Human body full-size respiratory rhythm respiration three-dimensional geometric model and construction method thereof

Abstract

The invention relates to the field of intelligent medical treatment, in particular to a human body full-size respiratory rhythm respiration three-dimensional geometric model and a construction method thereof. The method includes: acquiring image data and anatomical data of a respiratory tract; performing first respiratory tract three-dimensional reconstruction based on the image data to obtain a first respiratory tract geometric model; performing second respiratory tract three-dimensional reconstruction based on the anatomical data to obtain a second respiratory tract geometric model; the first respiratory tract geometric model and the second respiratory tract geometric model are spliced to obtain the full-size respiratory tract geometric model, the full-size respiratory tract geometric model is constructed, and compared with an existing model, the full-size respiratory tract geometric model is more comprehensive and fits the whole particle motion process in the actual respiratory tract; basic conditions are provided for research on the rule of respiratory tract particle deposition, and good clinical value is achieved.

Description

Technical Field

[0001] The present application relates to the field of intelligent medical treatment, and specifically to a full-scale human respiratory rhythmic breathing three-dimensional geometric model and its construction method, equipment, program product and computer-readable storage medium. Background Art

[0002] In recent years, respiratory diseases have become the second largest disease endangering human health. Targeted drug delivery is to use an aerosol generator to atomize drugs into aerosols and deliver them to the respiratory lesion area through the respiratory airflow. It has the characteristics of direct action, rapid onset, high lung concentration, low systemic concentration, and small side effects. It is widely used in the treatment of respiratory diseases. The efficacy of targeted drug delivery depends on the total deposition dose of the drug in the lungs and the local deposition dose of the target lung tissue. Existing studies have shown that the deposition dose of aerosol particles is related to many factors, such as particle size, breathing mode, and particle physicochemical properties (hydrophilicity, density), among which particle size is the most important factor. The respiratory tract is a high-humidity environment. Hydrophilic particles will grow rapidly when they enter the respiratory tract. As the respiratory airflow flows, particles are prone to collision and agglomeration, which in turn affects the particle size. The structure of the respiratory tract is extremely complex and is a non-transparent tissue. Therefore, it is of great significance to predict the deposition law of particles of different sizes and different physicochemical properties under different breathing modes. The particle deposition numerical model mainly integrates the three-dimensional geometric model of the respiratory tract, the airflow flow physical field, and the particle deposition physical field. It can obtain both high-resolution particle deposition distribution parameters and dynamic information of particles. Therefore, it is an ideal means to study the inhalation deposition mechanism. However, the construction of the three-dimensional geometric model of the respiratory tract focuses on the upper respiratory tract and the upper reaches of the lower respiratory tract, and the construction of the three-dimensional geometric model of the entire respiratory tract is rare, which in turn affects the study of the deposition laws of different particles in the particle deposition numerical model.

[0003] Secondly, the control equations for the construction of the existing particle deposition physics field are mature, but the influencing factors are not fully included. The existing particle deposition physics field model still has major shortcomings, mainly reflected in the insufficient consideration of the interaction between particles, particles, respiratory tract, and respiratory microenvironment. At the same time, the existing numerical simulation models have not been verified by in vivo experiments, and the correctness of the models cannot be guaranteed. Therefore, the existing models will have large deviations in predicting the deposition distribution of aerosol particles, especially hydrophilic particles.

[0004] In addition, particles collide in the respiratory tract, the particle collision time is extremely short, and the particle size changes extremely quickly. The movement trajectory morphology of the same particle in different generations of airways across scales is extremely different. The existing mathematical models and numerical models are not perfect enough to study the collision and agglomeration process during particle diffusion and deposition. At the same time, the numerical models all use segmented research and cannot obtain the complete deposition process in the respiratory tract. At the same time, they have not been verified by in vivo experimental results, and thus the research results have significant deviations. Summary of the invention

[0005] In view of the above problems, the present invention proposes a human full-scale respiratory rhythmic breathing three-dimensional geometric model and a construction method thereof, which specifically includes: Obtain respiratory imaging data, anatomical data, and respiratory tidal volume; Performing three-dimensional reconstruction of the first respiratory tract based on the image data to obtain a first respiratory tract geometric model; Performing three-dimensional reconstruction of the second respiratory tract based on the anatomical data to obtain a second respiratory tract geometric model; The first respiratory tract geometric model and the second respiratory tract geometric model are spliced ​​together to obtain a full-size respiratory tract geometric model.

[0006] Constructing a geometric deformation physical field according to the respiratory tidal volume; The full-scale respiratory rhythm breathing three-dimensional geometric model is obtained through the full-scale respiratory geometry model and geometric deformation physical field Optionally, the first respiratory tract is G 0 - G 6 segmental respiratory tract; Optionally, the second respiratory tract is G 7 - G 23 segmental respiratory tract; The process of the second airway three-dimensional reconstruction is: Obtain anatomical data, including airway dimensions, average alveolar dimensions, and respiratory tract geometric characteristics; Perform two-dimensional tessellation of alveoli based on the airway size and the average alveolar size to obtain two-dimensional tessellation data; A second respiratory tract geometric model is obtained by performing three-dimensional stretching on the two-dimensional tessellated data and the respiratory tract geometric features; Optionally, the two-dimensional acinar tessellation arranges the arrangement of the bronchi by symmetrically truncating them; The object of the present invention is to provide a computer program product having a computer program or instructions thereon, characterized in that it includes: the computer program or instructions are executed by a processor to implement the above-mentioned full-size human respiratory rhythmic breathing three-dimensional geometric model and its construction method.

[0007] The object of the present invention is to provide a computer device, which includes a memory, a processor and a computer program or instructions stored in the memory, wherein the computer program or instructions are executed by the processor to implement the above-mentioned full-size human respiratory rhythmic breathing three-dimensional geometric model and its construction method.

