photographing technique, and the generated image is virtual.
  • a monochromatic radiographic image it can also be used for other processing such as angiography (DSA: Digital Subtraction Angiography) using energy subtraction.
  • DSA Digital Subtraction Angiography
  • the optimization method since the optimization method is used for each pixel during the calculation processing of the energy subtraction, it may take a predetermined time to generate the substance characteristic image.
  • the second embodiment a configuration will be described in which the calculation processing result of the energy subtraction is stored in advance in a table or the like, and the substance property image is acquired in a short time by referring to the table when the energy subtraction is processed.
  • the configuration of the present embodiment has an advantageous effect when real-time performance such as when seeing through is required.
  • the image generation unit 110 generates a material characteristic image of a substance contained in a plurality of radiation images taken with different radiation energies using a radiation energy spectrum, and corresponds to the plurality of shot radiation images.
  • a first table for example, table 501 shown in 5A of FIG. 5 and table 502 shown in 5B of FIG. 5 for outputting a material property image is generated (S410).
  • the noise reduction processing unit 111 reduces the noise component of the material characteristic image. That is, the noise reduction processing unit 111 reduces the noise component of the substance characteristic image output from the first table (for example, table 501 shown in 5A of FIG. 5 and table 502 shown in 5B of FIG. 5).
  • the image generation unit 110 acquires from a composite image (S412, 404) acquired from a plurality of radiation images, a noise-reduced first material property image (S413,405), and different radiation energy spectra. Using the combined spectrum obtained, a second table (for example, 601 in FIG. 6) was generated based on the calculation result of generating a noise-reduced second material characteristic image (S414), and the composite image (404) and noise were generated. The second material property image (S415, 406) corresponding to the reduced first material property image (405) is output from the second table (601).
  • step S410 the image generation unit 110 performs energy subtraction processing in advance and tabulates the calculation results.
  • the image generation unit 110 has a plurality of substances constituting the human body, from the combination of the thicknesses of the substances that can be taken (for example, the combination of the thickness of fat as the first substance and the thickness of bone as the second substance). Solve the simultaneous equations of Equation 2 to obtain a low-energy image ( XL ) and a high-energy image (X H ). For the analysis of two simultaneous equations, it is possible to use an optimization method such as Newton's Rapson method.
  • the image generation unit 110 determines the correspondence between the low-energy image ( XL ) acquired from the simultaneous equations of the equations 1 and 2, the high-energy image (X H ), and the thickness of the first substance (for example, fat). It is generated as the table 501 shown in 5A of FIG. 5 and stored in the storage unit 108. Further, the image generation unit 110 corresponds to the low energy image ( XL ) acquired from the simultaneous equations of the equations 1 and 2, the high energy image (X H ), and the thickness of the second substance (for example, bone). The relationship is generated as the table 502 shown in 5B of FIG. 5 and stored in the storage unit 108.
  • the image generation unit 110 can improve the accuracy of the table by finely setting the step size on the vertical axis and the horizontal axis of the table to be generated. If the step size is set finely, it takes time to generate the table. Therefore, the image generation unit 110 can arbitrarily set the step size of the table so as to obtain desired accuracy.
  • the vertical and horizontal axes of the table may be set linearly or logarithmically. Further, the setting of the vertical axis and the horizontal axis of the table does not have to be continuous because it is possible to interpolate using the values of the setting values in the vicinity.
  • step S411 the image generation unit 110 uses the table (5A in FIG. 5 and 5B in FIG. 5) generated in step S410, the low-energy image 401 ( XL ), and the high-energy image 402 ( XH ). Acquire a substance separation image which is a substance property image.
  • the image generating unit 110 based on the value of each pixel of the low-energy image 401 (X L) and high energy image 402 (X H), table (fat thickness table (5A in Figure 5), bone With reference to the thickness table (5B in FIG. 5), a substance separation image (fat thickness image and bone thickness image) which is a substance property image is acquired.
  • the values in the table can be interpolated by a method such as bilinear interpolation or logarithmic interpolation.
  • step S412 Synthesis of radiation image and radiation spectrum
  • the synthesis unit 112 converts the low-energy radiation image (low-energy image 401 ( XL )) and the high-energy radiation image (high-energy image 402 ( XH )) into images. Based on this, a combined radiation image (hereinafter referred to as a composite radiation image) is generated. That is, the combining unit 112, the low-energy image 401 (X L) and high energy image 402 (X H), based on the equation (3), to generate a composite radiographic image 404 (Xproc).
  • step S413 Noise reduction processing
