CN117132729A - Multi-mode fine breast model design method, device, equipment and medium - Google Patents

Multi-mode fine breast model design method, device, equipment and medium Download PDF

Info

Publication number
CN117132729A
CN117132729A CN202310913595.9A CN202310913595A CN117132729A CN 117132729 A CN117132729 A CN 117132729A CN 202310913595 A CN202310913595 A CN 202310913595A CN 117132729 A CN117132729 A CN 117132729A
Authority
CN
China
Prior art keywords
breast
gland
fine
model
mammary gland
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.)
Pending
Application number
CN202310913595.9A
Other languages
Chinese (zh)
Inventor
邱睿
武祯
王佳豪
张辉
李君利
刘烨祺
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.)
Tsinghua University
Original Assignee
Tsinghua University
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 Tsinghua University filed Critical Tsinghua University
Priority to CN202310913595.9A priority Critical patent/CN117132729A/en
Publication of CN117132729A publication Critical patent/CN117132729A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
    • G06T17/10—Constructive solid geometry [CSG] using solid primitives, e.g. cylinders, cubes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T15/00—3D [Three Dimensional] image rendering
    • G06T15/08—Volume rendering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
    • G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30068—Mammography; Breast
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Geometry (AREA)
  • Computer Graphics (AREA)
  • Software Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Apparatus For Radiation Diagnosis (AREA)

Abstract

The application relates to a multi-mode fine breast model design method, a device, equipment and a medium, wherein the method comprises the following steps: constructing the shape of the breast to obtain the shape of the target breast; based on the target breast shape, dividing a breast area according to preset breast characteristic parameters to obtain a skin area, a fat area and a fibrous gland area; respectively performing first refined structure growth on the skin region, second refined structure growth on the fat region and third refined structure growth on the fibrous gland region to obtain a fine mammary gland mathematical model; and voxelizing the fine mammary gland mathematical model to obtain a fine mammary gland voxel model, and adjusting the fine mammary gland voxel model based on a preset gland percentage adjustment strategy to obtain the multi-mode fine mammary gland model. Therefore, the problems that the breast metrology model is not fine enough, an image close to clinic cannot be acquired, gland distribution is not considered in the breast imaging model and the like are solved, the imaging texture authenticity of the breast model is ensured, and the clinic accuracy of the dosage is ensured.