[0008] The object of the present invention is to provide a computer-readable storage medium having a computer program or instruction stored thereon, wherein the computer program or instruction is executed by a processor to implement the above-mentioned full-size human respiratory rhythmic breathing three-dimensional geometric model and its construction method.

[0009] The object of the present invention is to provide a method for constructing a respiratory tract particle deposition model of rhythmic breathing, comprising: Obtain airflow pattern type, particle properties, respiratory tidal volume, and full-scale airway geometry model; Determine the airflow physics field and particle tracking physics field according to the fluid flow type and particle properties; A rhythmic breathing respiratory tract particle deposition model is obtained through the airflow flow physical field, particle tracking physical field, and full-scale respiratory tract rhythmic breathing three-dimensional geometric model.

[0010] Optionally, the full-size airway geometric model is obtained according to the above-mentioned human full-size airway rhythmic breathing three-dimensional geometric model and its construction method; Optionally, the airflow physical field is obtained by training a first neural network, and the first neural network learns the correspondence between the fluid flow type and the airflow physical field for classification training; Optionally, the particle tracking physical field is to perform force analysis on fluid particles according to the airflow state of rhythmic breathing and particle properties to obtain force results, and construct a particle momentum equation based on the force results; Optionally, the force parameters calculated by the particle momentum equation include one or more of the following: drag force, gravity, Brownian force, pressure gradient force, virtual mass force, lift force, Bassett force; The method also includes a particle collision and agglomeration physical field, which is determined according to the fluid flow type and particle properties, and a rhythmic breathing respiratory tract particle deposition model is obtained through the airflow physical field, particle tracking physical field, particle collision and agglomeration physical field, geometric deformation physical field, and full-scale respiratory tract geometric model; Optionally, the particle collision and agglomeration physical field is constructed based on a particle collision and agglomeration equation calculated based on the flow pattern type of the fluid and the airflow changes of the particle properties under rhythmic breathing; Optionally, the calculation of the particle collision agglomeration equation includes a first collision calculation and a second collision calculation, wherein the collision parameters of the first collision calculation include Brownian collision, inertial collision, and turbulent collision; and the collision parameters of the second collision calculation include Brownian collision and inertial collision. Optionally, the first collision calculation is used to calculate G 0 - G 6 The second collision calculation is used to calculate the particle collision of the respiratory tract.G 7 - G 23 Particle collisions in the respiratory tract; Optionally, the geometric deformation physical field is obtained based on the respiratory deformation equation by calculating the expansion amplitude of the alveoli based on the respiratory tidal volume and constructing the respiratory tract deformation equation through the expansion amplitude; Optionally, the geometric deformation physical field is obtained by training a third neural network, and the third neural network learns the correspondence between the respiratory tidal volume and the geometric deformation physical field for classification training.

[0011] The respiratory tract particle deposition model also includes model correction, obtaining animal inhalation experiment result data, correcting the respiratory tract particle deposition model based on the animal inhalation experiment result, and obtaining a corrected respiratory tract particle deposition model; Optionally, the model correction is replaced by: obtaining human inhalation experiment result data and animal inhalation experiment result data, and correcting the model based on the human inhalation experiment result data. G 0 - G 6 Segmental airway particle deposition model, calibrated based on animal inhalation test results G 7 - G 23 The segmental respiratory tract particle deposition model is used to obtain the calibrated respiratory tract particle deposition model.

[0012] The object of the present invention is to provide a computer program product having a computer program or instructions thereon, including: the computer program or instructions are executed by a processor to implement the above-mentioned method for constructing a respiratory particle deposition model based on rhythmic breathing.

[0013] The object of the present invention is to provide a computer device, which includes a memory, a processor and a computer program or instructions stored in the memory, wherein the computer program or instructions are executed by the processor to implement the above-mentioned method for constructing a respiratory particle deposition model based on rhythmic breathing.

[0014] An object of the present invention is to provide a computer-readable storage medium having a computer program or instruction stored thereon, wherein the computer program or instruction is executed by a processor to implement the above-mentioned method for constructing a respiratory particle deposition model based on rhythmic breathing.

[0015] The object of the present invention is to provide a method for predicting the distribution of respiratory tract particle deposition, comprising: Obtain respiratory rate, tidal volume, and particle parameters; Calculate the airflow velocity based on the respiratory frequency, and calculate the expansion amplitude of the alveoli based on the tidal volume; The airflow velocity, the expansion amplitude of the alveoli, and the particle parameters are input into the respiratory tract particle deposition model obtained by the above-mentioned method for constructing a respiratory tract particle deposition model based on rhythmic breathing, and the model parameters are set to obtain a set respiratory tract particle deposition model; The deposition rate and deposition distribution are predicted by the set respiratory particle deposition model.

[0016] The method also includes analyzing the particle deposition mechanism, recording the movement trajectory of the particles during the deposition process, marking the deposition mechanism of each particle by comparing the particle deposition mechanism standard, and obtaining the deposition mechanism of different particles in different regions based on the particle movement trajectory and the particle deposition mechanism; Optionally, the particle parameters include one or more of the following: quantity, particle size, density, hydrophilicity and hydrophobicity; Optionally, the respiratory particle deposition model parameter setting sets the boundary conditions of the bronchi and bronchiolar terminals of the airflow flow physics field to the airflow velocity, sets the deformation parameters of the acinus in the geometric deformation physics field to the expansion amplitude of the acinus, and sets the particle parameters in the particle tracking physics field; Optionally, the parameter setting further includes setting the boundary condition of the oral cavity to zero pressure; obtaining particle collision parameters, and setting the particle collision parameters in a particle tracking physical field; Optionally, the deposition rate includes one or more of the following: regional deposition rate, total deposition rate; Optionally, the region includes one or more of the following: pharynx, trachea, left bronchus, right bronchus, left bronchiole, right bronchiole, left acinus, right acinus.