  • the noise reduction processing unit 111 performs noise reduction processing on the substance separation image 403 acquired in step S411.
  • the noise reduction processing unit 111 can perform noise reduction processing on a separated fat image or bone image as a target of noise reduction processing. For example, when acquiring a clearer bone image, the noise reduction processing unit 111 performs noise reduction processing on the fat image, and when acquiring a clearer fat image, the noise reduction processing unit 111 performs bone reduction processing. Noise reduction processing may be performed on the image.
  • the noise reduction processing unit 111 performs noise reduction processing on the fat image and acquires the smoothed fat image 405.
  • the image generation unit 110 generates a table 601 that associates the thickness of the first substance (for example, fat) with the pixel value of the synthetic radiation image and the thickness of the second substance (for example, bone), and stores the storage unit 108.
  • FIG. 6 is a diagram illustrating a table 601 for acquiring a noise-reduced substance characteristic image.
  • the horizontal axis is set to the pixel value of the synthetic radiation image 404 (Xproc), and the vertical axis is set to the thickness of the first substance (fat).
  • the thickness of one substance (fat) and the distribution of the thickness of the second substance (bone) are associated with each other.
  • the image generation unit 110 determines the thickness d B of the second material (bone). Then, the image generator 110, the thickness of the obtained second material (bone) d B, the pixel value of the combined radiographic image 404 (XProc), first material table associating the thickness d A of the (fat) Generate 601.
  • step S415 the image generation unit 110 generated an image (smoothed) of the thickness of the table 601 and the synthetic radiation image 404 (Xproc) generated in step S414 and the noise-reduced first substance (fat) in step S413.
  • an image (bone image 406) of the thickness of the second substance (bone) is acquired.
  • Image generating unit 110 refers to the table 601, the first substance (fat) thick d A and the second material from the pixel values of the composite radiation image 404 (XProc) from which noise has been reduced in step S413 (bone) it is possible to obtain the thickness d B, to obtain a bone image from which noise has been reduced from the thickness d B of the second material (bone).
  • a substance separation image which is a substance characteristic image can be acquired while reducing the calculation load based on the equations 1 and 2. Can be done. Further, by referring to the table 601 in step S414, it is possible to acquire a substance separation image which is a noise-reduced substance characteristic image while reducing the calculation load based on the equation (4). According to the process of the second embodiment described with reference to FIG. 4, it can be applied to moving image shooting that requires higher speed image processing.
  • FIG. 7 is a diagram illustrating the effect of the embodiment of the present invention.
  • 7A of FIG. 7 is a diagram illustrating the configuration of a phantom for measuring the effect.
  • an image by low-energy radiation (low-energy image) and an image by high-energy radiation (high-energy image) are acquired, and an embodiment of the present invention (for example, the first embodiment) is obtained.
  • the result of determining the thickness of the substance constituting the phantom based on the process of the second embodiment) is the waveform 701 of 7B in FIG.
  • the result of using the energy subtraction method disclosed in Patent Document 1 is the waveform 702 of 7B in FIG. 7.
  • the composition of the phantom is composed of acrylic resin (PMMA) and aluminum (AL).
  • Acrylic resin (PMMA) corresponds to fat, which is a soft substance
  • aluminum (AL) corresponds to bone, which is a hard substance. ..
  • the thickness of aluminum (AL) is constant at 1 cm.
  • acrylic resin (PMMA) thickness 15 cm, 20 cm, and 25 cm.
  • the horizontal axis shows the thickness of acrylic resin (PMMA), and the vertical axis shows the error with respect to the thickness of aluminum (AL) of 1 cm.
  • the thickness error of aluminum (AL) is ⁇ 7 with respect to the thickness of acrylic resin (PMMA). It varies in the range of% to + 4%.
  • the thickness error of aluminum (AL) is about ⁇ 1 with respect to the thickness of acrylic resin (PMMA). It is within the range of%.
  • An embodiment of the present invention can reduce an error as compared with the processing of Patent Document 1 which is a prior art, and the embodiment of the present technology is effective in the case of measurement in which quantitativeness is required in image processing. ..
  • the acrylic resin (PMMA) has a different thickness, and is characterized by excellent quantitativeness even if beam hardening occurs.
  • the present invention supplies a program that realizes one or more functions of the above-described embodiment to a system or device via a network or storage medium, and one or more processors in the computer of the system or device reads and executes the program. It can also be realized by the processing to be performed. It can also be realized by a circuit (for example, ASIC) that realizes one or more functions.
  • a circuit for example, ASIC
  • 100 Radiation imaging system
  • 101 Radiation source
  • 102 FPD (Radiation detection device)
  • 104 Radiation generator
  • 105 Control unit
  • 106 Monitor (display unit)
  • 107 Operation unit
  • 108 Storage unit
  • 109 Image processing unit
  • 110 Generation unit
  • 111 Noise reduction processing unit
  • 112 Synthesis unit
  • 120 Image processing device

