WO2018129889A1 - 基于医学影像的组织元素质量比例解构方法及几何模型建立方法 - Google Patents

基于医学影像的组织元素质量比例解构方法及几何模型建立方法 Download PDF

Info

Publication number
WO2018129889A1
WO2018129889A1 PCT/CN2017/092745 CN2017092745W WO2018129889A1 WO 2018129889 A1 WO2018129889 A1 WO 2018129889A1 CN 2017092745 W CN2017092745 W CN 2017092745W WO 2018129889 A1 WO2018129889 A1 WO 2018129889A1
Authority
WO
WIPO (PCT)
Prior art keywords
tissue
medical image
mass ratio
image data
geometric model
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/CN2017/092745
Other languages
English (en)
French (fr)
Inventor
刘渊豪
萧明城
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.)
Neuboron Medtech Ltd
Original Assignee
Neuboron Medtech Ltd
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 Neuboron Medtech Ltd filed Critical Neuboron Medtech Ltd
Priority to EP17891253.1A priority Critical patent/EP3556431A4/en
Priority to JP2019537223A priority patent/JP6938646B2/ja
Priority to RU2019122188A priority patent/RU2736917C1/ru
Publication of WO2018129889A1 publication Critical patent/WO2018129889A1/zh
Priority to US16/439,980 priority patent/US11590360B2/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
    • A61N5/00—Radiation therapy
    • A61N5/10—X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
    • A61N5/00—Radiation therapy
    • A61N5/10—X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy
    • A61N5/103—Treatment planning systems
    • A61N5/1031—Treatment planning systems using a specific method of dose optimization
    • 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
    • A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
    • A61N5/00—Radiation therapy
    • A61N5/10—X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy
    • A61N5/103—Treatment planning systems
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three-dimensional [3D] modelling for computer graphics
    • 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
    • G16H30/00—ICT specially adapted for the handling or processing of medical images
    • G16H30/40—ICT specially adapted for the handling or processing of medical images for processing medical images, e.g. editing
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
    • A61N5/00—Radiation therapy
    • A61N5/10—X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy
    • A61N5/103—Treatment planning systems
    • A61N5/1031—Treatment planning systems using a specific method of dose optimization
    • A61N2005/1034—Monte Carlo type methods; particle tracking
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61N—ELECTROTHERAPY; MAGNETOTHERAPY; RADIATION THERAPY; ULTRASOUND THERAPY
    • A61N5/00—Radiation therapy
    • A61N5/10—X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy
    • A61N2005/1085—X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy characterised by the type of particles applied to the patient
    • A61N2005/109—Neutrons
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2210/00—Indexing scheme for image generation or computer graphics
    • G06T2210/41—Medical
    • 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
    • G16H50/50—ICT specially adapted for medical diagnosis, medical simulation or medical data mining; ICT specially adapted for detecting, monitoring or modelling epidemics or pandemics for simulation or modelling of medical disorders