[0017] The object of the present invention is to provide a computer program product having a computer program or instructions thereon, including: the computer program or instructions are executed by a processor to implement the above-mentioned method for predicting respiratory tract deposition distribution.

[0018] An object of the present invention is to provide a computer device, which includes a memory, a processor, and a computer program or instructions stored in the memory, wherein the computer program or instructions are executed by the processor to implement the above-mentioned method for predicting respiratory tract deposition distribution.

[0019] An object of the present invention is to provide a computer-readable storage medium having a computer program or instructions stored thereon, wherein the computer program or instructions are executed by a processor to implement the above-mentioned method for predicting respiratory tract deposition distribution.

[0020] Advantages of the present invention: 1. A method for constructing a full-size respiratory tract geometric model is proposed. In view of the problem that the particle deposition numerical model simulates the segmented study of particle deposition and lacks the continuous particle deposition movement throughout the whole process, a full-size respiratory tract geometric model is constructed. The upper and upper reaches of the lower respiratory tract are reconstructed in three dimensions through imaging data, and the densely packed alveoli in the lower respiratory tract are reconstructed in three dimensions through anatomical data. A full-size respiratory tract geometric model is obtained based on the three-dimensional reconstruction of the two parts, which is helpful for subsequent research on the deposition of different particles in the full-size respiratory tract.

[0021] 2. In view of the problem that the respiratory tract does not sufficiently consider the interaction between particles, particles, respiratory tract, and respiratory microenvironment, the present invention integrates multiple force influencing factors to construct a particle motion model for the movement of particles in the respiratory tract, which is used to fit the movement trajectory of particles in real scenes, thereby improving the accuracy of particle deposition distribution prediction.

[0022] 3. Adding the influencing factor of particle collision in the numerical model of particle deposition is helpful to fit the situation of particles under the action of rhythmic breathing in real respiratory scenarios, which is beneficial to improve the accuracy of particle deposition movement, and then improve the accuracy of particle deposition distribution, and obtain accurate deposition motion trajectory and deposition mechanism.

[0023] 4. Particle deposition prediction is performed based on the particle deposition model to obtain the movement process of particles in the full-size respiratory tract, the deposition of different particles in different areas, the deposition distribution and movement trajectory of different particles, and the deposition mechanism of different particles in the full-size respiratory tract based on the prediction results. This helps to understand the movement of different particles in the respiratory tract, and can not only target the distribution and deposition of harmful particles, but also perform auxiliary diagnosis such as efficacy analysis and dosage control for drug particles in targeted therapy.

[0024] 5. The present invention uses in vivo inhalation experimental data to calibrate the particle deposition numerical model, thereby improving the accuracy of the model and reducing prediction errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 A schematic diagram of a full-scale human respiratory rhythmic breathing three-dimensional geometric model and a construction method thereof provided in an embodiment of the present invention; Figure 2A schematic diagram of a system for constructing a full-scale human respiratory rhythmic breathing three-dimensional geometric model provided by an embodiment of the present invention; Figure 3 A schematic diagram of a device for constructing a full-scale human respiratory rhythmic breathing three-dimensional geometric model provided by an embodiment of the present invention; Figure 4 The embodiment of the present invention provides G 0 - G 6 Schematic diagram of three-dimensional reconstruction; Figure 5 The embodiment of the present invention provides G 7 - G 23 Schematic diagram of three-dimensional reconstruction; Figure 6 A schematic diagram of the process of constructing a full-scale rhythmic breathing respiratory tract particle deposition model provided by an embodiment of the present invention; Figure 7 A schematic diagram of directional displacement during breathing deformation provided by an embodiment of the present invention; Figure 8 A schematic flow chart of a deposition prediction method provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0028] In some of the processes described in the specification and claims of the present invention and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this article or executed in parallel. The sequence numbers of the operations, such as S101, S102, etc., are only used to distinguish between different operations, and the sequence numbers themselves do not represent any execution order. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this article are used to distinguish different messages, devices, modules, etc., do not represent the order of precedence, and do not limit the "first" and "second" to be different types.

[0029] Figure 1 The schematic diagram of the full-scale human respiratory rhythmic breathing three-dimensional geometric model and its construction method provided by the embodiment of the present invention specifically includes: S1. Obtain imaging data and anatomical data of the respiratory tract; In one embodiment, the image data includes one or more of the following: CT, MRI, ultrasound, X-ray, and nuclear imaging.

[0030] In one embodiment, the anatomical data includes the cross-sectional size of the airway opening, the length of the airway, and the size of the pulmonary alveoli. The data of the pulmonary alveoli are averaged to obtain the average size of the pulmonary alveoli.

[0031] S2. Performing three-dimensional reconstruction of the first respiratory tract based on the image data to obtain a first respiratory tract geometric model; In one embodiment, the first airway is G 0 - G 6 Segmental respiratory tract.

[0032] In a specific embodiment, the human respiratory system CT slice acquisition and G 0 - G 6 Reconstruction of three-dimensional airway geometry model (e.g. Figure 4 The process shown is: Adult males were selected for respiratory system CT slice collection with a resolution of 1 mm. After the scan was completed, the scanned slices were 3D reconstructed to obtain 3D CT tomographic slices of the respiratory system. Materialise's Mimics software was then used to process and reconstruct the 3D CT slices. G 0 - G 6 The specific reconstruction process is to manually outline the airway contour, where the airway color in the CT slice is black, and then use the default algorithm of Materialise's Mimics software to perform G 0 - G 6 Three-dimensional airway reconstruction.