Description

Multi-mode fine breast model design method, device, equipment and medium
Technical Field
The application relates to the technical field of medical images, in particular to a multi-mode fine breast model design method, device, equipment and medium.
Background
Breast dosimetry and breast imaging are two important subjects in the study of breast medical diagnosis, and development of three-dimensional imaging technologies such as breast CT (Computed Tomography, electronic computer tomography) and the like has led to deep understanding of breast anatomy.
Homogeneous models in the related art are believed to overestimate 30% of breast dose, and a range of heterogeneous breast models are increasingly proposed; the model comprises a radial Gauss distribution breast model and a three-dimensional bi-Gauss distribution breast model, but the models still have certain defects, the parameter source statistical data amount is small, the model cannot represent female gonad distribution, and a fine anatomical structure in the breast is lacking.
However, the above calculation about breast dosimetry is mainly based on a standard model, lacks a real anatomical structure, cannot acquire a real imaging texture, and focuses on improvement of texture authenticity in the current breast imaging study, and is an important factor affecting the clinical accuracy of the dose for glandular distribution.
In summary, considering the current situation of study on breast dosimetry and relative cleavage of breast imaging, there is currently a lack of a method for designing a relatively fine breast model for radiation protection, and the method needs to be solved.
Disclosure of Invention
The application provides a multi-mode fine breast model design method, device, equipment and medium, which are used for solving the problems that in the related technology, a breast metrology model is not fine enough and can not acquire an image close to clinic, a breast imaging model does not consider glandular distribution, relative rupture exists in the research of the breast metrology model and the like, and the clinic accuracy of dosage is ensured while the imaging texture authenticity of the breast model is ensured.
An embodiment of a first aspect of the present application provides a method for designing a multi-modal fine breast model, including the steps of:
constructing the shape of the breast to obtain the shape of the target breast; based on the target breast shape, dividing a breast area according to preset breast characteristic parameters to obtain a skin area, a fat area and a fibrous gland area; respectively performing first refined structure growth on the skin region, second refined structure growth on the fat region and third refined structure growth on the fibrous gland region to obtain a fine mammary gland mathematical model; and voxelizing the fine mammary gland mathematical model to obtain a fine mammary gland voxel model, and adjusting the fine mammary gland voxel model based on a preset gland percentage adjustment strategy to obtain the multi-mode fine mammary gland model.
Optionally, in some embodiments, the constructing the breast shape results in a target breast shape, comprising: performing shape fitting on the appearance of the breast based on a plurality of preset hypersurface to obtain an initial breast shape; and respectively carrying out upper side shaping deformation, inner side flattening deformation, shoulder connection deformation, breast sagging treatment and front side deflection transformation on the initial breast shape based on a preset deformation strategy to obtain a target breast shape.
Optionally, in some embodiments, the skin region and the fat region are divided by a hypersurface profile; and dividing fat globules obtained by adopting a preset sampling strategy between the fat region and the fibrous gland region.
Optionally, in some embodiments, the performing a first refined structural growth on the skin region, a second refined structural growth on the fat region, and a third refined structural growth on the fibrous gland region, respectively, results in a refined mammary mathematical model, comprising: growing skin, growing nipple and growing areola on the skin area, growing milk infusion sinuses in the nipple, performing Cooper ligament growth on a subcutaneous fat area and a rear fat area of the fat area, sampling fat globule center voxels of the fibrous gland area based on a preset target gland distribution requirement and a target sampling function, and rejecting according to a preset rejection probability to obtain an initial mammary gland model; and processing the initial mammary gland model based on preset Perlin parting noise to obtain the fine mammary gland mathematical model.
Optionally, in some embodiments, the target sampling function to evenly distributed sampling function ratio is in the definition domain bounded by:
wherein M is R which is equal to or more than 0 and equal to or less than R 0 In the time-course of which the first and second contact surfaces,is the distance between any point of the breast section and the center of the section, R 0 For maximum radius of breast section, f (r) is RAN with gland distribution as Gauss distribution, namely target sampling function, RAN (normalized radial adipose fraction, normalized radial fat distribution) represents normalized radial fat distribution, h (r) is RAN with gland distribution as uniform distribution, A and sigma are fitting parameters, VBD is physical quantity for describing gland content in breast, and k is->And sigma (sigma) 2 Is a ratio of (2).
Optionally, in some embodiments, the preset rejection probability is:
wherein r is h Is a random variable sampled from the gland uniform distribution h (R), M is R which is more than or equal to 0 and less than or equal to R 0 In the time-course of which the first and second contact surfaces,is of the upper bound, f (r) h ) As a random variable r h RAN at glandular Gauss distribution f (r), h (r) h ) As a random variable r h RAN with uniform distribution h (R), a and σ are fitting parameters, R 0 Is the maximum radius of the mammary gland section.
Optionally, in some embodiments, the preset Perlin typing noise is:
r≤r base (θ，φ)[1+αP f，l，p，o (θ，φ)]；
wherein r is the distance from any point on the spherical structure surface to the center of the sphere after the Perlin noise is added, and r base (theta, phi) is the distance from any point on the surface of the basic spherical structure to the center of the sphere, theta is the elevation angle of any point on the sphere, phi is the azimuth angle of any point on the sphere, alpha is the amplitude weight factor of the Perlin noise, f is the main frequency weight factor of the Perlin noise, l is the frequency doubling weight factor of the Perlin noise, p is the frequency quadrupling weight factor of the Perlin noise, and o is the eight-time frequency weight factor of the Perlin noise.
An embodiment of the second aspect of the present application provides a multi-modal fine breast model design apparatus, including:
the construction module is used for constructing the shape of the breast to obtain the shape of the target breast; the dividing module is used for dividing the breast area according to the preset breast characteristic parameters based on the target breast shape to obtain a skin area, a fat area and a fibrous gland area; the refining module is used for respectively carrying out first refining structure growth on the skin area, second refining structure growth on the fat area and third refining structure growth on the fibrous gland area to obtain a fine mammary gland mathematical model; and the voxelization module is used for voxelization processing of the fine mammary gland mathematical model to obtain a fine mammary gland voxel model, and adjusting the fine mammary gland voxel model based on a preset gland percentage adjustment strategy to obtain the multi-mode fine mammary gland model.
Optionally, in some embodiments, the building block comprises: the fitting unit is used for performing shape fitting on the appearance of the breast based on a plurality of preset hypersurface to obtain an initial breast shape; and the deformation unit is used for respectively carrying out upper side shaping deformation, inner side flattening deformation, shoulder connection deformation, breast sagging treatment and front side deflection transformation on the initial breast shape based on a preset deformation strategy to obtain a target breast shape.
Optionally, in some embodiments, the partitioning module includes: the skin area and the fat area are divided by adopting a hypersurface contour; and dividing fat globules obtained by adopting a preset sampling strategy between the fat region and the fibrous gland region.
Optionally, in some embodiments, the refinement module includes: the modeling unit is used for growing skin, growing nipple and growing areola on the skin area, growing milk infusion sinuses in the nipple, performing Cooper ligament growth on a subcutaneous fat area and a rear fat area of the fat area, sampling fat globule center voxels of the fibrous gland area based on a preset target gland distribution requirement and a target sampling function, and rejecting according to a preset rejection probability to obtain an initial mammary gland model; and the processing unit is used for processing the initial mammary gland model based on the preset Perlin parting noise to obtain the fine mammary gland mathematical model.
Optionally, in some embodiments, the target sampling function to evenly distributed sampling function ratio is in the definition domain bounded by:
wherein M is R which is equal to or more than 0 and equal to or less than R 0 In the time-course of which the first and second contact surfaces,is the distance between any point of the breast section and the center of the section, R 0 For maximum radius of mammary gland section, f (r) is RAN with gland distribution as Gauss distribution, namely target sampling function, RAN represents normalized radial fat distribution, h (r) is RAN with gland distribution as uniform distribution, A and sigma are fitting parameters, VBD is physical quantity for describing glandular content in mammary gland, and k is->And sigma (sigma) 2 Is a ratio of (2).
Optionally, in some embodiments, the preset rejection probability is:
wherein r is h Is a random variable sampled from the gland uniform distribution h (R), M is R which is more than or equal to 0 and less than or equal to R 0 In the time-course of which the first and second contact surfaces,is of the upper bound, f (r) h ) As a random variable r h RAN at glandular Gauss distribution f (r), h (r) h ) As a random variable r h RAN with uniform distribution h (R), a and σ are fitting parameters, R 0 Is the maximum radius of the mammary gland section.