Definitions

  • One aspect of the present invention relates to a tissue element mass ratio deconstruction method, and more particularly to a medical image based tissue element mass ratio deconstruction method; and another aspect of the present invention relates to a geometric model establishing method, and more particularly to a medical image based method.
  • the geometric model is built.
  • neutron capture therapy combines the above two concepts, such as boron neutron capture therapy, by the specific agglomeration of boron-containing drugs in tumor cells, combined with precise neutron beam regulation, providing better radiation than traditional radiation. Cancer treatment options.
  • BNCT Boron Neutron Capture Therapy
  • Three-dimensional models are widely used in scientific experimental analysis and scientific experimental simulation.
  • in the field of nuclear radiation and protection in order to simulate the absorbed dose of a living body under certain radiation conditions, it is often necessary to use computer technology to perform various processing on medical images to establish a precise lattice model required by Monte Carlo software, and combined with software. Simulation calculation.
  • the Monte Carlo method is currently a tool for accurately simulating the collision trajectory and energy distribution of nuclear particles within the three-dimensional space of the irradiation target.
  • the combination of the Monte Carlo method and the complex three-dimensional human anatomical model represents the leap of simulation in computer technology.
  • Accurate human body dose assessment is very beneficial for radiation therapy in diagnostic radiology.
  • a variety of human body models have been successfully established internationally and combined with Monte Carlo simulation procedures to absorb the absorbed dose of the human body in a radiation environment. Perform a computational assessment of accuracy.
  • the geometric description required for the successful conversion of the human three-dimensional anatomical model to the Monte Carlo program is a prerequisite for Monte Carlo simulation calculation, and it is also a hotspot and a difficult point in the international Monte Carlo simulation study.
  • Medical imaging such as Magnetic Resonance Imaging (MRI) or Computed Tomography (CT) can provide detailed tissue geometry information for human body features, providing a data foundation for solid modeling of human internal structures.
  • CT image data is converted into a voxel prosthesis model similar to the human anatomy, thereby simulating the migration history of the radiation and calculating the radiation absorbed dose with the action of the human tissue. Therefore, when the elements of a given voxel prosthesis are composed, they are as close as possible to the actual situation of the patient, so as to obtain the result closest to the real situation in the simulation calculation.
  • the human body tissues should be composed of specific element mass ratios instead of the fixed element mass ratios in a single interval. It can be seen that there are still considerable room for improvement in the current common voxel prosthetic component definition methods.
  • An aspect of the present invention provides a tissue image quality ratio deconstruction method based on a medical image, comprising the step of approximating an elemental mass ratio of tissue corresponding to unknown medical image data by tissue corresponding medical image data of a known element mass ratio.
  • the method uses the known tissue element mass ratio to find the element mass ratio of other tissues by medical image data, and provides more tissue types for establishing a voxel prosthesis based on medical image conversion, which can be used for the surrounding area.
  • the Carlo treatment planning system through a larger amount of tissue element mass ratio, specifies the correct physical action data such as the linear attenuation coefficient of photons, the action cross section of neutrons, and charged particles (protons, electrons) in Monte Carlo calculations. Or the quality barrier of the charged particles, providing a near-real particle migration environment.
  • said medical image data is associated with a photon linear attenuation coefficient by which an effective linear attenuation coefficient of tissue of a known element mass ratio can be obtained.
  • the step of approximating the element mass ratio of the tissue corresponding to the unknown medical image data by the medical image data corresponding to the tissue of the known element mass ratio comprises:
  • the photon linear attenuation coefficient of the basic element provided by the NIST is weighted by the medical image beam energy spectrum to obtain an average linear attenuation coefficient ⁇ i under the medical image beam;
  • the linear attenuation coefficient of each tissue is corrected to become an effective linear attenuation coefficient of each tissue.
  • the derivation algorithm is used to obtain the proportion distribution of the chemical elements in the bio-image corresponding to each medical image data, that is, mi .
  • the medical image data is a CT value
  • the CT value range is divided from -100 to +1530 into 1630/n groups.
  • the derivation algorithm is a maximum approximation algorithm.
  • Another aspect of the present invention provides a medical image-based geometric model establishing method, including:
  • the step of generating a geometric model is the step of generating a geometric model.
  • the geometric model establishing method can determine the number of tissue groups according to actual needs, thereby providing the tissue type, element composition and density more accurately, and the established geometric model is more matched with the medical image. The real situation reflected.
  • the geometric model building method is applied to neutron capture therapy, which further includes the steps of giving a B-10 concentration and establishing a 3D coding matrix with B-10 concentration information.
  • the number of organizational clusters is the number of organizational clusters manually defined by the user plus the number of four organizational groups in the database or the number of 14 organizational clusters or 70 organizational clusters or the organizational element mass decomposed method to determine 163 organizational clusters. number. If the corresponding number of organizational clusters is not established in the existing database, the user can customize a new number of organizational clusters. This avoids the fact that if the existing database does not completely match the corresponding number of organizational clusters, it can only approximate the selection, thereby effectively improving the accuracy of the modeling.
  • the geometric model building method further comprises the steps of establishing a 3D tissue coding matrix and establishing a 3D density coding matrix. According to the corresponding slice of the medical image, the corresponding tissue coding and density coding are established for each slice, thereby establishing a 3D tissue coding matrix and a 3D density coding matrix.
  • the geometry model includes the lattice card, cell card, surface card, and material card required for the Monte Carlo software input file.
  • the image finally obtains the lattice card, the cell card, the surface card and the material card required for the Monte Carlo software input file, thus providing a theoretical basis for the simulation calculation and obtaining accurate simulation calculation results.
  • Medical imaging data can be Magnetic Resonance Imaging (MRI), Computed Tomography (CT), Positron Emission Tomography (PET), PET-CT or X-ray imaging ( X-Ray imaging), the following example will be based on electronic computed tomography (CT) data, the file format of CT is usually DICOM.
  • CT Magnetic Resonance Imaging
  • CT Computed Tomography
  • PET Positron Emission Tomography
  • PET-CT PET-CT
  • X-Ray imaging X-ray imaging
  • the tissue element mass ratio deconstruction method and the geometric model establishing method based on medical image data in the embodiment of the present invention deconstruct the mass ratio of the living tissue element through the medical image data of the known tissue, which can directly correspond to the living tissue according to the CT value.
  • the composition of each element makes the voxel prosthesis more similar to the real situation, which can improve the accuracy of the dose calculation accuracy.
  • Figure 1 is a schematic diagram of a boron neutron capture reaction.
  • Figure 2 is a 10 B(n, ⁇ ) 7 Li neutron capture nuclear reaction equation.
  • FIG. 3 is a logic block diagram of a medical image based tissue element quality ratio deconstruction method in an embodiment of the present invention.
  • FIG. 4 is a logic block diagram of a method for establishing a geometric image based on a medical image in an embodiment of the present invention.
  • Figure 5 is a plot of the CT value (HU) and tissue density regression curve and the correlation coefficient.
  • medical image-based tissue element mass ratio deconstruction method and geometric model establishment method for neutron capture therapy are embodiments of the present invention.
  • the following is a brief introduction to neutron capture therapy, especially boron neutron capture therapy.
  • Neutron capture therapy has been increasingly used as an effective means of treating cancer in recent years, with boron neutron capture therapy being the most common, and neutrons supplying boron neutron capture therapy can be supplied by nuclear reactors or accelerators.
  • Embodiments of the invention take the accelerator boron neutron capture treatment as an example.
  • the basic components of the accelerator boron neutron capture treatment typically include an accelerator, target and heat removal for accelerating charged particles (eg, protons, deuterons, etc.).
  • Systems and beam shaping bodies in which accelerated charged particles interact with metal targets to produce neutrons, depending on the desired neutron yield and energy, the energy and current of the accelerated charged particles, and the physicochemical properties of the metal target.
  • the nuclear reactions that are often discussed are 7 Li(p,n) 7 Be and 9 Be(p,n) 9 B, both of which are endothermic.
  • the energy thresholds of the two nuclear reactions are 1.881 MeV and 2.055 MeV, respectively. Since the ideal neutron source for boron neutron capture therapy is the superheated neutron of the keV energy level, theoretically, if proton bombardment with energy only slightly higher than the threshold is used.
  • a metallic lithium target that produces relatively low-energy neutrons that can be used clinically without too much slow processing.
  • proton interaction cross sections for lithium metal (Li) and base metal (Be) targets and threshold energy Not high, in order to generate a sufficiently large neutron flux, a higher energy proton is usually used to initiate the nuclear reaction.
  • BNCT Boron Neutron Capture Therapy
  • boron-containing ( 10 B) drugs with 10 B(n, ⁇ ) 7 Li neutron capture and nuclear splitting reactions.
  • Two heavy charged particles of 4 He and 7 Li are produced. 1 and 2, which respectively show a schematic diagram of a boron neutron capture reaction and a 10 B(n, ⁇ ) 7 Li neutron capture nuclear reaction equation, the average energy of the two charged particles is about 2.33 MeV, which has high linearity.
  • Linear Energy Transfer (LET) short-range characteristics, the linear energy transfer and range of ⁇ particles are 150 keV/ ⁇ m and 8 ⁇ m, respectively, while the 7 Li heavy particles are 175 keV/ ⁇ m and 5 ⁇ m.
  • the total range of the two particles is approximately equivalent.
  • a cell size so the radiation damage caused by the organism can be limited to the cell level.
  • the boron-containing drug is selectively accumulated in the tumor cells, with appropriate neutron source, it can cause too much damage to normal tissues. Under the premise, the purpose of locally killing tumor cells is achieved.