[0033] S3, performing three-dimensional reconstruction of the second respiratory tract based on the anatomical data to obtain a second respiratory tract geometric model; In one embodiment, the second airway is G 7 - G 23 Segmental respiratory tract.

[0034] In one embodiment, the process of the second airway three-dimensional reconstruction is: Obtain anatomical data, including airway dimensions, average alveolar dimensions, and respiratory tract geometric characteristics; Perform two-dimensional tessellation of alveoli based on the airway size and the average alveolar size to obtain two-dimensional tessellation data; The second respiratory tract geometric model is obtained by performing three-dimensional stretching on the two-dimensional dense paving data and the respiratory tract geometric features.

[0035] In one embodiment, the two-dimensional alveolar paving is performed by symmetrically truncating the arrangement of the bronchi, and the geometric characteristic data of the respiratory tract include one or more of the following: the length of the respiratory tract and the size of the cross-section of the respiratory tract.

[0036] In one embodiment, the human body is idealized into a single path G 7 - G 18 Bronchial tubes and densely packed alveoli ( G 19 ~ G 23 ) 3D airway geometry model reconstruction (e.g. Figure 5 The process is as follows: Based on the human respiratory tract anatomical data and respiratory tract geometric characteristics obtained from the literature, taking into account the current level of computer development, the UNIGRAPHICS software is used to construct an idealized single-path symmetric human body. G 7 - G 18 The geometric model of the bronchus and the densely packed alveoli in the three-dimensional respiratory tract. The specific reconstruction process is to perform a two-dimensional densely packed arrangement of the alveoli according to the airway size and the average size of the alveoli. G 7 - G 18 The bronchi were arranged in a symmetrical cutout pattern and then stretched to form a three-dimensional idealized single path. G 7 - G 18 Reconstruction of the geometric model of bronchi and densely packed alveoli.

[0037] S4. Splicing the first respiratory tract geometric model with the second respiratory tract geometric model to obtain a full-size respiratory tract geometric model.

[0038] In one specific embodiment, ANSYS Spaceclaim software is used to G 0 ~ G 6 Three-dimensional respiratory tract geometry model, idealized single path G 7 ~ G 18The bronchial and densely packed alveolar 3D respiratory tract geometric models are spliced ​​to obtain a full-size single-path 3D respiratory tract geometric model of the human body. The specific splicing method is to first place the two models in the same file, then use the move function to match the positions of the two models, and after the correspondence is completed, use the fusion function to splice the two models to obtain a full-size single-path 3D respiratory tract geometric model of the human body.

[0039] The disclosed embodiment of the present invention also provides a computer program product or system, including a computer program, which, when executed by a processor, implements the above-mentioned full-size human respiratory rhythmic breathing three-dimensional geometric model and its construction method steps.

[0040] Figure 2 A schematic diagram of a system for constructing a full-scale human respiratory rhythmic breathing three-dimensional geometric model provided by an embodiment of the present invention specifically includes: Acquisition unit: acquires image data and anatomical data of the respiratory tract; A first reconstruction unit: performing a three-dimensional reconstruction of the first respiratory tract based on the image data to obtain a first respiratory tract geometric model; A second reconstruction unit: performing three-dimensional reconstruction of the second respiratory tract based on the anatomical data to obtain a second respiratory tract geometric model; Full-size unit: The first respiratory tract geometry model is spliced ​​with the second respiratory tract geometry model to obtain a full-size respiratory tract geometry model.

[0041] Figure 3 A schematic diagram of a device for constructing a full-scale human respiratory rhythmic breathing three-dimensional geometric model provided by an embodiment of the present invention specifically includes: A memory and a processor; the memory is used to store program instructions; the processor is used to call program instructions, and when the program instructions are executed, any one of the above-mentioned full-size respiratory rhythmic breathing three-dimensional geometric models of the human body and the construction method thereof is obtained.

[0042] The disclosed embodiment of the present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, any one of the above-mentioned full-size respiratory rhythmic breathing three-dimensional geometric models of the human body and the construction method thereof is obtained.

[0043] Figure 6 The present invention proposes a method for constructing a respiratory tract particle deposition model based on rhythmic breathing, which specifically includes: S101, obtain fluid flow type, particle properties, respiratory tidal volume, and full-size respiratory tract geometry model; In one embodiment, the full-size airway geometric model is obtained based on the above-mentioned human full-size airway rhythmic breathing three-dimensional geometric model and its construction method.

[0044] In another embodiment, the method for constructing the full-scale geometric respiratory tract model is specifically as follows: The CT image data of the tester is obtained, and the CT image is segmented by threshold value to obtain the respiratory area; then the respiratory area is reconstructed in three dimensions to obtain a real respiratory model including the nasal cavity, oral cavity, pharynx, main bronchus, lobar bronchus, and segmental bronchus; The idealized bronchial tree model was obtained by using the Weibel model and a three-dimensional rotation matrix with random angles to generate the bronchial tree centerline skeleton for subdivision modeling. A random point cloud was generated according to the size of the alveoli, and three-dimensional Thiessen polygons were generated from the random point cloud. The alveolar path was constructed in the three-dimensional Thiessen polygons using the particle swarm optimization algorithm / slime mold growth algorithm. The four-sided reconstruction and subdivision modeling were used for smoothing to obtain an idealized polyhedral alveolar model of the respiratory tract terminal. Before smoothing, the polygonal faces passing through the alveolar path were deleted and the edges of the two adjacent three-dimensional Thiessen polygons were connected to obtain the flow channel connecting the alveoli, and geometric repair was performed to create the entrance of the idealized polyhedral alveolar model.