Optionally, in some embodiments, the preset Perlin typing noise is:
r≤r base (θ，φ)[1+αP f，l，p，o (θ，φ)]；
wherein r is the distance from any point on the spherical structure surface to the center of the sphere after the Perlin noise is added, and r base (theta, phi) is the distance from any point on the surface of the basic spherical structure to the center of the sphere, theta is the elevation angle of any point on the sphere, phi is the azimuth angle of any point on the sphere, alpha is the amplitude weight factor of the Perlin noise, f is the main frequency weight factor of the Perlin noise, l is the frequency doubling weight factor of the Perlin noise, p is the frequency quadrupling weight factor of the Perlin noise, and o is the eight-time frequency weight factor of the Perlin noise.
An embodiment of a third aspect of the present application provides an electronic device, including: the system comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the processor executes the program to realize the multi-mode fine breast model design method according to the embodiment.
An embodiment of a fourth aspect of the present application provides a computer-readable storage medium having stored thereon a computer program for execution by a processor for implementing the multi-modal fine breast model design method as described in the above embodiment.
The method comprises the steps of constructing a breast shape to obtain a target breast shape, dividing a breast region according to preset breast characteristic parameters to obtain a skin region, a fat region and a fiber gland region, carrying out fine structural growth on the skin region, the fat region and the fiber gland region respectively to obtain a fine breast mathematical model, carrying out voxel processing on the model to obtain a fine breast voxel model, and adjusting the fine breast voxel model based on a preset gland percentage adjustment strategy to obtain the multi-mode fine breast model. Therefore, the problems that in the related technology, a breast metrology model is not fine enough, an image close to clinic cannot be acquired, glandular distribution is not considered in the breast imaging model, relative fracture and the like exist in the research of the breast metrology model and the glandular distribution, and the clinic accuracy of dosage is ensured while the imaging texture authenticity of the breast model is ensured are solved.
Additional aspects and advantages of the application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the application.
Drawings
The foregoing and/or additional aspects and advantages of the application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings, in which:
FIG. 1 is a flow chart of a multi-modal fine breast model design method provided in accordance with an embodiment of the present application;
FIG. 2 is a schematic illustration of a fine breast model building process provided in accordance with one embodiment of the present application;
FIG. 3 is a block diagram of a multi-modal fine breast model design apparatus according to an embodiment of the present application;
fig. 4 is a block schematic diagram of an electronic device according to an embodiment of the application.
Detailed Description
Embodiments of the present application are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative and intended to explain the present application and should not be construed as limiting the application.
The following describes a multi-modal fine breast model design method, apparatus, device and medium according to embodiments of the present application with reference to the accompanying drawings. Aiming at the problems that the breast metrology model mentioned in the background art is not fine enough and can not acquire an image close to clinic, the breast imaging model does not consider glandular distribution, and the research of the breast imaging model and the glandular distribution has relative fracture, the application provides a multi-mode fine breast model design method, in which a breast shape is constructed to obtain a target breast shape; based on the target breast shape, dividing a breast area according to preset breast characteristic parameters to obtain a skin area, a fat area and a fibrous gland area; respectively performing first refined structure growth on the skin region, second refined structure growth on the fat region and third refined structure growth on the fibrous gland region to obtain a fine mammary gland mathematical model; and voxelizing the fine mammary gland mathematical model to obtain a fine mammary gland voxel model, and adjusting the fine mammary gland voxel model based on a preset gland percentage adjustment strategy to obtain the multi-mode fine mammary gland model. Therefore, the problems that in the related technology, a breast metrology model is not fine enough, an image close to clinic cannot be acquired, glandular distribution is not considered in the breast imaging model, relative fracture and the like exist in the research of the breast metrology model and the glandular distribution, and the clinic accuracy of dosage is ensured while the imaging texture authenticity of the breast model is ensured are solved.
Prior to describing the embodiments of the present application, the related art related to the above background art will be described in detail.
Digital mammography (Digital Mammography, DM for short) is currently the main medical equipment for breast cancer screening, and has the advantages of high speed, low price, low radiation dose, and the like. With the rapid development of technology, digital mammography (Digital Breast tomosynthesis, DBT for short) has been proposed and developed in the past 20 years. DBT aims to acquire three-dimensional (3D) information of a breast by using a plurality of projections obtained at a limited angle. It can reduce the diagnostic interference caused by overlapping fibroglandular glands in the breast and improve the visibility of lesions. Studies have shown that DBT is superior to DM in diagnosing malignant tumors. Thus, in some cases, DM and DBT will be combined to diagnose pathology.
In recent years, DM and DBT devices have been rapidly developed. Their complexity and diversity present significant challenges for clinical application assessment and design optimization. The practice of clinical trials appears to be somewhat impractical and limited by various factors such as ethics, cost, time requirements or lack of ground truth.
To address these issues, virtual clinical trials (Virtual clinical trials, VCTs for short) provide an alternative method to effectively evaluate medical imaging techniques by simulating patients, imaging systems and interpreters. And VCTs can be performed quickly, economically, efficiently on a computer, and minimize risk to patients in clinical applications. Researchers have used VCTs for a variety of applications including: verification and optimization of a DBT reconstruction method; authenticity assessment of DM projection and DBT reconstruction; glandular dose assessment. Programming procedures have been developed to generate a large collection of models with fine detail and varying sizes, shapes and gland densities. Such as UPENN model, FDA breast model, patr Lei Da breast model and DeBRa breast model. Some studies attempt to produce images comparable to clinical images through current breast models, but they may not yet pass so-called "fool radiologists" visualization tests. VCTs researchers are still striving to improve the realism of breast texture.
In addition, medical radiation is the largest source of artificial ionizing radiation exposure to the public. With the widespread popularity of X-ray scanning, the public is increasingly exposed to medical radiation, leading to an increased risk of radiation, and breast dosimetry has therefore become of increasing interest in recent 20 years. Breast tissue is mainly composed of fibrous glandular tissue, which appears dense on mammography images, and adipose tissue, which appears almost transparent.
In the related art, the main basis for estimating the average glandular dose of the clinical breast is a dose conversion coefficient obtained by Dance and the like based on a homogeneous breast model. X-ray energy spectrum, breast compression thickness and breast gland percentage are considered to be the main factors affecting the average gland dose. In recent years, as the mammary anatomy is further advanced, researchers have found that glandular distribution is also an important factor affecting average glandular doses. The study by Hernandez et al found that homogeneous models overestimated the average glandular dose by 30%, up to 120%, compared to heterogeneous mammary models that consider glandular distribution.
On the other hand, regarding the optimization of radiation protection, the goal is to obtain optimized exposure parameters by balancing the risks and benefits of medical diagnostics. The radiation risk is mainly measured by average gland dosage, and the diagnosis effect is influenced by the image quality and the lesion type. In the related art, the optimized parameter acquisition method is mainly based on a simple model experiment, but the inherent structural noise texture existing in the mammary gland is not considered.
From the above description, it can be seen that the breast dosimetry has only a standard model, no fine structure exists, and no image close to clinic can be obtained. The mammary gland imaging model focuses on texture authenticity, glandular distribution and the like are not considered, and the accuracy of dose calculation is poor.
To sum up, in order to solve the problems in the related art, the application establishes a multi-mode fine breast model generation method aiming at the current research situation of relative fracture of breast dosimetry and breast imaging, and ensures the clinical accuracy of dose calculation while obtaining the texture close to clinical images. The multi-modal fine breast model design method of the present application will be described in detail with reference to specific examples.
Specifically, fig. 1 is a schematic flow chart of a method for designing a multi-modal fine breast model according to an embodiment of the present application.
As shown in FIG. 1, the method for designing the multi-modal fine breast model comprises the following steps:
in step S101, a breast shape is constructed to obtain a target breast shape.
Specifically, the present application first constructs a basic shape of the breast, and then deforms the basic shape, thereby obtaining a target breast shape.
Optionally, in some embodiments, constructing the breast shape to obtain the target breast shape includes: performing shape fitting on the appearance of the breast based on a plurality of preset hypersurface to obtain an initial breast shape; based on a preset deformation strategy, the initial breast shape is subjected to upper side shaping deformation, inner side flattening deformation, shoulder connection deformation, breast sagging treatment and front side deflection transformation respectively, so that the target breast shape is obtained.
In the stage of constructing the basic appearance of the mammary gland, the application adopts a hypersurface to carry out shape fitting on the appearance of the mammary gland, and the hypersurface is added with two parameters E relative to the hypersurface 1 ，∈ 2 The surface curvature change is more flexible, and the hypersurface expression of the ellipsoid is as follows:
wherein, E is 1 ，∈ 2 Is two index parameters in an ellipsoidal equation, and the parameter equation is as follows:
wherein, -pi/2 is less than or equal to phi and less than or equal to pi/2, -pi is less than or equal to theta and less than or equal to pi.
The breast shape of the application is composed of four hypersurface, and the mathematical description is as follows:
wherein r is t ，r b ，r l ，r r Respectively represent four ellipsoidal radii of the shape of the basic breast, r h Representing the height (height) of the nipple from the plane of the pectoral muscle, i.e. the length of the breast.
Further, after the most basic breast shape is constructed, there is a great difference between the shape and the actual breast, i.e. the basic shape needs to be deformed if a breast model conforming to the actual human body is to be constructed. Considering different aspects of the actual breast characteristics, the deformation process of the embodiment of the present application is divided into 5 steps of upper shaping deformation, inner flattening deformation, shoulder connection deformation, breast sagging treatment and front deflection deformation, which will be described in turn.
Regarding the upper shaping deformation, considering that the shape of the upper part of the breast is not a convex curved surface, it is necessary to treat the upper part of the breast to a certain extent so as to be concave to be close to the real human body structure. In the upper treatment, the method mainly adopted by the application is to utilize a polynomial to zoom the z coordinate of the upper half part of the breast to a certain extent, thereby achieving the purpose of plasticity, namely:
x′＝x，y′＝y，z′＝z×f(y)，z≥0； (4)
the polynomial includes four operator-defined variables (S 0 ，T 0 ，S 1 ，T 1 ) At the same time, S is required for the convenience of the operator to perform intuitive processing at the time of performing the operation 0 Equal to the slope when y=0, i.e. S 0 ＝f′(y)| y＝0 At the same time require T 0 Equal to the curvature when y=0, i.e. T 0 ＝f″(y)| y＝0 Corresponding S is 1 ＝f′(y)| y＝ymax And T 1 ＝f″(y)| y＝ymax Further, in order for the deformation not to affect the end points of the breast, f (0) =1 and f (y are required max ) =1. According to the above conditions, we obtain:
f(y′)＝Ay′ 5 +By′ 4 +Cy′ 3 +Dy′ 2 +Ey′+F； (5)
wherein:
y′＝y/y max
A＝-0.5T 0 -3S 0 -3S 1 +0.5T 1
B＝1.5T 0 +8S 0 +7S 1 -0.5T 1
C＝-0.5T 0 -6S 0 -4S 1 +0.5T 1
D＝0.5T 0
E＝S 0
F＝1
regarding medial flattening deformation, the tendency of the medial side of the breast to flatten as it approaches the sternum is known from anatomical knowledge, so the present application requires a degree of polynomial scaling of the medial side of the breast when creating the model, i.e., scaling the right side if it is the left chest and scaling the left side if it is the right chest. The specific coordinate transformation is as follows:
x′＝f(y)x，y′＝y，z′＝z； (6)
To ensure that the model can be consistent with the original model at the root, f (0) =1, f' (0) =1 is required. At the same time, for the convenience of the operator to intuitively adjust, thus requiring f (1) =g 0 ，f′(1)＝g 1 . The following expression is thus obtained:
f(y′)＝Ay′ 3 +By′ 2 +Cy′+D； (7)
wherein:
A＝g 1 +2-2g 0
B＝-g 1 -3+3g 0
C＝0
D＝1
regarding the shoulder connection deformation, since the chest and the shoulder are not simply connected together, there is a first-to-smooth transition so that the chest is directed upward toward the shoulder, and there is also some glandular tissue at this connection position, proper deformation of the shoulder connection of the breast is required. Considering an actual model, the application needs to perform certain translation treatment on the upper part of the model towards the shoulder direction. The corresponding coordinate transformations are:
x′＝x-f(z)，y′＝y，z′＝z z≥0； (8)
the present application takes into account the problem of excessive connection of the upper and lower breast parts, and therefore requires f (0) =0. The specific expression is as follows:
f(z′)＝h 0 z′+h 1 z′ 2 ； (9)
regarding the breast sagging deformation, the breast forms a natural sagging due to gravity in a natural state, and the sagging of the breast is continuously increased with age. Therefore, the application needs to perform certain sagging treatment on the established breast model, so that the breast model is more in line with the breast model in a real natural state. The process equations used here are as follows:
x′＝x，y′＝y，z′＝z-b 0 y-b 1 y 2 ； (10)
With respect to the front deflection deformation, since the actual breast is not directed directly ahead, but is directed to the outside of the human body along with the chest contour, the difference in the breast-directed position causes a change in the incident direction at the time of external irradiation, and further causes a difference in the external irradiation dose of each tissue. It is therefore necessary to translate the established breast model laterally to achieve the deflection effect, expressed as follows:
x′＝x-f(y)，y′＝y，z′＝z； (11)
wherein:
f(y)＝c 0 y+c 1 y 2
the formula can ensure that the translation is 0 at the breast root, namely y=0, so that the range of the breast root is ensured not to be changed.
After the above 5 deflection steps, the breast contour is substantially established. The modeling mode can adjust the breast shape through the parameters which are shallow and easily understood, and can generate the personalized breast model of the specific patient more conveniently so as to perform finer and more accurate dose research aiming at individuals.
In step S102, based on the target breast shape, breast region division is performed according to preset breast feature parameters to obtain a skin region, a fat region and a fibrous gland region.
The preset breast characteristic parameters are characteristic parameters of the breast of the Chinese female, such as subcutaneous fat thickness, skin thickness and the like.
Optionally, in some embodiments, the skin region and the fat region are divided by a hypersurface profile; the fat globules obtained by adopting a preset sampling strategy are divided between the fat region and the fibrous gland region.
Specifically, the skin area and the fat area are divided by adopting the hypersurface contour, and a large number of fat balls which are randomly sampled are adopted as boundaries between the fat area boundary and the fibrous gland area, so that the surface of the fibrous gland area is provided with an uneven texture structure.
In step S103, a fine mammary gland mathematical model is obtained by performing a first fine structure growth on the skin region, a second fine structure growth on the fat region, and a third fine structure growth on the fibrous gland region, respectively.
It is understood that skin areas include skin, nipple and areola, the milk antrum growing inside the nipple, fat areas including subcutaneous fat areas and posterior fat areas, where Cooper ligament growth is performed by the present application. In addition, the fibrous glandular region includes the milk delivery system, end leaflets, glands and fat, the present application is not specifically set forth with respect to its production process, and the following examples specifically describe the improved method of the present application for protecting optimization targets based on this model.
Optionally, in some embodiments, the first refined structural growth of the skin region, the second refined structural growth of the fat region, and the third refined structural growth of the fibrous glandular region, respectively, result in a refined breast mathematical model comprising: growing skin, growing nipple and growing areola on a skin area, growing milk infusion sinuses in the nipple, performing Cooper ligament growth on a subcutaneous fat area and a rear fat area of a fat area, sampling fat globule center voxels of a fibrous gland area based on a preset target gland distribution requirement and a target sampling function, and rejecting according to a preset rejection probability to obtain an initial mammary gland model; and processing the initial mammary gland model based on the preset Perlin parting noise to obtain a fine mammary gland mathematical model.
Specifically, the fat globule sampling method taking glandular distribution into consideration of the application, VBD (Volumetric breast density, mammary gland volume density) is a physical quantity describing glandular content in mammary glands, and the formula is
Wherein V is g And V a Representing the number of voxels of the gland (glandulr) and fat (adicose) in the breast model, respectively. The sagittal section of breast data acquired by breast CT may be approximately a circle. RGF (Radial glandular fraction, radial gland distribution) is defined as the area ratio of the glands in the radial direction of the center of the circle in the range of r to r+dr, i.e
Wherein A is g And A a Representing the number of voxels of gland and fat in a sagittal section of the breast model radially from r to r + dr, respectively. In the related art, regarding analysis of DBCT (dedicated breast CT, based on dedicated breast computed tomography) data, RGF (r) can be fitted radially using Gauss functions. It is assumed that each section of breast contains only two tissues: fat and gland, and VBD all equal, RGF of every tangent plane of mammary gland is:
wherein A, sigma is a fitting parameter, and the maximum radius of the tangent plane is R 0 。
In the same way, RAF (Radial adipose fraction, radial fat distribution is
Order theCan obtain identity
It will be appreciated that if a heterogeneous breast model is generated that contains only fat and gland voxels, directly following a fixed parameter a, σ will result in a difference between the final RGF and the target, this difference being due to the fact that during the sampling process, voxels near the center will be iteratively pumped to a size that will result in a rejection of the partial sampling result, especially when VBD is large, the central region of the actual distribution will flatten as the sampled data increases.
Furthermore, the fibrous glandular region is a very irregular shape, and it is a computationally expensive matter to sample the fat globule center point that corresponds to the final fat distribution within it. In order to meet the distribution requirement of the target gland while improving the sampling efficiency, the application adopts a selecting and sampling method to carry out the fat sphere center.
Specifically, in order to realize sampling from a known distribution density function f (x), the application selects a distribution density function h (x) which is the same as the value range of f (x), if:
the sampling method is selected as follows:
i.e. sample X from h (X) h With f (x) h )/(M·h(x h ) If the f (x) sampling profile is more complex, a simpler sampling profile h (x) can be found and then an additional calculation of f (x) is required for each sample h )/(M·h(x h ))。
RAN (radio access network) homo (RAN with gland distribution being uniform distribution (radial fat distribution after normalization)) as h (x), RAN gauss (RAN with gland distribution of Gauss distribution) is the target sampling function, f (x), then
Wherein M is R which is more than or equal to 0 and less than or equal to R 0 In the time-course of which the first and second contact surfaces,r is the distance from any point of the breast section to the center of the section, R 0 For maximum radius of mammary gland section, f (r) is RAN with gland distribution as Gauss distribution, namely target sampling function, RAN represents normalized radial fat distribution, h (r) is RAN with gland distribution as uniform distribution, A and sigma are fitting parameters, VBD is physical quantity for describing glandular content in mammary gland, and k is->And sigma (sigma) 2 Is a ratio of (2).
The rejection probability is:
wherein r is h Is a random variable sampled from the gland uniform distribution h (R), M is R which is more than or equal to 0 and less than or equal to R 0 In the time-course of which the first and second contact surfaces,is of the upper bound, f (r) h ) Is as followsMachine variable r h RAN at glandular Gauss distribution f (r), h (r) h ) As a random variable r h RAN with uniform distribution h (R), a and σ are fitting parameters, R 0 Is the maximum radius of the mammary gland section.
Therefore, the application samples fat globule center voxels in the mammary gland model, refuses the fat globule center voxels according to the refusing probability, and can obtain a fine mammary gland model which is distributed close to the target gland.
Further, the present application requires the use of a Perlin noise texture to enhance texture realism. The fine structure is a spherical or ellipsoidal structure, and the calculation formula of the basic surface of the spherical structure is as follows:
wherein R is the radius of the spherical structure, theta and phi are the elevation angle and the azimuth angle of any point on the sphere respectively, and R base And (theta, phi) is the distance from any point on the surface of the basic spherical structure to the center of the sphere. In the refining process, in order to make the internal structure of the breast more real, the application aims to add Perlin parting noises with different frequencies on the surface of the refined spherical structure, and the method specifically comprises the following steps:
r≤r base (θ，φ)[1+αP f，l，p，o (θ，φ)]； (21)
wherein r is the distance from any point on the spherical structure surface to the center of the sphere after the Perlin noise is added, and r base (theta, phi) is the distance from any point on the surface of the basic spherical structure to the center of the sphere, theta is the elevation angle of any point on the sphere, phi is the azimuth angle of any point on the sphere, alpha is the amplitude weight factor of the Perlin noise, f is the main frequency weight factor of the Perlin noise, l is the frequency doubling weight factor of the Perlin noise, p is the frequency quadrupling weight factor of the Perlin noise, and o is the eight-time frequency weight factor of the Perlin noise.
In step S104, the fine breast mathematical model is subjected to voxel processing to obtain a fine breast voxel model, and the fine breast voxel model is adjusted based on a preset gland percentage adjustment strategy to obtain a multi-modal fine breast model.
The mathematical model voxelization needs to consider the problems of voxelization algorithm optimization, voxelization precision and efficiency balance and the like, and particularly, a large number of matrixes need to be crossed and operated in the processes of catheter growth, fat globule sampling and the like.
Based on the above embodiments, the present application establishes a normal fine mammary gland mathematical model. Therefore, the application adopts a fine structure positioning algorithm to determine the voxelization range, and calculates the coordinate corresponding to each voxel center point in the range.
Further, the application judges whether each coordinate belongs to the inside of the tissue according to a mammary gland mathematical modeling equation, marks the coordinate as a corresponding tissue number, finally carries out voxelization operation according to the sequence from the outer layer to the inner layer, and replaces the outer layer tissue structure number with the inner layer tissue structure number.
Therefore, the multi-mode fine breast voxel model considering gland distribution is established based on the breast medical anatomical structure through self-defining of the appearance parameters and the anatomical parameters. Based on the above embodiment, it can be understood that the multi-mode fine breast model design method of the application makes up for the technical blank that gland distribution is not considered in foreign Virtual Clinical Test Systems (VCTs) in the related technology, and realizes the clinical accuracy of model dose calculation on the basis of guaranteeing real anatomical texture. In addition, the application combines the dual outstanding performance of the fine mammary gland model in the aspects of dosimetry and imaging, and can be used for researching radiation protection optimization technology.
In order for those skilled in the art to further understand the multi-modal fine breast model design method of the present application, the following illustrative examples schematically illustrate the steps of carrying out the method.
Specifically, fig. 2 is a schematic diagram of a fine breast model construction process according to an embodiment of the present application, and as shown in fig. 2, the design of the multi-modal fine breast model includes: the method comprises four stages of building breast appearance, breast area division, fine structure growth and model voxel.
In the mammary gland appearance constructing stage, the application constructs a basic appearance based on typical parameters of female mammary gland appearance, and obtains an initial mammary gland appearance through 5 times of deformation including upper side plasticity, inner side flattening, shoulder connection, mammary gland sagging and front side deflection.
In the mammary gland region dividing stage, the application divides the model into a skin region, a fibrous gland region and a fat region according to region parameters based on the generated initial mammary gland appearance.
In the growth stage of the refined structure, the application carries out further refined structure growth on the divided mammary gland region, generates a catheter system, a cooper ligament, fat globules and the like, and establishes a fine mammary gland mathematical model; adding Perlin noise to the surface of the sphere in it serves to enhance the texture realism of the internal anatomy.
In the voxelization stage, the application realizes voxelization based on a fine mammary gland mathematical model through a fine structure positioning algorithm, and establishes a fine mammary gland voxel model.
Therefore, the multi-mode fine breast model for radiation protection optimization is constructed through four stages of breast appearance, breast region division, fine structure growth and model voxel, and the model ensures the clinical accuracy of dosage while ensuring the imaging texture authenticity of the breast model.
According to the multi-mode fine breast model design method provided by the embodiment of the application, the target breast shape is obtained by constructing the breast shape, the skin area, the fat area and the fiber gland area are obtained by dividing the breast area according to the preset breast characteristic parameters, the fine structure growth is respectively carried out on the skin area, the fat area and the fiber gland area, the fine breast mathematical model is obtained, the fine breast voxel model is obtained by carrying out voxel processing on the model, and the fine breast voxel model is regulated based on the preset gland percentage regulation strategy, so that the multi-mode fine breast model is obtained. Therefore, the problems that in the related technology, a breast metrology model is not fine enough, an image close to clinic cannot be acquired, glandular distribution is not considered in the breast imaging model, relative fracture and the like exist in the research of the breast metrology model and the glandular distribution, and the clinic accuracy of dosage is ensured while the imaging texture authenticity of the breast model is ensured are solved.
The multi-modal fine breast model design apparatus according to the embodiment of the present application will be described next with reference to the accompanying drawings.
FIG. 3 is a block diagram of a multi-modal fine breast model design apparatus according to an embodiment of the present application.