  • the purpose of embodiments of the present invention is to establish a medical image-based tissue element mass ratio deconstruction method closer to a living tissue structure and a geometric model establishing method including the deconstruction method for performing tissue dose simulation calculation of boron neutron capture treatment.
  • Medical imaging data can be Magnetic Resonance Imaging (MRI), Computed Tomography (CT), Positron Emission Tomography (PET), PET-CT or X-ray imaging ( X-Ray imaging), in this embodiment, based on the data of the computed tomography scan, the file format of the CT is usually DICOM.
  • MRI Magnetic Resonance Imaging
  • CT Computed Tomography
  • PET Positron Emission Tomography
  • PET-CT PET-CT
  • X-Ray imaging X-ray imaging
  • the tissue element mass ratio deconstruction method is based on the relationship between the chemical element mass ratio distribution and the CT value in the living body, and the CT value, also known as the Hounsfield Unit (HU or H), is the linear attenuation of each tissue and water.
  • the coefficient ratio is the unit of the attenuation coefficient of the reactive light, which is defined as Equation 1:
  • the linear attenuation coefficient of each organization is the product of the mass ratio of each element (m i ; 0 ⁇ m i ⁇ 1) and the linear attenuation coefficient ( ⁇ i ) of each element.
  • the linear attenuation coefficient of each element in the energy range of the X-ray generated by CT is weighted and summed by the photon linear attenuation coefficient of the basic elements provided by the National Institute of Standards and Technology (NIST).
  • the average linear attenuation coefficient was obtained as the linear attenuation coefficient ( ⁇ i ) of each element calculated in the present example.
  • the product of the detector efficiency and the enhancement factor in the CT scanner is used as the efficiency weighting factor C to correct the linear attenuation coefficient of each organization to become the effective linear attenuation coefficient of each organization.
  • the mass ratio of the tissue elements of the similar CT values is taken as an initial hypothesis, and the derivation algorithm can obtain the proportion distribution of the chemical elements in the living body corresponding to each CT value, that is, mi , so that the CT image data can be obtained.
  • the element composition ratio corresponding to each voxel unit, and then a three-dimensional voxel prosthesis model is established to perform a treatment plan simulation based on Monte Carlo algorithm.
  • MCNP software is preferably used for simulation.
  • CT values correspond to the organization of 70 different elements (% by weight of elements)
  • the medical image-based tissue element quality ratio deconstruction method disclosed in the embodiment of the present invention mainly includes the following processes:
  • the photon linear attenuation coefficient of the basic elements provided by NIST is weighted by the CT beam energy spectrum, and the average linear attenuation coefficients ⁇ i and ⁇ water of each element and liquid water under the CT beam are obtained.
  • the CT value n is a group, and the CT value range is divided into a 1630/n group from -100 to +1530. In this embodiment, n is 10, and a total of 163 groups start with -100.
  • the derivation algorithm in this embodiment is preferably a maximum approximation algorithm.
  • the method for establishing a geometric model based on medical image data disclosed in the embodiments of the present invention mainly includes the following processes:
  • the software will discriminate the position of each voxel point on the CT image, and divide all voxel points into "falling within the ROI boundary" and “fall outside the ROI boundary”;
  • the user can manually define the relative unique tissue type and tissue density for each ROI according to actual needs, or automatically match the CT value with the tissue type and density to avoid
  • the given CT value covers a wide range of ROI (such as mucosal chamber) unique tissue type (element composition) and density, resulting in error in dose calculation;
  • the definition of the organization type is automatically performed. According to the difference in CT value, the organization can be divided into 4 or 14 or 70 or 163 different elements. The user can select according to the actual judgment.
  • the user manually inputs the normal blood boron-containing drug concentration, the tumor-blood boron concentration ratio, the tissue-blood boron concentration ratio, and the like, and the B-10 element is prepared in all voxel points;
  • the software will integrate information such as tissue type (element composition), tissue density, and tissue B-10 concentration to generate a three-dimensional MCNP lattice model, and write a lattice card and a cell in the format specified by the MCNP software input file.
  • tissue type element composition
  • tissue density tissue density
  • tissue B-10 concentration tissue B-10 concentration
  • the method for establishing a geometric model based on medical image data disclosed by the present invention includes two Example.
  • a first embodiment of the present invention provides a geometric model establishing method based on medical image data, comprising: a step of reading medical image data; a step of defining a tissue type by a conversion relationship between medical image data and a tissue type; determining an organization a step of grouping a number; a step of defining a tissue density by a conversion relationship between medical image data and density; a step of creating a 3D coding matrix with tissue and density information; and a step of generating a geometric model.
  • the number of tissue clusters can be determined according to actual needs, thereby providing the tissue type, element composition and density more accurately, and the established geometric model is more compatible with the reality reflected by the medical image data. happening.
  • the geometric model building method further includes the steps of giving a B-10 concentration and establishing a 3D encoding matrix with B-10 concentration information.
  • the number of organizational clusters is 163 organizations defined by the number of organizational groups manually defined by the user plus the number of four organizational groups in the database or the number of 14 tissue groups or 70 tissue groups or the tissue element mass ratio deconstruction method of the present invention.
  • the number of clusters If there is no corresponding number of organizational clusters in the existing database, then a new number of organizational clusters can be determined based on the experimental situation. This avoids the fact that if the existing database does not completely match the corresponding number of organizational clusters, it can only approximate the selection, thereby effectively improving the accuracy of the modeling.
  • the geometric model building method further includes the steps of establishing a 3D tissue coding matrix and establishing a 3D density coding matrix. According to the corresponding transformation relationship of the slices of the medical image data, the corresponding tissue coding and density coding are established for each slice, thereby establishing a 3D tissue coding matrix and a 3D density coding matrix.
  • the geometry model includes the lattice card, cell card, surface card, and material card required for the MCNP software input file.
  • the lattice card, the cell card, the surface card and the material card required for the Monte Carlo software input file are finally obtained, thereby providing a theoretical basis for the simulation calculation and obtaining accurate simulation calculation results.
  • a second embodiment of the present invention provides a geometric model establishing method based on medical image data, comprising: a step of reading medical image data; a step of defining or reading an ROI boundary; and determining whether the medical image voxel is within the ROI boundary Step: If yes, enter the step of manually defining the tissue type and density by the user given a specific tissue and density for each voxel within the boundary of each ROI or entering between medical image data and tissue type/density The conversion relationship automatically defines the ROI organization type and density step. If not, the step of automatically defining the tissue type by the conversion relationship between the medical image data and the tissue type is entered and the tissue density is defined by the conversion relationship between the medical image data and the density. Step; the step of creating a 3D coding matrix with tissue and density information; the step of generating a geometric model.
  • the so-called ROI refers to the region of interest, and the user can manually define the organization type, element composition, and density of the ROI.
  • the definition of the organization type is determined according to the conversion relationship between the medical image data and the tissue type/density, and the number of tissue groups is determined according to actual needs, thereby providing the organization type, element composition and density more accurately, and establishing The geometric model is more closely matched to the real situation reflected by the medical image data.
  • the geometric model building method includes the steps of giving a B-10 concentration and establishing a 3D encoding matrix with B-10 concentration information.
  • the number of organizational clusters is the number of organizational clusters manually defined by the user plus the number of four tissue clusters or 14 tissue clusters or 70 tissue clusters existing in the database, or obtained by the above-described tissue element mass ratio deconstruction method disclosed in the present invention.
  • the geometric model building method further includes the steps of establishing a 3D tissue coding matrix and establishing a 3D density coding matrix. According to the corresponding transformation relationship of the slices of the medical image data, the corresponding tissue coding and density coding are established for each slice, thereby establishing a 3D tissue coding matrix and a 3D density coding matrix.
  • the geometry model includes the lattice card, cell card, surface card, and material card required for the MCNP software input file.
  • the lattice card, the cell card, the surface card and the material card required for the Monte Carlo software input file are finally obtained, thereby providing a theoretical basis for the simulation calculation and obtaining accurate simulation calculation results.
  • CT value-tissue type and CT value-tissue density conversion relationship diagram in the existing database in the geometric model establishing method based on medical image data disclosed in the present invention will be described below.
  • ICRU-46 International Commission on Radiation Units and Measurements, Photon, electron, proton and neutron interaction data for body tissues, ICRU-46, Tech. Rep., 1992.
  • ICRU-46 International Commission on Radiation Units and Measurements, Photon, electron, proton and neutron interaction data for body tissues. It contains air, adult brain, adult skin, cranium, and its corresponding density and element composition are shown in Table 3.
  • HU L and HU U are the lower and upper limits of CT value, respectively.
  • the method uses the known tissue element mass ratio to calculate the element mass ratio of other tissues by using the computed tomography image data, so that more different CT data can be considered in the preparation of the voxel prosthesis, and more tissue types are provided.
  • the prosthesis is closer to the real situation.
  • the prosthesis can be used in a Monte Carlo-based treatment planning system to specify the correct physical action data such as the linear attenuation coefficient of photons and the action cross section of neutrons in Monte Carlo calculations through a larger amount of tissue element mass ratio.
  • the quality barrier with charged particles provides a near-real particle migration environment.