[0045] The real respiratory tract model, the idealized bronchial tree model, and the idealized polyhedral alveolar model at the end of the respiratory tract were spliced ​​together to obtain a full-size respiratory tract geometric model.

[0046] S102, determining the airflow physical field and the particle tracking physical field according to the fluid flow type and particle properties; In one embodiment, the fluid flow state includes laminar flow and turbulent flow; laminar flow is a flow state of the fluid, characterized in that the fluid particles flow along smooth layers, and each layer of fluid flows smoothly through the adjacent layers with little or no mixing. In the laminar flow state, the liquid particles do not interfere with each other, and the flow of the liquid is linear or laminar, parallel to the axis of the pipe. Turbulent flow is another fluid flow state, characterized in that the movement of the liquid particles is chaotic, and in addition to the movement parallel to the axis of the pipe, there is also violent lateral movement. In the turbulent state, the liquid particles will mix with each other, and the flow state is more complicated.

[0047] In one embodiment, the particle tracking physical field performs force analysis on fluid particles according to the airflow state of rhythmic breathing and particle properties to obtain force results, and constructs a particle momentum equation based on the force results.

[0048] In one embodiment, the force parameters calculated by the particle momentum equation include one or more of the following: drag force, gravity, Brownian force, pressure gradient force, virtual mass force, lift, and Bassett force.

[0049] In one embodiment, the airflow flow physical field analyzes the airflow under rhythmic breathing by analyzing the fluid flow type and particle properties to obtain an airflow analysis result, and constructs an airflow flow equation based on the airflow analysis result.

[0050] In another embodiment, the airflow flow physical field is obtained by training a first neural network, and the first neural network learns the correspondence between the airflow flow type and the airflow flow physical field for classification training.

[0051] In one embodiment, the particle tracking physical field is obtained by training a second neural network, and the second neural network learns the correspondence between flow fluid type and particle properties and the particle tracking physical field for classification training.

[0052] In one embodiment, the method further includes determining a particle collision and agglomeration physical field, determining the particle collision and agglomeration physical field according to the fluid flow type and particle properties, and obtaining a rhythmic breathing respiratory tract particle deposition model through the airflow physical field, particle tracking physical field, particle deposition and agglomeration physical field, geometric deformation physical field, and full-scale respiratory tract geometric model; In one embodiment, the particle collision and agglomeration physical field is constructed based on the particle collision and agglomeration equation calculated based on the flow pattern type of the fluid and the airflow changes of the particle properties under rhythmic breathing; In one embodiment, the calculation of the particle collision agglomeration equation includes a first collision calculation and a second collision calculation, wherein the collision parameters of the first collision calculation include Brownian collision, inertial collision, and turbulent collision; and the collision parameters of the second collision calculation include Brownian collision and inertial collision.

[0053] In one embodiment, the first collision calculation is used to calculate G 0 - G 6 The second collision calculation is used to calculate the particle collision of the respiratory tract. G 7 - G 23 Particle collision in the respiratory tract.

[0054] In another embodiment, the particle collision and agglomeration physical field is obtained by training a fourth neural network, and the fourth neural network learns the correspondence between flow fluid type and particle properties and the particle deposition and agglomeration physical field for classification training.

[0055] In a specific embodiment, the air flow equation construction process is: Considering that the airflow in the respiratory tract is laminar and incompressible, the continuity equation and momentum equation are: Continuity equation:

[0056] The momentum equation has two parts: G 0 - G 6 Turbulence:

[0057]

[0058]

[0059] G 7 -G 23 Laminar Flow:

[0060] Where u represents the fluid velocity, ρ represents the air density, p represents the air flow pressure, v represents the air viscosity, and k represents the turbulent kinetic energy. represents the turbulent dissipation rate, is the turbulent Prandtl number, Shear stress tensor, is k dissipation, yes dissipation, is a constant.

[0061] In a specific embodiment, the particle momentum equation is constructed as follows: against G 0 - G 23 The particle momentum equation is obtained by force analysis of micro-nanoparticles in the respiratory tract. The particle momentum equation is expressed as:

[0062] Among them, drag represents drag force, gravitation represents gravity, Brown represents Brownian force, pressure gradient represents pressure gradient force, virtual mass represents virtual mass force, and lift represents lift.

[0063] In a specific embodiment, G 0 - G 6 and G 7 - G 23 The expressions and magnitudes of the forces are:

[0064]

[0065] in, is the aerodynamic viscosity, is the particle mass, is the particle velocity, is the particle density, is the particle size, is the Cunningham correction coefficient, is the acceleration due to gravity, is a random number, represents the Boltzmann constant, is the Kelvin temperature, represents the particle radius, is the time step, is the virtual quality coefficient, Represents relaxation time.

[0066] Considering that the magnitude of the Bassett force is much smaller than other forces, the Bassett force is discarded and the particle momentum equation is obtained by integrating the remaining forces.

[0067] In a specific embodiment, the particle collision agglomeration equation is constructed as follows: Considering G 0 - G 6 The main reasons for the collision of micro-nano particles caused by rhythmic breathing turbulent flow are Brownian collision, inertial collision and turbulent collision. Therefore, the particle collision probability equation is:

[0068] Considering G 7 -G 23 The main causes of micro-nanoparticle collisions under rhythmic breathing laminar flow conditions are Brownian collisions and inertial collisions, so the particle collision probability equation is:

[0069] G 0 -G 23 The collision aggregation equation is:

[0070]

[0071]

[0072] Where P represents the collision probability, represents the mass of particle "1" at time t, represents the particle radius, represents the particle density, represents the velocity difference between the two particles, , Represents the distance between two particles.

[0073] S103, determining a geometric deformation physical field according to the respiratory tidal volume; In one embodiment, the geometric deformation physical field is obtained by calculating the expansion amplitude of the alveoli based on the respiratory tidal volume and constructing a respiratory deformation equation through the expansion amplitude, and then based on the respiratory deformation equation.