As shown in fig. 3, the multi-modal fine breast model design apparatus 10 includes: a construction module 100, a partitioning module 200, a refinement module 300, and a voxelization module 400.
Wherein, the construction module 100 is configured to construct a breast shape to obtain a target breast shape; the dividing module 200 is configured to divide a breast area according to preset breast feature parameters based on a target breast shape to obtain a skin area, a fat area and a fibrous gland area; a refinement module 300 for respectively performing a first refinement structure growth on the skin region, a second refinement structure growth on the fat region, and a third refinement structure growth on the fibrous gland region to obtain a refined mammary gland mathematical model; the voxelization module 400 is configured to voxeize the fine mammary gland mathematical model to obtain a fine mammary gland voxel model, and adjust the fine mammary gland voxel model based on a preset gland percentage adjustment strategy to obtain a multi-mode fine mammary gland model.
Optionally, in some embodiments, building module 100 includes: fitting unit and deformation unit.
The fitting unit is used for performing shape fitting on the appearance of the breast based on a plurality of preset hypersurface to obtain an initial breast shape; and the deformation unit is used for respectively carrying out upper side shaping deformation, inner side flattening deformation, shoulder connection deformation, breast sagging treatment and front side deflection transformation on the initial breast shape based on a preset deformation strategy to obtain the target breast shape.
Optionally, in some embodiments, the partitioning module 200 includes: the skin area and the fat area are divided by adopting a hypersurface contour; the fat globules obtained by adopting a preset sampling strategy are divided between the fat region and the fibrous gland region.
Optionally, in some embodiments, the refinement module 300 includes: a modeling unit and a processing unit.
The modeling unit is used for growing skin, growing nipple and growing areola on a skin area, growing milk antrum in the nipple, performing Cooper ligament growth on a subcutaneous fat area and a rear fat area of the fat area, sampling fat globule center voxels of a fiber gland area based on a preset target gland distribution requirement and a target sampling function, and rejecting according to a preset rejection probability to obtain an initial mammary gland model; and the processing unit is used for processing the initial mammary gland model based on the preset Perlin parting noise to obtain a fine mammary gland mathematical model.
Optionally, in some embodiments, the target sampling function to evenly distributed sampling function ratio is in the definition domain bounded by:
wherein M is R which is equal to or more than 0 and equal to or less than R 0 In the time-course of which the first and second contact surfaces,r is the distance from any point of the breast section to the center of the section, R 0 For maximum radius of mammary gland section, f (r) is RAN with gland distribution as Gauss distribution, namely target sampling function, RAN represents normalized radial fat distribution, h (r) is RAN with gland distribution as uniform distribution, A and sigma are fitting parameters, VBD is physical quantity for describing glandular content in mammary gland, and k is->And sigma (sigma) 2 Is a ratio of (2).
Optionally, in some embodiments, the preset rejection probability is:
wherein r is h Is a random variable sampled from the gland uniform distribution h (R), M is R which is more than or equal to 0 and less than or equal to R 0 In the time-course of which the first and second contact surfaces,is of the upper bound, f (r) h ) As a random variable r h RAN at glandular Gauss distribution f (r), h (r) h ) As a random variable r h RAN with uniform distribution h (R), a and σ are fitting parameters, R 0 Is the maximum radius of the mammary gland section.
Optionally, in some embodiments, the preset Perlin typing noise is:
r≤r base (θ，φ)[1+αP f，l，p，o (θ，φ)]；
wherein r is the distance from any point on the spherical structure surface to the center of the sphere after the Perlin noise is added, and r base (theta, phi) is the distance from any point on the surface of the basic spherical structure to the center of the sphere, theta is the elevation angle of any point on the sphere, phi is the azimuth angle of any point on the sphere, alpha is the amplitude weight factor of the Perlin noise, f is the main frequency weight factor of the Perlin noise, l is the frequency doubling weight factor of the Perlin noise, p is the frequency quadrupling weight factor of the Perlin noise, and o is the eight-time frequency weight factor of the Perlin noise.
It should be noted that the foregoing explanation of the embodiment of the method for designing a multi-modal fine breast model is also applicable to the multi-modal fine breast model designing apparatus of this embodiment, and will not be repeated here.
According to the multi-mode fine breast model design device provided by the embodiment of the application, the target breast shape is obtained by constructing the breast shape, the skin area, the fat area and the fiber gland area are obtained by dividing the breast area according to the preset breast characteristic parameters, the fine structure growth is respectively carried out on the skin area, the fat area and the fiber gland area, the fine breast mathematical model is obtained, the fine breast voxel model is obtained by carrying out voxel processing on the model, and the fine breast voxel model is regulated based on the preset gland percentage regulation strategy, so that the multi-mode fine breast model is obtained. Therefore, the problems that in the related technology, a breast metrology model is not fine enough, an image close to clinic cannot be acquired, glandular distribution is not considered in the breast imaging model, relative fracture and the like exist in the research of the breast metrology model and the glandular distribution, and the clinic accuracy of dosage is ensured while the imaging texture authenticity of the breast model is ensured are solved.
Fig. 4 is a schematic structural diagram of an electronic device according to an embodiment of the present application. The electronic device may include:
Memory 401, processor 402, and a computer program stored on memory 401 and executable on processor 402.
The processor 402, when executing the program, implements the multi-modal fine breast model design method provided in the above embodiments.
Further, the electronic device further includes:
a communication interface 403 for communication between the memory 401 and the processor 402.
A memory 401 for storing a computer program executable on the processor 402.
The memory 401 may include high speed RAM (Random Access Memory ) memory, and may also include non-volatile memory, such as at least one disk memory.
If the memory 401, the processor 402, and the communication interface 403 are implemented independently, the communication interface 403, the memory 401, and the processor 402 may be connected to each other by a bus and perform communication with each other. The bus may be an ISA (Industry Standard Architecture ) bus, a PCI (Peripheral Component, external device interconnect) bus, or EISA (Extended Industry Standard Architecture ) bus, among others. The buses may be divided into address buses, data buses, control buses, etc. For ease of illustration, only one thick line is shown in fig. 4, but not only one bus or one type of bus.
Alternatively, in a specific implementation, if the memory 401, the processor 402, and the communication interface 403 are integrated on a chip, the memory 401, the processor 402, and the communication interface 403 may perform communication with each other through internal interfaces.
The processor 402 may be a CPU (Central Processing Unit ) or ASIC (Application Specific Integrated Circuit, application specific integrated circuit) or one or more integrated circuits configured to implement embodiments of the present application.
Embodiments of the present application also provide a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the multi-modal fine breast model design method as above.
In the description of the present specification, a description referring to terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. In this specification, schematic representations of the above terms are not necessarily directed to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or N embodiments or examples. Furthermore, the different embodiments or examples described in this specification and the features of the different embodiments or examples may be combined and combined by those skilled in the art without contradiction.
Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying a relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In the description of the present application, "N" means at least two, for example, two, three, etc., unless specifically defined otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and additional implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order from that shown or discussed, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the embodiments of the present application.
It is to be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above-described embodiments, the N steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. As with the other embodiments, if implemented in hardware, may be implemented using any one or combination of the following techniques, as is well known in the art: discrete logic circuits having logic gates for implementing logic functions on data signals, application specific integrated circuits having suitable combinational logic gates, programmable gate arrays, field programmable gate arrays, and the like.
Those of ordinary skill in the art will appreciate that all or a portion of the steps carried out in the method of the above-described embodiments may be implemented by a program to instruct related hardware, where the program may be stored in a computer readable storage medium, and where the program, when executed, includes one or a combination of the steps of the method embodiments.
While embodiments of the present application have been shown and described above, it will be understood that the above embodiments are illustrative and not to be construed as limiting the application, and that variations, modifications, alternatives and variations may be made to the above embodiments by one of ordinary skill in the art within the scope of the application.