Landscapes

  • Health & Medical Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Public Health (AREA)
  • Physics & Mathematics (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • General Health & Medical Sciences (AREA)
  • Animal Behavior & Ethology (AREA)
  • Pathology (AREA)
  • Veterinary Medicine (AREA)
  • Medical Informatics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Graphics (AREA)
  • Geometry (AREA)
  • Software Systems (AREA)
  • Primary Health Care (AREA)
  • Epidemiology (AREA)
  • Biophysics (AREA)
  • Pulmonology (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Optics & Photonics (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Radiation-Therapy Devices (AREA)
  • Apparatus For Radiation Diagnosis (AREA)

Abstract

一种基于医学影像的组织元素质量比例解构方法及几何模型建立方法,组织元素质量比例解构方法包括通过已知元素质量比例的组织对应的医学影像数据近似未知医学影像数据对应的组织的元素质量比例的步骤；几何模型建立方法包括：读取医学影像的步骤；根据组织元素质量比例解构方法定义组织种类的步骤；决定组织分群数的步骤；通过医学影像与密度之间的转换关系定义组织密度的步骤；建立带有组织和密度信息的3D编码矩阵的步骤；产生几何模型的步骤。透过已知组织的医学影像数据来解构生物体组织元素质量比例,在制作体素假体时更近似于真实情况,可提升剂量计算准确度精进治疗质量。