[0074] Specifically, the respiratory deformation equation is generated based on the expansion amplitude of the alveoli in the full-size respiratory tract geometric model and the displacement of each plane in each movement direction obtained by calculating the respiratory tidal volume.

[0075] In another embodiment, the geometric deformation physical field is obtained by training a third neural network, and the third neural network learns the correspondence between the respiratory tidal volume and the geometric deformation physical field for classification training.

[0076] In one embodiment, the first neural network, the second neural network, the third neural network, and the fourth neural network are trained by inputting different input data into different or the same neural networks, and the neural networks include but are not limited to one or more of the following: convolutional neural network, void convolutional neural network, residual network, and Transformer.

[0077] In a specific embodiment, the process of constructing the respiratory tract deformation equation is: The expansion ratio is used to determine the expansion amplitude of the airway according to the tidal volume of the breath. Since the breathing waveform is approximately sinusoidal breathing, the motion equation of each plane of the alveoli is:

[0078] Among them, i, j, k represent the numbers of each plane perpendicular to the x, y, and z directions respectively, and l(i or j or k) represents the product of the displacement of l(i) or l(j) or l(k) relative to the center of the topmost plane (red plane) of the model and the expansion ratio. Figure 7 shown.

[0079] S104, obtaining a rhythmic breathing respiratory tract particle deposition model through the airflow physics field, particle tracking physics field, geometric deformation physics field, and full-scale respiratory tract geometry model.

[0080] In one embodiment, the airflow flow physical field and the particle tracking physical field are determined according to the airflow equation and the particle momentum equation generated according to the fluid flow type and the particle properties. The deformation motion of the respiratory tract is calculated based on the full-scale respiratory tract geometric model and the characteristics of rhythmic breathing to generate the respiratory tract deformation equation, and the geometric deformation physical field is generated through the respiratory tract deformation equation. The full-scale respiratory tract geometric model, the airflow flow physical field, the particle tracking physical field, and the geometric deformation physical field are integrated to form a full-scale respiratory tract rhythmic breathing particle deposition numerical model.

[0081] In another embodiment, the airflow flow equation and particle momentum equation generated according to the fluid flow type and the characteristics of rhythmic breathing are used to calculate the airflow flow physical field and the particle tracking physical field. On this basis, and in accordance with the fluid flow type and the characteristics of rhythmic breathing, a particle collision equation is generated to calculate the particle collision and agglomeration physical field. Then, based on the full-size respiratory tract geometric model and the characteristics of rhythmic breathing, the deformation movement of the respiratory tract is calculated to generate the respiratory tract deformation equation, and the geometric deformation physical field is generated through the respiratory tract deformation equation. The airflow physical field, the particle collision and agglomeration physical field, and the geometric deformation physical field are used to constitute the rhythmic breathing movement state.

[0082] In one embodiment, the respiratory tract particle deposition model further includes model correction, obtaining animal inhalation experiment result data, correcting the respiratory tract particle deposition model based on the animal inhalation experiment result, and obtaining a corrected respiratory tract particle deposition model. The correction adjusts the deposition model parameters according to the experimental results to achieve correction.

[0083] In another embodiment, the model correction is replaced by: obtaining human inhalation experiment result data and animal inhalation experiment result data, and correcting the model based on the human inhalation experiment result data. G 0 - G 6 Segmental airway particle deposition model, calibrated based on animal inhalation test results G 7 - G 23 The segmental respiratory tract particle deposition model is used to obtain the calibrated respiratory tract particle deposition model.

[0084] In a specific embodiment, the airflow equation is first selected according to the fluid flow state, and then the force analysis of the particle motion process is performed to determine the final force on the particles (such as gravity, drag force, Brownian force, etc.) to construct the particle momentum equation and the particle collision and agglomeration equation, and the expansion amplitude of the alveoli (dynamic grid parameters) is determined according to the respiratory tidal volume to construct the respiratory deformation equation. Finally, computational fluid dynamics software is used to select the airflow flow physics field, particle tracking physics field, and particle collision and agglomeration physics field according to the constructed equations, and combined with the human body's full-scale single-path rhythmic breathing three-dimensional respiratory tract geometric model to obtain a full-scale single-path rhythmic breathing three-dimensional high-humidity respiratory tract particle deposition numerical model. Correction using human inhalation experiment results G 0 - G 6 The accuracy of the numerical model. Considering the similarities between the guinea pig respiratory tract and human G 7 - G 23 The fluid flow and particle motion in the respiratory tract are similar, with the same Reynolds number and Strouhal number for the fluid flow and the same Stokes number for the particles. Therefore, the parameters of the existing guinea pig particle deposition numerical model calibrated by experimental results are used to correct the G 7 - G 23 Accuracy of numerical models.

[0085] The disclosed embodiments of the present invention further provide a computer program product or system, including a computer program, which, when executed by a processor, implements the above-mentioned method steps for constructing a respiratory particle deposition model based on rhythmic breathing.

[0086] The system for constructing a respiratory particle deposition model based on rhythmic breathing provided by an embodiment of the present invention specifically includes: Acquisition module: obtains fluid flow type, particle properties, respiratory tidal volume, and full-size respiratory tract geometry model; Motion module: Constructs a fluid particle motion model based on the fluid flow type and particle properties to determine the airflow physical field and particle tracking physical field; Deformation module: constructing a respiratory deformation model according to the respiratory tidal volume and determining a geometric deformation physical field; Deposition module: A rhythmic breathing respiratory tract particle deposition model is obtained through the airflow flow physics field, particle tracking physics field, geometric deformation physics field, and full-scale respiratory tract geometry model.