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Medical Informatics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • General Health & Medical Sciences (AREA)
  • Radiology & Medical Imaging (AREA)
  • Pathology (AREA)
  • Surgery (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Molecular Biology (AREA)
  • Optics & Photonics (AREA)
  • Animal Behavior & Ethology (AREA)
  • Veterinary Medicine (AREA)
  • Public Health (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Quality & Reliability (AREA)
  • Apparatus For Radiation Diagnosis (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)
PCT/JP2020/016777 2019-05-30 2020-04-16 画像処理装置、画像処理方法及びプログラム Ceased WO2020241110A1 (ja)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US17/529,327 US12136191B2 (en) 2019-05-30 2021-11-18 Image processing apparatus, image processing method, and non-transitory computer-readable storage medium

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2019101483A JP7313194B2 (ja) 2019-05-30 2019-05-30 画像処理装置、画像処理方法及びプログラム
JP2019-101483 2019-05-30

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/529,327 Continuation US12136191B2 (en) 2019-05-30 2021-11-18 Image processing apparatus, image processing method, and non-transitory computer-readable storage medium

Publications (1)

Publication Number Publication Date
WO2020241110A1 true WO2020241110A1 (ja) 2020-12-03

Family

ID=73545395

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2020/016777 Ceased WO2020241110A1 (ja) 2019-05-30 2020-04-16 画像処理装置、画像処理方法及びプログラム

Country Status (3)

Country Link
US (1) US12136191B2 (enExample)
JP (1) JP7313194B2 (enExample)
WO (1) WO2020241110A1 (enExample)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022181022A1 (ja) * 2021-02-26 2022-09-01 キヤノン株式会社 画像処理装置及び方法、放射線撮像システム、プログラム

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020003744A1 (ja) * 2018-06-27 2020-01-02 キヤノン株式会社 放射線撮影装置、放射線撮影方法およびプログラム

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH03285475A (ja) * 1990-02-14 1991-12-16 Fuji Photo Film Co Ltd エネルギーサブトラクション画像生成方法
JP2008062035A (ja) * 2006-09-08 2008-03-21 General Electric Co <Ge> 対象の多重スペクトル画像を形成する方法及びシステム
JP2008125900A (ja) * 2006-11-22 2008-06-05 Ge Medical Systems Global Technology Co Llc Ｘ線断層撮影装置
JP2009506854A (ja) * 2005-09-06 2009-02-19 コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ 複数エネルギー窓をもつ計算機断層撮影におけるデータ処理および解析
JP2014210180A (ja) * 2013-04-04 2014-11-13 株式会社東芝 Ｘ線コンピュータ断層撮影装置
WO2018187735A1 (en) * 2017-04-07 2018-10-11 Photo Diagnostic Systems, Inc. Method for artifact reduction using monoenergetic data in computed tomography

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5485371A (en) * 1990-02-14 1996-01-16 Fuji Photo Film Co., Ltd. Method for forming energy subtraction radiation images, and method and apparatus for smoothing radiation images
JP2003180669A (ja) * 2001-12-18 2003-07-02 Shimadzu Corp Ｘ線撮影装置
DE602004003521T2 (de) * 2003-06-26 2007-10-04 Fujifilm Corporation Verfahren, Vorrichtung und Programm zur Bildverarbeitung und zur Detektion von abnormalen Schatten
EP1712179B1 (en) * 2005-04-13 2010-12-15 Tanita Corporation Trunk visceral fat measuring method and apparatus
US7742568B2 (en) * 2007-06-09 2010-06-22 Spectrum San Diego, Inc. Automobile scanning system
EP2676128B1 (en) * 2011-02-18 2019-08-21 Smiths Heimann GmbH System and method for multi-scanner x-ray inspection
SK6148Y1 (sk) 2011-08-22 2012-06-04 Boris Hudak Bike with modifiable design to climb, descent and terrain roughness
US9458524B2 (en) * 2013-03-05 2016-10-04 Cabot Corporation Methods to recover cesium or rubidium from secondary ore
JP6747787B2 (ja) * 2014-08-22 2020-08-26 キヤノンメディカルシステムズ株式会社 光子計数型ｘ線ｃｔ装置
JP7054329B2 (ja) 2017-10-06 2022-04-13 キヤノン株式会社 画像処理装置、画像処理方法及びプログラム