Description

基于医学影像的组织元素质量比例解构方法及几何模型建立方法 技术领域
本发明一方面涉及一种组织元素质量比例解构方法，尤其涉及一种基于医学影像的组织元素质量比例解构方法；本发明另一方面涉及一种几何模型建立方法，尤其涉及一种基于医学影像的几何模型建立方法。
背景技术
随着原子科学的发展，例如钴六十、直线加速器、电子射束等放射线治疗已成为癌症治疗的主要手段之一。然而传统光子或电子治疗受到放射线本身物理条件的限制，在杀死肿瘤细胞的同时，也会对射束途径上大量的正常组织造成伤害；另外由于肿瘤细胞对放射线敏感程度的不同，传统放射治疗对于较具抗辐射性的恶性肿瘤(如：多行性胶质母细胞瘤(glioblastoma multiforme)、黑色素细胞瘤(melanoma))的治疗成效往往不佳。
为了减少肿瘤周边正常组织的辐射伤害，化学治疗(chemotherapy)中的标靶治疗概念便被应用于放射线治疗中；而针对高抗辐射性的肿瘤细胞，目前也积极发展具有高相对生物效应(relative biological effectiveness,RBE)的辐射源，如质子治疗、重粒子治疗、中子捕获治疗等。其中，中子捕获治疗便是结合上述两种概念，如硼中子捕获治疗，借由含硼药物在肿瘤细胞的特异性集聚，配合精准的中子射束调控，提供比传统放射线更好的癌症治疗选择。
硼中子捕获治疗(Boron Neutron Capture Therapy,BNCT)是利用含硼(B-10)药物对热中子具有高捕获截面的特性，借由10B(n,α)7Li中子捕获及核分裂反应产生4He和7Li两个重荷电粒子，两粒子的总射程约相当于一个细胞大小，因此对于生物体造成的辐射伤害能局限在细胞层级，当含硼药物选择性地聚集在肿瘤细胞中，搭配适当的中子射源，便能在不对正常组织造成太大伤害的前提下，达到局部杀死肿瘤细胞的目的。
三维模型广泛应用于科学实验分析、科学实验模拟领域。比如在核辐射与防护领域，为了模拟生物体在一定辐射条件下的吸收剂量，常常需要利用计算机技术对医学影像进行各种处理建立精确的蒙特卡罗软件需要的晶格模型，并结合软件进行模拟计算。
蒙特卡罗方法是目前能够对辐照目标内部三维空间核粒子碰撞轨迹和能量分布进行精确模拟的工具，蒙特卡罗方法与复杂的三维人体解剖模型相结合代表了模拟在计算机技术中的跃进。在诊断放射检查中，精确的人体器官剂量评估对于放射治疗是非常有益的。目前，国际上已经成功建立多种人体模型并结合蒙特卡罗模拟程序，对人体在辐射环境下的吸收剂量 进行精确性的计算评估。人体三维解剖模型成功转换为蒙特卡罗程序所需要的几何描述是进行蒙特卡罗模拟计算的前提条件，也是目前国际上蒙特卡罗模拟研究的热点和难点。
核磁共振成像(Magnetic Resonance Imaging,MRI)或电子计算机断层扫描(Computed Tomography,CT)等医学影像能够针对人体体内特征提供较为详细的组织几何结构信息，为人体内部结构的实体建模提供了数据基础。其中，以计算机断层扫描为例，将CT影像数据转换成与人体解剖构造相近的体素假体模型，从而模拟辐射的迁移历程，及与人体组织的作用情形来计算辐射吸收剂量。因此，在给定体素假体的各项元素组成时，尽可能地与病人真实情况吻合，以在模拟计算时获得最接近真实情况的结果。然而目前对体素假体的材料定义，普遍由固定区间范围的CT值对应固定元素组成比例的组织，如Vanderstraeten等人于2007年发表的文献(将在下文详述)所述，CT值＝-100～20对应为脂肪，CT值＝20～100对应为肌肉。但是人体各组织应由特定的元素质量比例组成，而非由单一区间内的固定元素质量比例代表，由此可知目前常见的体素假体成份定义方法仍有相当大的改善空间。
因此，有必要提出一种提高治疗计划精确性的基于医学影像的组织元素质量比例解构方法及包含该解构方法的几何模型建立方法。
发明内容
本发明的一方面提供一种基于医学影像的组织元素质量比例解构方法，包括通过已知元素质量比例的组织对应的医学影像数据近似未知医学影像数据对应的组织的元素质量比例的步骤。该方法透过已知的组织元素质量比例，以医学影像数据求出其他组织的元素质量比例，提供更多组织种类用于建立基于医学影像转换的体素假体，该假体可用于蒙地卡罗治疗计划系统，透过更大量的组织元素质量比例，在蒙特卡罗计算时进而指定正确的物理作用数据如光子的线性衰减系数、中子的作用截面、与荷电粒子(质子、电子或其它荷电粒子)的质量阻挡本领，提供接近于真实的粒子迁移环境。
作为一种优选地，所述医学影像数据与光子线性衰减系数相关，通过所述医学影像数据可以获得已知元素质量比例的组织的有效线性衰减系数。
更进一步，所述通过已知元素质量比例的组织对应的医学影像数据近似未知医学影像数据对应的组织的元素质量比例的步骤包括：
将NIST提供的基本元素的光子线性衰减系数以所述医学影像射束能谱加权，求出所述医学影像射束下的平均线性衰减系数μi；
以所述医学影像的侦检器效率与增建因子乘积作为效率加权因子C，来修正各组织的线性衰减系数使其成为各组织的有效线性衰减系数即
Figure PCTCN2017092745-appb-000001
以相近医学影像数据的组织元素质量比例作为初始假设，以推演算法求出各医学影像数据所对应的生物体内化学元素质量比例分布即mi。
作为一种优选地，所述医学影像数据为CT值，所述CT值HU与有效线性衰减系数μeff的关系为HU＝1000(μeff/μwater-1)，所述CT值以n为一群组，将CT值范围由-100至+1530分为1630/n群组。所述推演算法为最大近似算法。
本发明的另一方面提供一种基于医学影像的几何模型建立方法，包括：
读取医学影像的步骤；
根据医学影像数据与组织种类之间的转换关系或上述组织元素质量比例解构方法定义组织种类的步骤；
决定组织分群数的步骤；
通过医学影像与密度之间的转换关系定义组织密度的步骤；
建立带有组织和密度信息的3D编码矩阵的步骤；
产生几何模型的步骤。
该几何模型建立方法根据医学影像的数据与组织种类之间的转换关系，可以根据实际需要确定组织分群数，从而更加精确地提供组织种类、元素组成及密度，建立的几何模型更加匹配于医学影像反应出的真实情况。
作为一种优选地，该几何模型建立方法应用于中子捕获治疗，其进一步包括给定B-10浓度的步骤和建立带有B-10浓度信息的3D编码矩阵的步骤。标记有B-10浓度信息的几何模型，便可清楚地知道，各个组织内的含硼药物浓度，然后进行中子捕获治疗照射模拟时，则更加真实地反应出实际情况。
组织分群数为用户手动定义的组织分群数加上数据库中已有的4种组织分群数或14种组织分群数或70种组织分群数或所述组织元素质量比例解构方法确定的163种组织分群数。如果在已有数据库中并未建立有相对应的组织分群数，那么可以由用户自定义一个新的组织分群数。这样即避免了如果已有数据库中不能完全匹配相对应的组织分群数，只能近似选择的情况，从而有效地提高的建模的精确度。
更加优选地，该几何模型建立方法进一步包括建立3D组织编码矩阵的步骤和建立3D密度编码矩阵的步骤。根据医学影像的切片通过相对应的转换关系，每一张切片建立相应的组织编码和密度编码，从而建立起3D组织编码矩阵和3D密度编码矩阵。
几何模型包括蒙特卡罗软件输入档所需的晶格卡、栅元卡、曲面卡和材料卡。通过医学 影像最终获得蒙特卡罗软件输入档所需的晶格卡、栅元卡、曲面卡和材料卡，从而为模拟计算提供理论依据并获得精确的模拟计算结果。
医学影像数据可以为核磁共振成像(Magnetic Resonance Imaging,MRI)、电子计算机断层扫描(Computed Tomography,CT)、正电子发射型计算机断层扫描(Positron Emission Tomography,PET)、PET-CT或X射线成像(X-Ray imaging)，下文实施例中将基于电子计算机断层扫描(CT)的数据来阐述，CT的文件格式通常为DICOM。但本领域技术人员熟知地，还可以使用其他的医学影像数据，只要该医学影像数据能够被转换成体素假体组织模型，就能够应用于本发明揭示的基于医学影像数据的组织元素质量比例解构方法及几何模型建立方法中。
本发明实施例的基于医学影像数据的组织元素质量比例解构方法及几何模型建立方法透过已知组织的医学影像数据来解构生物体组织元素质量比例，其可依CT值直接对应生物体组织中的各项元素组成，使在制作体素假体时更近似于真实情况，可提升剂量计算准确度精进治疗质量。
附图说明
图1是硼中子捕获反应示意图。
图2是10B(n,α)7Li中子捕获核反应方程式。
图3是本发明实施例中的基于医学影像的组织元素质量比例解构方法的逻辑框图。
图4是本发明实施例中的基于医学影像的几何模型建立方法的逻辑框图。
图5是CT值(HU)与组织密度回归曲线公式及相关系数图。
具体实施方式
下面结合附图对本发明的实施例做进一步的详细说明，以令本领域技术人员参照说明书文字能够据以实施。
作为一种优选地，以用于中子捕获治疗的基于医学影像的组织元素质量比例解构方法及几何模型建立方法为本发明的实施例。下面将简单介绍一下中子捕获治疗，尤其是硼中子捕获治疗。
中子捕获治疗作为一种有效的治疗癌症的手段近年来的应用逐渐增加，其中以硼中子捕获治疗最为常见，供应硼中子捕获治疗的中子可以由核反应堆或加速器供应。本发明的实施例以加速器硼中子捕获治疗为例，加速器硼中子捕获治疗的基本组件通常包括用于对带电粒子(如质子、氘核等)进行加速的加速器、靶材与热移除系统和射束整形体，其中加速带电粒子与金属靶材作用产生中子，依据所需的中子产率与能量、可提供的加速带电粒子能量与 电流大小、金属靶材的物化性等特性来挑选合适的核反应，常被讨论的核反应有7Li(p,n)7Be及9Be(p,n)9B，这两种反应皆为吸热反应。两种核反应的能量阀值分别为1.881MeV和2.055MeV，由于硼中子捕获治疗的理想中子源为keV能量等级的超热中子，理论上若使用能量仅稍高于阀值的质子轰击金属锂靶材，可产生相对低能的中子，不须太多的缓速处理便可用于临床，然而锂金属(Li)和铍金属(Be)两种靶材与阀值能量的质子作用截面不高，为产生足够大的中子通量，通常选用较高能量的质子来引发核反应。
硼中子捕获治疗(Boron Neutron Capture Therapy,BNCT)是利用含硼(10B)药物对热中子具有高捕获截面的特性，借由10B(n,α)7Li中子捕获及核分裂反应产生4He和7Li两个重荷电粒子。参照图1和图2，其分别示出了硼中子捕获反应的示意图和10B(n,α)7Li中子捕获核反应方程式，两荷电粒子的平均能量约为2.33MeV，具有高线性转移(Linear Energy Transfer,LET)、短射程特征，α粒子的线性能量转移与射程分别为150keV/μm、8μm，而7Li重荷粒子则为175keV/μm、5μm，两粒子的总射程约相当于一个细胞大小，因此对于生物体造成的辐射伤害能局限在细胞层级，当含硼药物选择性地聚集在肿瘤细胞中，搭配适当的中子射源，便能在不对正常组织造成太大伤害的前提下，达到局部杀死肿瘤细胞的目的。
在中子捕获治疗中，为了模拟生物体在一定辐射条件下的吸收剂量以帮助医生制定治疗计划，常常需要利用计算机技术对医学影像进行各种处理建立精确的蒙特卡罗软件需要的晶格模型，并结合蒙特卡罗软件进行模拟计算。本发明实施例的目的在于建立更接近生物体组织构造的基于医学影像的组织元素质量比例解构方法及包含该解构方法的几何模型建立方法，以进行硼中子捕获治疗的组织剂量模拟计算。医学影像数据可以为核磁共振成像(Magnetic Resonance Imaging,MRI)、电子计算机断层扫描(Computed Tomography,CT)、正电子发射型计算机断层扫描(Positron Emission Tomography,PET)、PET-CT或X射线成像(X-Ray imaging)，本实施例中将基于电子计算机断层扫描的数据来阐述，CT的文件格式通常为DICOM。但本领域技术人员熟知地，还可以使用其他的医学影像数据，只要该医学影像数据能够被转换成体素假体模型，就能够应用于本发明揭示的基于医学影像数据的组织元素质量比例解构方法及几何模型建立方法中。
本发明实施例的组织元素质量比例解构方法是基于生物体内化学元素质量比例分布与CT值之间的关系，CT值，又称为Hounsfield Unit(HU或H)，是各组织与水的线性衰减系数比值，为反应光衰减系数之单位，其定义如公式一：
Figure PCTCN2017092745-appb-000002
而各组织的线性衰减系数即为各元素质量比例(mi；0≤mi≤1)与各元素线性衰减系数(μi)之乘积和
Figure PCTCN2017092745-appb-000003