[0087] The device for constructing a respiratory tract particle deposition model based on rhythmic breathing provided in an embodiment of the present invention specifically includes: A memory and a processor; the memory is used to store program instructions; the processor is used to call the program instructions, and when the program instructions are executed, any one of the above-mentioned methods for constructing a respiratory particle deposition model based on rhythmic breathing is implemented.

[0088] The disclosed embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, any one of the above-mentioned methods for constructing a respiratory particle deposition model based on rhythmic breathing is implemented.

[0089] Figure 8 A method for predicting respiratory tract particle deposition distribution provided in an embodiment of the present invention includes: Obtain respiratory rate, tidal volume, and particle parameters; Calculate the expansion amplitude of the alveoli and the airflow velocity based on the respiratory rate and tidal volume; The airflow velocity, the expansion amplitude of the alveoli, and the particle parameters are input into the respiratory tract particle deposition model obtained by the above-mentioned method for constructing a respiratory tract particle deposition model based on rhythmic breathing, and the model parameters are set to obtain a complete respiratory tract particle deposition model; The deposition rate and distribution are predicted by setting up the respiratory particle deposition model.

[0090] In one embodiment, the method further comprises calculating particle deposition rules, and calculating the deposition rules of different particles according to the regional deposition rate and the total deposition rate.

[0091] In one embodiment, the method also includes analyzing the particle deposition mechanism, recording the movement trajectory of the particles during the deposition process, marking the deposition mechanism of each particle by comparing the particle deposition mechanism standard, and obtaining the deposition mechanism of different particles in different regions based on the particle movement trajectory and the particle deposition mechanism judgment standard.

[0092] In one embodiment, the particle parameters include one or more of the following: quantity, particle size, density, and hydrophilicity.

[0093] In one embodiment, the respiratory particle deposition model parameters are set to set the boundary conditions of the bronchi and bronchiolar ends of the airflow flow physics field to the airflow velocity, set the deformation parameters of the alveoli in the geometric deformation physics field to the expansion amplitude of the alveoli, and set the particle parameters in the particle deposition physics field.

[0094] In one embodiment, the parameter setting further includes setting the boundary condition of the oral cavity to zero pressure; obtaining particle collision parameters, and setting the particle collision parameters in the particle deposition physical field.

[0095] In one embodiment, the deposition rate includes one or more of the following: regional deposition rate, total deposition rate.

[0096] In one embodiment, the region includes one or more of the following: pharynx, trachea, left bronchus, right bronchus, left bronchiole, right bronchiole, left acinus, right acinus.

[0097] In a specific embodiment, the input respiratory parameters, particle attributes and other parameters are set in the particle deposition numerical model. The airflow velocity at the outlet and the expansion amplitude of the alveoli are calculated according to the respiratory frequency and tidal volume. The boundary condition of the outlet (the end of the bronchi and bronchioles) is set to the airflow velocity in the airflow flow physics field, and the boundary condition of the inlet (oral cavity) is set to the pressure of 0Pa. The deformation parameters of the alveoli are set in the geometric deformation physics field. The particle attribute parameters and the number of released particles are set in the particle deposition physics field, and the particle collision parameters are set.

[0098] Computational fluid dynamics software was used to run the particle deposition numerical model to predict the particle deposition law. After the calculation was completed, the particle deposition amount in different areas and the total amount of deposited particles in the complete respiratory tract were post-processed using computational fluid dynamics software. The particle deposition law was then calculated and the total deposition rate was calculated. DF total and regional sedimentation rate DF i Indicates that i Represents respiratory stage.

[0099]

[0100]

[0101] After the calculation is completed, the movement trajectory of the particles during the deposition process is tracked, and the deposition mechanism of each particle is marked by comparing the particle deposition mechanism standard, thereby analyzing the deposition mechanism of particles of different sizes in different regions.

[0102] The disclosed embodiments of the present invention further provide a computer program product or system, including a computer program, which implements the above-mentioned steps of the method for predicting respiratory tract particle deposition distribution when executed by a processor.

[0103] The prediction system for respiratory tract particle deposition distribution provided by the embodiment of the present invention specifically includes: Parameter unit: obtain respiratory rate, tidal volume, and particle parameters; Expansion unit: calculates the expansion amplitude and airflow velocity of the alveoli based on the respiratory rate and tidal volume; Integration unit: inputting the airflow velocity, the expansion amplitude of the alveoli, and the particle parameters into the respiratory tract particle deposition model obtained by the above-mentioned method for constructing the respiratory tract particle deposition model based on rhythmic breathing, setting the model parameters, and obtaining the set respiratory tract particle deposition model; Prediction unit: The deposition rate and deposition distribution are predicted by setting up the respiratory particle deposition model.

[0104] The device for predicting the distribution of respiratory tract particle deposition provided by the embodiment of the present invention specifically includes: A memory and a processor; the memory is used to store program instructions; the processor is used to call the program instructions, and when the program instructions are executed, any one of the above-mentioned methods for predicting respiratory particle deposition distribution is implemented.

[0105] The disclosed embodiments of the present invention further provide a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, is any one of the above-mentioned methods for predicting respiratory tract particle deposition distribution.

[0106] The verification results of this verification embodiment show that assigning inherent weights to indications can improve the performance of the method relative to the default setting. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the coupling or direct coupling or communication connection between each other shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms. The unit described as a separate component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units. A person of ordinary skill in the art may understand that all or part of the steps in the various methods of the above-mentioned embodiments may be completed by instructing the relevant hardware through a program, and the program may be stored in a computer-readable storage medium, and the storage medium may include: a read-only memory (ROM), a random access memory (RAM), a disk or an optical disk, etc.