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH03285475A (ja) * 1990-02-14 1991-12-16 Fuji Photo Film Co Ltd エネルギーサブトラクション画像生成方法
JP2009506854A (ja) * 2005-09-06 2009-02-19 コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ 複数エネルギー窓をもつ計算機断層撮影におけるデータ処理および解析
JP2008062035A (ja) * 2006-09-08 2008-03-21 General Electric Co <Ge> 対象の多重スペクトル画像を形成する方法及びシステム
JP2008125900A (ja) * 2006-11-22 2008-06-05 Ge Medical Systems Global Technology Co Llc Ｘ線断層撮影装置
JP2014210180A (ja) * 2013-04-04 2014-11-13 株式会社東芝 Ｘ線コンピュータ断層撮影装置
WO2018187735A1 (en) * 2017-04-07 2018-10-11 Photo Diagnostic Systems, Inc. Method for artifact reduction using monoenergetic data in computed tomography

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
IWAMOTO, SHIN- ICHIRO: "Educational lecture materials] X-ray attenuation coefficient in diagnostic imaging", MEDICAL IMAGING AND INFORMATION SCIENCES, vol. 32, no. 3, 2015, pages 54 - 62 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022181022A1 (ja) * 2021-02-26 2022-09-01 キヤノン株式会社 画像処理装置及び方法、放射線撮像システム、プログラム

Also Published As

Publication number Publication date
US20220076397A1 (en) 2022-03-10
US12136191B2 (en) 2024-11-05
JP2020192268A (ja) 2020-12-03
JP7313194B2 (ja) 2023-07-24

Similar Documents

Publication Publication Date Title
US9345443B2 (en) Calibration free dual energy radiography method
US8054940B2 (en) Image capturing system for medical use, image processing apparatus, and image processing method
JP7093233B2 (ja) 放射線撮影装置、放射線撮影方法およびプログラム
EP3612095B1 (en) Beam hardening correction in x-ray dark-field imaging
JP2017143943A (ja) 放射線画像処理装置、方法およびプログラム
US11813095B2 (en) Radiation imaging apparatus, radiation imaging method, and non-transitory computer-readable storage medium
JPWO2018235823A1 (ja) Ｘ線装置、ｘ線検査方法、及びデータ処理装置
JPS6287136A (ja) 像を較正する方法
JP2018038646A (ja) 画像処理装置、方法およびプログラム
JP7169853B2 (ja) 画像処理装置、放射線撮影装置、および画像処理方法
JP2017080343A (ja) 放射線撮像システム、放射線画像の情報処理装置、放射線画像の情報処理方法、及び、そのプログラム
US12136191B2 (en) Image processing apparatus, image processing method, and non-transitory computer-readable storage medium
JP2017080342A (ja) 放射線撮像システム、放射線画像の情報処理装置、放射線画像の情報処理方法、及び、そのプログラム
JPH0448453B2 (enExample)
JP7144988B2 (ja) 放射線撮影装置、放射線撮影方法およびプログラム
JP7144987B2 (ja) 放射線撮影装置、放射線撮影方法およびプログラム
EP3773214B1 (en) Cross directional bilateral filter for ct radiation dose reduction
WO2021095447A1 (ja) 画像処理装置、放射線撮影装置、画像処理方法及びプログラム
US20100061654A1 (en) Scatter estimation and reduction method and apparatus
JP2018117900A (ja) 放射線撮影装置、放射線撮影システム、放射線撮影方法、及びプログラム
Butzer MARS-CT: Biomedical Spectral x-ray imaging with Medipix
Zhang et al. Perfusion Computed Tomography Image Reconstruction Using Spatio-Temporal Constrained Edge-preserving Prior

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: 20813111

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20813111

Country of ref document: EP

Kind code of ref document: A1