通过美国国家标准与技术研究院(National Institute of Standards and Technology,NIST)提供的基本元素的光子线性衰减系数，将各元素在经过CT产生的X射线的能量范围内的线性衰减系数进行加权求和得到平均线性衰减系数，作为本实施例中进行计算的各元素线性衰减系数(μi)。并将CT扫描机内的侦检器效率与增建因子乘积作为效率加权因子C，来修正各组织的线性衰减系数使其成为各组织的有效线性衰减系数即
Figure PCTCN2017092745-appb-000004
再以W.Schneider等人于2000年发表的文献(W.Schneider,T.Bortfeld,and W.Schlegel,"Correlation between CT numbers and tissue parameters needed for Monte Carlo simulations of clinical dose distributions",Phys.Med.Biol.45,459–478(2000).)为标准，其列出了70种CT值(有效线性衰减系数)对应的软组织或骨组织的各项元素比例分布，如下表一。虽然该文献已列出了具体CT值对应的组织元素质量比例，但可定义的CT值数量仍然偏少，有些数值间隔区间较大，使得该区间内的值难以精确定义。因此，本发明实施例以相近CT值的组织元素质量比例作为初始假设，以推演算法能够求出各CT值所对应的生物体内化学元素质量比例分布即mi，如此则可依CT影像数据获得各体素单元对应的元素组成比例，进而建立三维体素假体模型进行基于蒙特卡罗算法的治疗计划模拟，本实施例中优选采用MCNP软件进行模拟。
表一.CT值对应70种不同元素组成的组织(元素重量百分比)
Figure PCTCN2017092745-appb-000005
Figure PCTCN2017092745-appb-000006
Figure PCTCN2017092745-appb-000007
参阅图3，本发明实施例揭示的基于医学影像的组织元素质量比例解构方法主要包括如下流程：
1.将NIST提供的基本元素的光子线性衰减系数以CT射束能谱加权，求出各元素与液态水在CT射束下的平均线性衰减系数μi和μwater。
2.以W.Schneider等人于2000年发表的文献为标准，能够得到70种软组织与骨组织的CT值及对应的元素质量比例，将已知组织(本实施例中从表一中任选一组织)的各元素质量比例乘上该元素的平均线性衰减系数μi再求和，可得到该组织的理论线性衰减系数μtissue。
3.利用CT值定义HU＝1000(μeff/μwater-1)将上述已知组织的CT值转换成有效线性衰减系数μeff。
4.求出有效与理论线性衰减系数比值C，即C＝μeff/μtissue，作为效率加权因子，其包含计算器断层侦检器效率与计算器断层扫瞄过程中的增建因子。
5.以CT值n为一群组，将CT值范围由-100至+1530分为1630/n群组，本实施例中n为10，则共163个群组，以-100为起点，利用CT值定义HU＝1000(μeff/μwater-1)求出每个区间端点值之有效线性衰减系数μeff，后代入公式C＝μeff/μtissue求出CT值对应的组织理论线性衰减系数μtissue。
6.利用特定CT值邻近的已知组织元素质量比例作为初始假设，以各元素质量比例乘上该元素的平均线性衰减系数μi再求和，并以该CT值对应的组织理论线性衰减系数μtissue作为有效解，以推演算法求出该CT值对应的各元素质量比例，从而完善163群组生物体组织的元素质量比例。本实施例中推演算法优选最大近似算法。
简单来说，本发明实施例揭示的基于医学影像数据的几何模型建立方法主要包括如下流程：
1、输入计算机断层影像(DICOM格式)，CT影像将呈现于对应接口上；
2、自动读取DICOM档案中已定义的ROI边界，亦可另外新增ROI；
3、软件将对CT影像上每一体素点(voxel)的位置进行判别，将所有体素点分为「落于ROI边界内」及「落于ROI边界外」；
4、针对ROI边界内的体素点，可根据实际需求，由使用者手动对每一ROI定义相对的唯一组织种类及组织密度，或自动进行CT值与组织种类、密度的匹配，以避免因给定CT值涵盖范围较大的ROI(如黏膜腔室)唯一组织种类(元素组成)及密度，而造成剂量计算的误差；
5、而针对ROI边界外的体素点，则自动进行组织种类的定义，依据CT值大小差异，可区分为4或14或70或163种不同元素组成的组织，用户可根据实际判断，选择使用ICRU-46报告表列的4种组织(将在下文详述)，或使用Vanderstraeten等人于2007年发表文献中的14种不同元素组成的组织(将在下文详述)，或使用W.Schneider等人于2000年发表的文献中的70种不同元素组成的组织，或使用上述本发明揭示的组织元素质量比例解构方法获得的163种组织；
6、对于未手动定义密度的体素点，依据CT值大小差异，将自动给定密度，共可区分出96个密度分群；
7、使用者手动输入正常血液含硼药物浓度、肿瘤-血液硼浓度比值、组织-血液硼浓度比值等参数，将编写B-10元素于所有体素点中；
8、软件将整合组织种类(元素组成)、组织密度、组织B-10浓度等信息，产生三维MCNP晶格模型，并编写出MCNP软件输入档规定格式的晶格卡(lattice card)、栅元卡(cell card)、曲面卡(surface card)及材料卡(material card)。
具体来说，请参照图4，本发明揭示的基于医学影像数据的几何模型建立方法包括两个 实施例。
本发明的第一实施例提供一种基于医学影像数据的几何模型建立方法，包括：读取医学影像数据的步骤；通过医学影像数据与组织种类之间的转换关系定义组织种类的步骤；决定组织分群数的步骤；通过医学影像数据与密度之间的转换关系定义组织密度的步骤；建立带有组织和密度信息的3D编码矩阵的步骤；产生几何模型的步骤。
根据医学影像的数据与组织种类之间的转换关系，可以根据实际需要确定组织分群数，从而更加精确地提供组织种类、元素组成及密度，建立的几何模型更加匹配于医学影像数据反应出的真实情况。
几何模型建立方法进一步包括给定B-10浓度的步骤和建立带有B-10浓度信息的3D编码矩阵的步骤。标记有B-10浓度信息的几何模型，便可清楚地知道，各个体素点内的含硼药物浓度，然后进行中子捕获治疗照射模拟时，则更加真实地反应出实际情况。
组织分群数为用户手动定义的组织分群数加上数据库中已有的4种组织分群数或14种组织分群数或70种组织分群数或本发明的组织元素质量比例解构方法确定的163种组织分群数。如果在已有数据库中并未建立有相对应的组织分群数，那么可以根据实验情况确定一个新的组织分群数。这样即避免了如果已有数据库中不能完全匹配相对应的组织分群数，只能近似选择的情况，从而有效地提高的建模的精确度。
几何模型建立方法进一步包括建立3D组织编码矩阵的步骤和建立3D密度编码矩阵的步骤。根据医学影像数据的切片通过相对应的转换关系，每一张切片建立相应的组织编码和密度编码，从而建立起3D组织编码矩阵和3D密度编码矩阵。
几何模型包括MCNP软件输入档所需的晶格卡、栅元卡、曲面卡和材料卡。通过医学影像数据最终获得蒙特卡罗软件输入档所需的晶格卡、栅元卡、曲面卡和材料卡，从而为模拟计算提供理论依据并获得精确的模拟计算结果。
本发明的第二实施例提供一种基于医学影像数据的几何模型建立方法，包括：读取医学影像数据的步骤；定义或读取ROI边界的步骤；判断医学影像体素是否在ROI边界内的步骤：如果是，则进入通过为每个ROI边界内体素给定一个特定组织与密度的方式进行用户手动定义组织种类与密度的步骤或或进入通过医学影像数据与组织种类/密度之间的转换关系自动定义ROI组织种类与密度的步骤，如果否，则进入通过医学影像数据与组织种类之间的转换关系自动定义组织种类的步骤并通过医学影像数据与密度之间的转换关系定义组织密度的步骤；建立带有组织和密度信息的3D编码矩阵的步骤；产生几何模型的步骤。
所谓ROI是指感兴趣区域，用户可以手动定义ROI的组织种类、元素组成以及密度。如 果不在ROI边界内，根据医学影像的数据与组织种类/密度之间的转换关系进行组织种类的定义，并根据实际需要确定组织分群数，从而更加精确地提供组织种类、元素组成及密度，建立的几何模型更加匹配于医学影像数据反应出的真实情况。
几何模型建立方法包括给定B-10浓度的步骤和建立带有B-10浓度信息的3D编码矩阵的步骤。标记有B-10浓度信息的几何模型，便可清楚地知道，各个体素点内的含硼药物浓度，然后进行中子捕获治疗照射模拟时，则更加真实地反应出实际情况。
组织分群数为用户手动定义的组织分群数加上数据库中已有的4种组织分群数或14种组织分群数或70种组织分群数，或使用上述本发明揭示的组织元素质量比例解构方法获得的163种组织分群数。如果单一ROI边界内涵盖CT值范围较大(如黏膜腔室)，则可自动进行CT值与组织种类、密度的匹配，以避免因给定唯一组织种类(元素组成)及密度，而造成剂量计算的误差。如果在已有数据库中并未建立有相对应的组织分群数，那么可以根据实验情况确定一个新的组织分群数。这样即避免了如果已有数据库中不能完全匹配相对应的组织分群数，只能近似选择的情况，从而有效地提高的建模的精确度。
几何模型建立方法进一步包括建立3D组织编码矩阵的步骤和建立3D密度编码矩阵的步骤。根据医学影像数据的切片通过相对应的转换关系，每一张切片建立相应的组织编码和密度编码，从而建立起3D组织编码矩阵和3D密度编码矩阵。
几何模型包括MCNP软件输入档所需的晶格卡、栅元卡、曲面卡和材料卡。通过医学影像数据最终获得蒙特卡罗软件输入档所需的晶格卡、栅元卡、曲面卡和材料卡，从而为模拟计算提供理论依据并获得精确的模拟计算结果。
请参照图5，下面将阐述本发明揭示的基于医学影像数据的几何模型建立方法中已有数据库中的CT值-组织种类及CT值-组织密度转换关系图表。
引用Vanderstraeten等人于2007年发表之文献(Barbara Vanderstraeten et al,“Convension of CT numbers into tissue parametersfor Monte Carlo dose calculations:a multi-centre study”,Phys.Med.Biol.52(2007)539–562.)，依据CT值大小不同，可转换出1种空气(air)、1种肺组织(lung)、2种软组织(脂肪(adipose)及肌肉软组织(muscle))、10种骨组织(bone)，换言之，不同CT值共对应出14种不同元素组成的组织，如下表二。
表二.CT值对应14种不同元素组成的组织(元素重量百分比)
Figure PCTCN2017092745-appb-000008
Figure PCTCN2017092745-appb-000009
引用ICRU-46号报告(International Commission on Radiation Units and Measurements,Photon,electron,proton and neutron interaction data for body tissues,ICRU-46,Tech.Rep.,1992.)，撷取4种人体脑部主要组织，包含空气(air)、脑组织(adult brain)、皮肤(adult skin)、头骨(cranium)，其对应之密度与元素组成如表三。
表三.CT值对应4种不同元素组成的组织(元素重量百分比)
Figure PCTCN2017092745-appb-000010
同样引用Vanderstraeten等人的文献，该份文献统整了医院真实的实验值，整理出CT值对应组织密度的关系公式，如图5所示；本发明揭示的基于医学影像数据的几何模型建立方法使用图5的三组回归公式，将CT值(-1000～2000)区分为96组密度分群，如表四。
表四.CT值与质量密度的转换.HUL和HUU分别为CT值的下限和上限
Figure PCTCN2017092745-appb-000011
Figure PCTCN2017092745-appb-000012
本方法透过已知的组织元素质量比例，以计算机断层影像数据求出其他组织的元素质量比例，使在制作体素假体时能考虑更多不同的CT数据，提供更多组织种类，使假体更加贴近真实情况。该假体可用于基于蒙特卡罗算法的治疗计划系统，透过更大量的组织元素质量比例，在蒙特卡罗计算时进而指定正确的物理作用数据如光子的线性衰减系数、中子的作用截面、与荷电粒子(质子、电子或其它荷电粒子)的质量阻挡本领，提供接近于真实的粒子迁移环境。
尽管上面对本发明说明性的具体实施方式进行了描述，以便于本技术领域的技术人员理解本发明，但应该清楚，本发明不限于具体实施方式的范围，对本技术领域的普通技术人员来讲，只要各种变化在所附的权利要求限定和确定的本发明的精神和范围内，这些变化是显而易见的，都在本发明要求保护的范围之内。

Claims (10)

  1. 一种基于医学影像的组织元素质量比例解构方法，包括通过已知元素质量比例的组织对应的医学影像数据近似未知医学影像数据对应的组织的元素质量比例的步骤。
  2. 根据权利要求1所述的基于医学影像的组织元素质量比例解构方法，其特征在于：所述医学影像数据与光子线性衰减系数相关，通过所述医学影像数据可以获得已知元素质量比例的组织的有效线性衰减系数。
  3. 根据权利要求2所述的基于医学影像的组织元素质量比例解构方法，其特征在于：所述通过已知元素质量比例的组织对应的医学影像数据近似未知医学影像数据对应的组织的元素质量比例的步骤包括：
    将NIST提供的基本元素的光子线性衰减系数以所述医学影像射束能谱加权，求出所述医学影像射束下的平均线性衰减系数μi；
    以所述医学影像的侦检器效率与增建因子乘积作为效率加权因子C，来修正各组织的线性衰减系数使其成为各组织的有效线性衰减系数即
    Figure PCTCN2017092745-appb-100001
    以相近医学影像数据的组织元素质量比例作为初始假设，以推演算法求出各医学影像数据所对应的生物体内化学元素质量比例分布即mi。
  4. 根据权利要求3所述的基于医学影像的组织元素质量比例解构方法，其特征在于：所述医学影像数据为CT值，所述CT值HU与有效线性衰减系数μeff的关系为HU＝1000(μeff/μwater-1)，所述CT值以n为一群组，将CT值范围由-100至+1530分为1630/n群组。
  5. 根据权利要求3所述的基于医学影像的组织元素质量比例解构方法，其特征在于：所述推演算法为最大近似算法。
  6. 一种基于医学影像的几何模型建立方法，包括：
    读取医学影像数据的步骤；
    根据医学影像数据与组织种类之间的转换关系或上述权利要求之一所述的组织元素质量比例解构方法定义组织种类的步骤；
    决定组织分群数的步骤；
    通过医学影像数据与密度之间的转换关系定义组织密度的步骤；
    建立带有组织和密度信息的3D编码矩阵的步骤；
    产生几何模型的步骤。
  7. 根据权利要求6所述的基于医学影像的几何模型建立方法，其特征在于：所述几何模型建 立方法应用于中子捕获治疗，所述几何模型建立方法包括给定B-10浓度的步骤和建立带有B-10浓度信息的3D编码矩阵的步骤。
  8. 根据权利要求6所述的基于医学影像的几何模型建立方法，其特征在于：所述组织分群数为用户手动定义的组织分群数加上数据库中已有的4种组织分群数或14种组织分群数或70种组织分群数或所述组织元素质量比例解构方法确定的163种组织分群数。
  9. 根据权利要求8所述的基于医学影像的几何模型建立方法，其特征在于：所述几何模型建立方法进一步包括建立3D组织编码矩阵的步骤和建立3D密度编码矩阵的步骤。
  10. 根据权利要求6所述的基于医学影像的几何模型建立方法，其特征在于：所述几何模型包括蒙特卡罗软件输入档所需的晶格卡、栅元卡、曲面卡和材料卡。
PCT/CN2017/092745 2017-01-11 2017-07-13 基于医学影像的组织元素质量比例解构方法及几何模型建立方法 Ceased WO2018129889A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
EP17891253.1A EP3556431A4 (en) 2017-01-11 2017-07-13 MEDICAL IMAGE-BASED METHOD FOR DECONSTRUCTING MASS RATIO OF TISSUE ELEMENTS AND METHOD FOR ESTABLISHING A GEOMETRIC MODEL
JP2019537223A JP6938646B2 (ja) 2017-01-11 2017-07-13 医用画像に基づく組織元素質量比率の脱構築方法及び幾何的モデルの構築方法
RU2019122188A RU2736917C1 (ru) 2017-01-11 2017-07-13 Способ анализа элементов и отношений масс элементов ткани и способ построения геометрической модели на основе медицинского изображения
US16/439,980 US11590360B2 (en) 2017-01-11 2019-06-13 Method for analyzing elements and mass ratios of elements of tissue and method for establishing geometric model based on medical image

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710017476.X 2017-01-11
CN201710017476.XA CN108295384B (zh) 2017-01-11 2017-01-11 基于医学影像的组织元素质量比例解构方法及几何模型建立方法

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US16/439,980 Continuation US11590360B2 (en) 2017-01-11 2019-06-13 Method for analyzing elements and mass ratios of elements of tissue and method for establishing geometric model based on medical image

Publications (1)

Publication Number Publication Date
WO2018129889A1 true WO2018129889A1 (zh) 2018-07-19

Family

ID=62839195

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/092745 Ceased WO2018129889A1 (zh) 2017-01-11 2017-07-13 基于医学影像的组织元素质量比例解构方法及几何模型建立方法

Country Status (7)