[0107] A person skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be implemented by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. The above-mentioned medium storage can be a read-only memory, a disk or an optical disk, etc.

[0108] The above is a detailed introduction to a computer device provided by the present invention. For a person skilled in the art, according to the concept of the embodiments of the present invention, there may be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A full-scale human respiratory rhythmic breathing three-dimensional geometric model and a construction method thereof, characterized in that: include: Acquire imaging data and anatomical data of the respiratory tract; Performing three-dimensional reconstruction of the first respiratory tract based on the image data to obtain a first respiratory tract geometric model; Performing three-dimensional reconstruction of the second respiratory tract based on the anatomical data to obtain a second respiratory tract geometric model; The first respiratory tract geometric model and the second respiratory tract geometric model are spliced ​​together to obtain a full-size respiratory tract geometric model.

2. The human body full-scale respiratory rhythmic breathing three-dimensional geometric model and its construction method according to claim 1, characterized in that: The process of the second airway three-dimensional reconstruction is: Obtain anatomical data, including airway dimensions, average alveolar dimensions, and respiratory tract geometric characteristics; Perform two-dimensional tessellation of alveoli based on the airway size and the average alveolar size to obtain two-dimensional tessellation data; The second respiratory tract geometric model is obtained by performing three-dimensional stretching on the two-dimensional dense paving data and the respiratory tract geometric features.

3. A method for constructing a respiratory particle deposition model based on rhythmic breathing, characterized in that: include: Obtain fluid flow pattern type, particle properties, respiratory tidal volume, and full-scale respiratory tract geometry model; The full-scale airway geometry model The human body full-scale respiratory rhythmic breathing three-dimensional geometric model and its construction method according to claim 1 or 2 are obtained; Determine the airflow physics field and particle tracking physics field according to the fluid flow type and particle properties; Constructing a geometric deformation physical field according to the respiratory tidal volume; The rhythmic breathing respiratory tract particle deposition model is obtained through the airflow flow physics field, particle tracking physics field, geometric deformation physics field, and full-scale respiratory tract geometry model.

4. The method for constructing a respiratory particle deposition model based on rhythmic breathing according to claim 3, characterized in that: The particle tracking physical field performs force analysis on fluid particles according to the airflow state of rhythmic breathing and the particle properties to obtain force results, and constructs a particle momentum equation based on the force results. The force parameters calculated by the particle momentum equation include one or more of the following: drag, gravity, Brownian force, pressure gradient force, virtual mass force, lift, and Bassett force.

5. The method for constructing a respiratory particle deposition model based on rhythmic breathing according to claim 3, characterized in that: The method also includes a particle collision and agglomeration physical field, which is determined according to the fluid flow type and particle properties, and a rhythmic breathing respiratory tract particle deposition model is obtained through the airflow flow physical field, particle tracking physical field, particle deposition and agglomeration physical field, geometric deformation physical field, and full-scale respiratory tract geometric model.

6. The method for constructing a respiratory particle deposition model based on rhythmic breathing according to claim 3, characterized in that: The respiratory deformation model of the geometric deformation physical field is obtained by calculating the expansion amplitude of the alveoli based on the respiratory tidal volume and constructing the respiratory tract deformation equation through the expansion amplitude, and then based on the respiratory deformation equation.

7. The method for constructing a respiratory particle deposition model based on rhythmic breathing according to claim 3, characterized in that: The respiratory tract particle deposition model also includes model correction, obtaining animal inhalation experiment result data, correcting the respiratory tract particle deposition model based on the animal inhalation experiment result, and obtaining a corrected respiratory tract particle deposition model; Alternatively, the model correction is replaced by: obtaining human inhalation experiment result data and animal inhalation experiment result data, and correcting the model based on the human inhalation experiment result data. G 0- G 6-segment respiratory tract particle deposition model, calibrated based on animal inhalation test results G 7- G 23 The segmental respiratory tract particle deposition model is used to obtain the calibrated respiratory tract particle deposition model.

8. A method for predicting respiratory tract particle deposition distribution, characterized in that: include: Obtain respiratory rate, tidal volume, and particle parameters; Calculate the expansion amplitude of the alveoli and the airflow velocity based on the tidal volume of the respiratory frequency; Inputting the airflow velocity, the expansion amplitude of the alveoli, and the particle parameters into the respiratory tract particle deposition model obtained by the method for constructing a respiratory tract particle deposition model based on rhythmic breathing according to any one of claims 3 to 7, setting the model parameters, and obtaining a set respiratory tract particle deposition model; The deposition rate and distribution are predicted by setting up the respiratory particle deposition model.

9. A computer program product having a computer program or instructions thereon, characterized in that: include: The computer program or instructions are executed by the processor to implement the full-size human respiratory rhythmic breathing three-dimensional geometric model and its construction method as described in any one of claims 1-2, or to implement the construction method of the respiratory particle deposition model based on rhythmic breathing as described in any one of claims 3-7, or to execute the prediction method of respiratory deposition distribution as described in claim 8.

10. A computer device comprising a memory, a processor and a computer program or instruction stored in the memory, characterized in that: The computer program or instructions are executed by the processor to implement the full-size human respiratory rhythmic breathing three-dimensional geometric model and its construction method as described in any one of claims 1-2, or to implement the construction method of the respiratory particle deposition model based on rhythmic breathing as described in any one of claims 3-7, or to implement the prediction method of respiratory deposition distribution as described in claim 8.

11. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: The computer program or instructions are executed by the processor to implement the full-size human respiratory rhythmic breathing three-dimensional geometric model and its construction method as described in any one of claims 1-2, or to implement the construction method of the respiratory particle deposition model based on rhythmic breathing as described in any one of claims 3-7, or to execute the prediction method of respiratory deposition distribution as described in claim 8.

Images