Claims (10)

1. The multi-mode fine mammary gland model design method is characterized by comprising the following steps of:
constructing the shape of the breast to obtain the shape of the target breast;
based on the target breast shape, dividing a breast area according to preset breast characteristic parameters to obtain a skin area, a fat area and a fibrous gland area;
respectively performing first refined structure growth on the skin region, second refined structure growth on the fat region and third refined structure growth on the fibrous gland region to obtain a fine mammary gland mathematical model; and
And voxelizing the fine mammary gland mathematical model to obtain a fine mammary gland voxel model, and adjusting the fine mammary gland voxel model based on a preset gland percentage adjustment strategy to obtain the multi-mode fine mammary gland model.
2. The method of claim 1, wherein said constructing a breast shape to obtain a target breast shape comprises:
performing shape fitting on the appearance of the breast based on a plurality of preset hypersurface to obtain an initial breast shape;
and respectively carrying out upper side shaping deformation, inner side flattening deformation, shoulder connection deformation, breast sagging treatment and front side deflection transformation on the initial breast shape based on a preset deformation strategy to obtain a target breast shape.
3. The method of claim 1, wherein the step of determining the position of the substrate comprises,
the skin area and the fat area are divided by adopting a hypersurface contour;
and dividing fat globules obtained by adopting a preset sampling strategy between the fat region and the fibrous gland region.
4. A method according to claim 1 or 3, wherein said respectively performing a first refined structural growth on said skin region, a second refined structural growth on said fat region, and a third refined structural growth on said fibrous glandular region results in a refined mammary mathematical model comprising:
Growing skin, growing nipple and growing areola on the skin area, growing milk infusion sinuses in the nipple, performing Cooper ligament growth on a subcutaneous fat area and a rear fat area of the fat area, sampling fat globule center voxels of the fibrous gland area based on a preset target gland distribution requirement and a target sampling function, and rejecting according to a preset rejection probability to obtain an initial mammary gland model;
and processing the initial mammary gland model based on preset Perlin parting noise to obtain the fine mammary gland mathematical model.
5. The method of claim 4, wherein the target sampling function to evenly distributed sampling function ratio is in the upper bound of the defined domain:
wherein M is R which is more than or equal to 0 and less than or equal to R 0 In the time-course of which the first and second contact surfaces,is the distance between any point of the breast section and the center of the section, R 0 For maximum radius of mammary gland section, f (r) is RAN with gland distribution as Gauss distribution, namely target sampling function, RAN represents normalized radial fat distribution, h (r) is RAN with gland distribution as uniform distribution, A and sigma are fitting parameters, VBD is physical quantity for describing glandular content in mammary gland, and k is->And sigma (sigma) 2 Is a ratio of (2).
6. The method of claim 5, wherein the preset rejection probability is:
wherein r is h Is a random variable sampled from the gland uniform distribution h (R), M is R which is more than or equal to 0 and less than or equal to R 0 In the time-course of which the first and second contact surfaces,is of the upper bound, f (r) h ) As a random variable r h RAN at glandular Gauss distribution f (r), h (r) h ) As a random variable r h RAN with uniform distribution h (R), a and σ are fitting parameters, R 0 Is the maximum radius of the mammary gland section.
7. The method of claim 6, wherein the predetermined Perlin typing noise is:
r≤r base (θ,φ)[1+αP f,l,p,o (θ,φ)]；
wherein r is the distance from any point on the spherical structure surface to the center of the sphere after the Perlin noise is added, and r base (theta, phi) is the distance from any point on the surface of the basic spherical structure to the center of the sphere, theta is the elevation angle of any point on the sphere, phi is the azimuth angle of any point on the sphere, alpha is the amplitude weight factor of the Perlin noise, f is the main frequency weight factor of the Perlin noise, l is the frequency doubling weight factor of the Perlin noise, p is the frequency quadrupling weight factor of the Perlin noise, and o is the eight-time frequency weight factor of the Perlin noise.
8. A multi-modal fine breast model design apparatus, comprising:
the construction module is used for constructing the shape of the breast to obtain the shape of the target breast;
The dividing module is used for dividing the breast area according to the preset breast characteristic parameters based on the target breast shape to obtain a skin area, a fat area and a fibrous gland area;
the refining module is used for respectively carrying out first refining structure growth on the skin area, second refining structure growth on the fat area and third refining structure growth on the fibrous gland area to obtain a fine mammary gland mathematical model; and
and the voxelization module is used for voxelization processing of the fine mammary gland mathematical model to obtain a fine mammary gland voxel model, and adjusting the fine mammary gland voxel model based on a preset gland percentage adjustment strategy to obtain the multi-mode fine mammary gland model.
9. An electronic device, comprising: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor executing the program to implement the multi-modal fine breast model design method of any one of claims 1-7.
10. A computer-readable storage medium having stored thereon a computer program, characterized in that the program is executed by a processor for realizing the multi-modal fine breast model design method as claimed in any one of claims 1 to 7.
CN202310913595.9A 2023-07-24 2023-07-24 Multi-mode fine breast model design method, device, equipment and medium Pending CN117132729A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310913595.9A CN117132729A (en) 2023-07-24 2023-07-24 Multi-mode fine breast model design method, device, equipment and medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310913595.9A CN117132729A (en) 2023-07-24 2023-07-24 Multi-mode fine breast model design method, device, equipment and medium