Country Link
US (1) US11590360B2 (zh)
EP (1) EP3556431A4 (zh)
JP (1) JP6938646B2 (zh)
CN (1) CN108295384B (zh)
RU (1) RU2736917C1 (zh)
TW (1) TWI646945B (zh)
WO (1) WO2018129889A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI854879B (zh) * 2022-11-07 2024-09-01 大陸商中硼（廈門）醫療器械有限公司 中子捕獲治療系統及劑量評估方法
CN119693300A (zh) * 2024-11-13 2025-03-25 中国科学院遗传与发育生物学研究所 一种无损的生物样品组分分布的获取方法

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI695387B (zh) * 2019-01-25 2020-06-01 上銀科技股份有限公司 將電腦斷層掃描儀的掃描資訊轉換為骨骼參數的方法
US10846855B2 (en) 2019-04-24 2020-11-24 Hiwin Technologies Corp. Method for converting scan information of a computed tomography scanner into bone parameters
CN111128317A (zh) * 2019-11-20 2020-05-08 中国辐射防护研究院 一种电离辐射组织等效材料配方设计方法及系统
WO2021164026A1 (zh) * 2020-02-21 2021-08-26 西门子（中国）有限公司 模体构建方法和装置以及体模制造方法
WO2021170155A1 (en) * 2020-02-26 2021-09-02 Proton Therapy Group S.R.O. A method of binary proton radiotherapy
CN113797447A (zh) * 2020-06-11 2021-12-17 中硼(厦门)医疗器械有限公司 放射治疗系统及其治疗计划生成方法
CN114367061B (zh) * 2020-10-14 2023-06-23 中硼(厦门)医疗器械有限公司 硼中子捕获治疗系统及其治疗计划生成方法
WO2022143333A1 (zh) * 2020-12-31 2022-07-07 中硼(厦门)医疗器械有限公司 放射线照射系统及其控制方法
CN114152635B (zh) * 2021-10-15 2024-05-31 中国人民解放军军事科学院军事医学研究院 中子外照射后人体血管内中子能谱的等效模拟装置
CN118037667B (zh) * 2024-02-21 2024-08-27 国科离子医疗科技有限公司 物质元素密度分布的确定方法、装置、电子设备和存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070282575A1 (en) * 2006-06-05 2007-12-06 Cambridge Research & Instrumentation, Inc. Monte Carlo simulation using GPU units on personal computers
CN101751697A (zh) * 2010-01-21 2010-06-23 西北工业大学 一种基于统计模型的三维场景重建方法
CN104267425A (zh) * 2014-10-16 2015-01-07 中国科学院合肥物质科学研究院 一种基于CT数据的内照射HPGe探测器探测效率确定方法
CN104537699A (zh) * 2014-12-22 2015-04-22 上海联影医疗科技有限公司 获取ct图像的方法
CN106474634A (zh) * 2015-11-17 2017-03-08 南京中硼联康医疗科技有限公司 基于医学影像数据的几何模型建立方法

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1658878A1 (en) * 2004-11-17 2006-05-24 The European Community, represented by the European Commission BNCT treatment planning
JP5339592B2 (ja) * 2009-01-28 2013-11-13 独立行政法人国立がん研究センター 陽子線治療におけるポジトロン放出核種のアクティビティ分布のシミュレーション方法
JP5641503B2 (ja) * 2010-10-15 2014-12-17 国立大学法人 筑波大学 マルチステップ・ラティス・ボクセル法
US20130231548A1 (en) * 2011-08-30 2013-09-05 Qi Imaging, Llc Deformable registration for tissue analytics
US10292670B2 (en) * 2013-12-04 2019-05-21 Elekta Ltd. Method and system for dose calculation based on continuous material indexing
WO2016038604A1 (en) * 2014-09-11 2016-03-17 B. G. Negev Technologies And Applications Ltd. (Ben-Gurion University) Interactive segmentation
JP6164662B2 (ja) * 2015-11-18 2017-07-19 みずほ情報総研株式会社 治療支援システム、治療支援システムの動作方法及び治療支援プログラム
TWI594732B (zh) * 2015-12-28 2017-08-11 Nat Chung-Shan Inst Of Science And Tech Three-dimensional median filter applied to computed tomography
TWI593392B (zh) * 2015-12-28 2017-08-01 Metal Detection and Artifact Removal Methods in Computerized Tomography Images

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070282575A1 (en) * 2006-06-05 2007-12-06 Cambridge Research & Instrumentation, Inc. Monte Carlo simulation using GPU units on personal computers
CN101751697A (zh) * 2010-01-21 2010-06-23 西北工业大学 一种基于统计模型的三维场景重建方法
CN104267425A (zh) * 2014-10-16 2015-01-07 中国科学院合肥物质科学研究院 一种基于CT数据的内照射HPGe探测器探测效率确定方法
CN104537699A (zh) * 2014-12-22 2015-04-22 上海联影医疗科技有限公司 获取ct图像的方法
CN106474634A (zh) * 2015-11-17 2017-03-08 南京中硼联康医疗科技有限公司 基于医学影像数据的几何模型建立方法

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
"International Commission on Radiation Units and Measurements, Photon, electron, proton and neutron interaction data for body tissues, ICRU-46", TECH. REP., 1992
BARBARA VANDERSTRAETEN ET AL.: "Conversion of CT numbers into tissue parameters for Monte Carlo dose calculations: a multi-center study", PHYS. MED. BIOL., vol. 52, 2007, pages 539 - 562, XP020113168, DOI: doi:10.1088/0031-9155/52/3/001
See also references of EP3556431A4
W. SCHNEIDERT. BORTFELDW. SCHLEGEL: "Correlation between CT numbers and tissue parameters needed for Monte Carlo simulations of clinical dose distributions", PHYS. MED. BIOL., vol. 45, 2000, pages 459 - 478

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI854879B (zh) * 2022-11-07 2024-09-01 大陸商中硼（廈門）醫療器械有限公司 中子捕獲治療系統及劑量評估方法
CN119693300A (zh) * 2024-11-13 2025-03-25 中国科学院遗传与发育生物学研究所 一种无损的生物样品组分分布的获取方法

Also Published As

Publication number Publication date
JP6938646B2 (ja) 2021-09-22
JP2020503959A (ja) 2020-02-06
EP3556431A1 (en) 2019-10-23
CN108295384B (zh) 2020-02-28
TW201825050A (zh) 2018-07-16
CN108295384A (zh) 2018-07-20
TWI646945B (zh) 2019-01-11
US20190329067A1 (en) 2019-10-31
US11590360B2 (en) 2023-02-28
EP3556431A4 (en) 2020-01-15
RU2736917C1 (ru) 2020-11-23

Similar Documents

Publication Publication Date Title
TWI639134B (zh) Method for establishing geometric model based on medical image
TWI646945B (zh) 基於醫學影像的組織元素質量比例解構方法及幾何模型建立方法
TWI647657B (zh) Method for establishing smooth geometric model based on medical image data
Papagiannis et al. Current state of the art brachytherapy treatment planning dosimetry algorithms
WO2022078175A1 (zh) 硼中子捕获治疗系统及其治疗计划生成方法
CN104258505B (zh) 肿瘤放射治疗剂量个体化验证仿真体模及其建立和应用
WO2017097035A1 (zh) 射束的照射角度评价方法
CN113877073B (zh) 放射治疗系统及其治疗计划生成方法
CN109308733A (zh) 基于医学影像数据的几何模型建立方法及剂量计算方法
CN117982812A (zh) 硼中子捕获治疗系统及其治疗计划生成方法
CN114566252B (zh) 一种基于生成对抗网络的核医学活度-剂量自动转换方法
Shirazi An Advanced Method to Voxelize a Tissue Using CT Scan Images and Medical Imaging

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: 17891253

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2019537223

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2017891253

Country of ref document: EP

Effective date: 20190716

WWE Wipo information: entry into national phase

Ref document number: 2019122188

Country of ref document: RU