Publications (1)

Publication Number Publication Date
CN117132729A true CN117132729A (en) 2023-11-28

Family

ID=88851769

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310913595.9A Pending CN117132729A (en) 2023-07-24 2023-07-24 Multi-mode fine breast model design method, device, equipment and medium

Country Status (1)

Country Link
CN (1) CN117132729A (en)

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100014738A1 (en) * 2008-07-21 2010-01-21 Tomtec Imaging Systems Gmbh Method and system for breast cancer screening
WO2011137443A1 (en) * 2010-04-30 2011-11-03 The Trustees Of The University Of Pennsylvania Methods for developing breast phantoms
JP2017116326A (en) * 2015-12-22 2017-06-29 株式会社島津製作所 Image photographing apparatus for breast examination
CN108550150A (en) * 2018-04-17 2018-09-18 上海联影医疗科技有限公司 Acquisition methods, equipment and the readable storage medium storing program for executing of breast density
CN110458813A (en) * 2019-03-08 2019-11-15 腾讯科技（深圳）有限公司 Image-region localization method, device and Medical Image Processing equipment
US20190388045A1 (en) * 2018-06-20 2019-12-26 Shimadzu Corporation Image capturing apparatus for breast examination
MA47566A1 (en) * 2019-12-03 2021-06-30 Univ Hassan Ii Use of il-17 as a prognostic marker in breast cancer
CN114446480A (en) * 2020-11-05 2022-05-06 中国科学院深圳先进技术研究院 Mammary gland phantom generation method, device, equipment and medium
CN115620870A (en) * 2022-09-20 2023-01-17 清华大学 Automatic planning method and system based on dose prediction and parameter optimization
CN116433695A (en) * 2023-06-13 2023-07-14 天津市第五中心医院 Mammary gland region extraction method and system of mammary gland molybdenum target image

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100014738A1 (en) * 2008-07-21 2010-01-21 Tomtec Imaging Systems Gmbh Method and system for breast cancer screening
WO2011137443A1 (en) * 2010-04-30 2011-11-03 The Trustees Of The University Of Pennsylvania Methods for developing breast phantoms
JP2017116326A (en) * 2015-12-22 2017-06-29 株式会社島津製作所 Image photographing apparatus for breast examination
CN108550150A (en) * 2018-04-17 2018-09-18 上海联影医疗科技有限公司 Acquisition methods, equipment and the readable storage medium storing program for executing of breast density
US20190388045A1 (en) * 2018-06-20 2019-12-26 Shimadzu Corporation Image capturing apparatus for breast examination
CN110458813A (en) * 2019-03-08 2019-11-15 腾讯科技（深圳）有限公司 Image-region localization method, device and Medical Image Processing equipment
MA47566A1 (en) * 2019-12-03 2021-06-30 Univ Hassan Ii Use of il-17 as a prognostic marker in breast cancer
CN114446480A (en) * 2020-11-05 2022-05-06 中国科学院深圳先进技术研究院 Mammary gland phantom generation method, device, equipment and medium
CN115620870A (en) * 2022-09-20 2023-01-17 清华大学 Automatic planning method and system based on dose prediction and parameter optimization
CN116433695A (en) * 2023-06-13 2023-07-14 天津市第五中心医院 Mammary gland region extraction method and system of mammary gland molybdenum target image

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
任丽;邱睿;李君利;: "中国成年女性精细乳房模型的建立及其在外照射防护中的应用", 辐射防护, no. 02, 20 March 2016 (2016-03-20), pages 65 - 71 *
夏冰清;柴维敏;: "磁共振扩散加权成像在乳腺疾病诊治中的应用进展", 诊断学理论与实践, no. 05, 25 October 2016 (2016-10-25), pages 528 - 531 *
王文静;邱;睿;任丽;李君利;李春艳;武祯;: "精细乳房模型及其在乳腺X射线摄影剂量评估中的应用", 原子能科学技术, no. 11, 20 November 2016 (2016-11-20), pages 2085 - 2091 *

Similar Documents

Publication Publication Date Title
Rhee et al. Automatic contouring system for cervical cancer using convolutional neural networks
Bakic et al. Development and characterization of an anthropomorphic breast software phantom based upon region‐growing algorithm
CN107072624B (en) System and method for automated treatment planning
Schreier et al. Clinical evaluation of a full-image deep segmentation algorithm for the male pelvis on cone-beam CT and CT
CN107545584A (en) The method, apparatus and its system of area-of-interest are positioned in medical image
US20080107323A1 (en) Computer Diagnosis of Malignancies and False Positives
US9089307B2 (en) Three-dimensional analysis of lesions represented by image data
CN108268897A (en) A kind of multi-modal multi-level computer aided diagnosing method of ultrasound of tumor of breast
Jarosik et al. Breast lesion classification based on ultrasonic radio-frequency signals using convolutional neural networks
CN108885781A (en) For synthesizing the method and system of virtual high dose or high kV computed tomography images according to low dosage or low kV computed tomography images
CN104794749A (en) Method for establishing three-dimensional electromagnetic breast simulation model on basis of clinical MRI (magnetic resonance imaging) images
Chang et al. Development of realistic multi-contrast textured XCAT (MT-XCAT) phantoms using a dual-discriminator conditional-generative adversarial network (D-CGAN)
KR20190125592A (en) 3D Blood Vessel Construction Method using medical images
Oliveira et al. Development of clinically informed 3-D tumor models for microwave imaging applications
WO2023125683A1 (en) Systems and methods for image reconstruction
CN117132729A (en) Multi-mode fine breast model design method, device, equipment and medium
Oya et al. Investigation of clinical target volume segmentation for whole breast irradiation using three-dimensional convolutional neural networks with gradient-weighted class activation mapping
US9192340B2 (en) Imaging method and apparatus with optimized grayscale value window determination
WO2020253745A1 (en) Systems and methods for image processing
Danch-Wierzchowska et al. FEM-based MRI deformation algorithm for breast deformation analysis
Yang et al. Reconstruction of three-dimensional breast-tumor model using multispectral gradient vector flow snake method
Bao et al. Automatic identification and segmentation of orbital blowout fractures based on artificial intelligence
CN116993907A (en) Method, device and equipment for establishing compression breast statistical model based on DBT image
Rodriguez Ruiz Artificial intelligence and tomosynthesis for breast cancer detection
Kim 3D volume extraction of cerebrovascular structure on brain magnetic resonance angiography data sets

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination