A method for simulating diabetic retinopathy process based on agent model and electronic device

Abstract

The present invention discloses a method and electronic device for simulating the process of diabetic retinopathy based on an agent model. The method comprises the following steps: discretizing a three-dimensional model of a patient's retinal blood vessels in a three-dimensional cube, and setting the obtained cube grid of a preset size as an agent of an environmental point; setting an agent of a basic unit containing an environmental point with a three-dimensional model of a retinal blood vessel, and configuring a row of basic units connected in a ring as a blood vessel unit; respectively initializing and setting the attributes of the basic unit, the attributes of the blood vessel unit, and the attributes of the environmental point; respectively updating the respective attributes of the basic unit, the environmental point, and the blood vessel unit according to corresponding behavior rules, and the shape of the three-dimensional model of the patient's retinal blood vessels changes in real time according to the updated attributes. The present invention not only simulates the effects of key events on retinal blood vessel lesions during retinal blood vessel lesions and reveals the dynamic changes of the disease, but also makes personalized predictions based on the characteristics of the patient.

Description

Technical Field

[0001] The present invention relates to the technical field of biological simulation, and in particular to a method for simulating a diabetic retinopathy process based on an agent model and an electronic device. Background Art

[0002] Diabetic retinopathy is one of the common complications of diabetic patients, and its pathogenesis involves a variety of complex biological processes. In diabetic patients, long-term hyperglycemia can damage retinal blood vessels, causing retinal microcirculation disorders and vascular lesions. One of the important manifestations is the abnormal proliferation of new blood vessels, known as proliferative diabetic retinopathy (PDR). This process is closely related to the molecular signaling pathways and cytokine regulation related to angiogenesis.

[0003] The simulation of diabetic retinopathy has important scientific significance and clinical application value. The realization of the simulation of diabetic retinopathy process helps to deeply understand the disease mechanism, optimize the treatment plan, discover new treatment methods, and provide support for personalized medicine and medical education.

[0004] However, traditional biological simulation methods have the following disadvantages when applied to diabetic retinopathy simulation:

[0005] (1) Traditional biological simulation methods use mathematical models that can only describe simple biological processes and cannot fully consider the complex characteristics of diseases;

[0006] (2) Traditional biological simulation methods use static models, which can only provide the disease status at a certain point in time and are difficult to reveal the dynamic changes of the disease;

[0007] (3) The mathematical models of traditional biological simulation methods can often only provide average concentration results, making it difficult to make personalized predictions based on the characteristics of different patients. Summary of the invention

[0008] In order to address the shortcomings of traditional biological simulation methods, such as difficulty in comprehensively considering the complex characteristics of the disease, difficulty in revealing the dynamic changes of the disease, and difficulty in making personalized predictions based on the characteristics of different patients, an embodiment of the present invention provides a method and electronic device for simulating the diabetic retinopathy process based on an agent model.

[0009] The specific technical solutions are as follows:

[0010] In a first aspect, an embodiment of the present invention provides a method for simulating a diabetic retinopathy process based on an agent model, comprising the following steps:

[0011] Constructing a three-dimensional discrete coordinate system, and constructing a three-dimensional cube in the three-dimensional discrete coordinate system for simulating a retinal vascular environment;

[0012] The definition content includes the basic unit of attributes, the vascular unit and the environmental point, and configures the simulated environment behavior rules for controlling the update of the attributes over time for the environmental point, configures the signal pathway antagonism behavior rules for controlling the basic unit attributes to change with the environmental point attributes for the basic unit, and configures the vascular three-dimensional model change behavior rules for controlling the vascular unit attributes to change with the basic unit attributes for the vascular unit; wherein the basic unit is defined as an agent representing a part of the epithelial cell; the vascular unit is defined as the smallest unit of the retinal vascular entity composed of the basic units; the environmental point is defined as an agent of the retinal environment entity where the retinal blood vessels to be simulated are located;

[0013] Performing cube discretization on the three-dimensional model of the patient's retinal blood vessels constructed in the three-dimensional cube, and setting the obtained cube grid of a preset size as the environmental point position;

[0014] Assuming that an environmental point with a three-dimensional model of retinal blood vessels accommodates one of the basic units, a row of the basic units connected in a ring is configured as a blood vessel unit, and the number of basic units contained in each blood vessel unit is recorded;

[0015] Initializing and setting the attributes of the basic unit, the attributes of the blood vessel unit and the attributes of the environmental point respectively; wherein the attributes of the environmental point include the blood glucose concentration of the patient;

[0016] The basic unit updates their respective attributes according to the signal pathway antagonism behavior rules, the environmental points according to the simulated environment behavior rules, and the vascular unit according to the vascular three-dimensional model change behavior rules, and the shape of the patient's retinal vascular three-dimensional model changes in real time according to the updated attributes.

[0017] Preferably, the basic unit definition includes the following:

[0018] The type of the basic unit is a proxy representing a physical part of a vascular epithelial cell;

[0019] The properties of the basic unit include VEGFR concentration, activated VEGFR concentration, effective VEGFR concentration in activated VEGFR, Dll4 protein concentration, notch concentration, activated notch concentration, effective notch concentration in activated notch, and basic state of the basic unit;

[0020] The vascular unit definition includes the following:

[0021] The type of the blood vessel unit, the smallest physical unit of the blood vessel composed of basic units;

[0022] The properties of the vascular unit include the number of basic units, the maximum number and the minimum number of basic units, VEGFR concentration, activated VEGFR concentration, effective VEGFR concentration in activated VEGFR, Dll4 protein concentration, notch concentration, activated notch concentration, and effective notch concentration in activated notch;

[0023] The environmental point definition includes the following contents:

[0024] The type of the environmental point is a proxy for the physical object of the retina where the retinal blood vessels to be simulated are located;

[0025] The attributes of the environmental points include oxygen concentration, VEGF concentration, and blood glucose concentration of the patient.

[0026] The basic status includes:

[0027] Stateless N: The basic unit that constitutes the main unit of blood vessels;

[0028] Base state B: connecting the stateless N basic unit and the stem state S basic unit of the pseudopod;

[0029] Stalk state S: a part of the pseudopod, a basic unit connecting the basic unit of the base state B and the basic unit of the tip state T;

[0030] Tip state T: The tip cell is responsible for selecting the extension direction. When the pseudopodia are just generated, there are only the tip state T basic units. After the pseudopodia are extended, they are composed of the stem state S basic units and the tip state T basic units.

[0031] Preferably, the simulated environmental behavior rules for updating the control attributes of the environmental point configuration over time include:

[0032] Configure the oxygen concentration control equation for the environmental point, the equation is as follows:

[0033]

[0034] in:

[0035] C o is the oxygen concentration at each environmental point in the retina at time t, and the initial value is obtained from the patient's most recent test result;

[0036] D o is the diffusion coefficient of oxygen in the retina, a fixed constant;

[0037] F(C o) is the oxygen consumption rate equation for each environmental point, specifically:

[0038]

[0039] A is the normal oxygen consumption rate coefficient at each environmental point, a fixed constant;

[0040] C lim The minimum oxygen concentration required for normal physiological activities of each basic unit is a fixed constant. The basic unit below this constant value is in an oxygen-deficient state.

[0041] G(ρ) is the amount of oxygen transported from the peripheral blood vessels at the current environmental point at time t, which is obtained based on the environmental point where each basic unit contained in the vascular unit is located. The calculation equation is as follows:

[0042]

[0043] α o is the oxygen transport rate constant in blood vessels;

[0044] G avg is the average blood glucose concentration during T3;

[0045] ρ(x, y, z, t) is the capillary density of the current environmental point at time t, which is obtained according to the number of basic units accommodated in the environmental points around the environmental point. For example, the current environmental point is the ratio of the number of basic units existing in a cube whose central side length is N times the side length of the cube grid to the total number of environmental points;

[0046] Configure the VEGF concentration control equation for the environmental point, the equation is as follows:

[0047]

[0048] in:

[0049] C v is the VEGF concentration at each environmental point at time t;

[0050] γ is the degradation rate of VEGF at the environmental site, a fixed constant;

[0051] λ is the VEGF production rate at the environmental site, a fixed constant;

[0052] D v is the VEGF diffusion constant at the environmental site;

[0053] S(C0) is the governing equation for the VEGF concentration generated by oxygen concentration, and the equation is as follows:

[0054]

[0055] Configure the control equation for blood glucose concentration for the environmental point and input according to the patient's specific situation:

[0056]

[0057] in:

[0058] C g is the blood glucose concentration at the current environmental point at time t;

[0059] C g (t) is the equation of blood glucose concentration variation over time at the environmental point, and the initial value is obtained by patient testing;

[0060] F(t) is the increase in blood glucose concentration over time after food intake;

[0061] I(t) is the equation of insulin concentration changing with time, and the initial value of insulin concentration is obtained from the patient's test;

[0062] H(t) is the equation for the change of glucagon concentration over time, and the initial value is obtained from the patient test;

[0063] k1 is the influence coefficient of insulin concentration on blood glucose concentration, which is a fixed constant;

[0064] k2 is the coefficient of influence of glucagon concentration on blood glucose concentration, which is a fixed constant;

[0065] L(t) is the equation for the change of blood glucose concentration consumed by liver effect over time.

[0066] Preferably, the signal pathway antagonistic behavior rules configured for the basic unit to control the basic unit properties changing with the environmental point properties include:

[0067] Configuring the signal pathway antagonistic behavior rules of VEGF binding to the basic unit VEGFR in the environmental point includes:

[0068] (1) VEGF in the environmental point binds to the basic unit VEGFR, VEGFR is activated, and after T1 time, Dll4 protein is generated in the current basic unit;

[0069] (2) Because Dll4 protein is produced in the basic unit, the concentration of Dll4 protein increases. After binding with notch in the adjacent basic unit, the concentration of Dll4 protein in the basic unit returns to the initial concentration after R1 time;

[0070] Configuring the signaling pathway antagonistic behavior rules of the Dll4 protein in the basic unit binding to the basic unit notch, including:

[0071] (1) The Dll4 protein in the basic unit binds to the notch in the adjacent basic unit, and the notch is activated. After T2 time, VEGFR is generated in the current basic unit; wherein the adjacent basic unit is an environmental point that accommodates the basic unit among the 26 environmental points connected to the basic unit;

[0072] (2) Because the basic unit generates VEGFR, the VEGFR concentration increases. After combining with VEGF in the 26 environmental points adjacent to the environmental point where the basic unit is located, the VEGFR concentration in the basic unit returns to the initial concentration after R2 time.

[0073] Preferably, the control equation of the concentration of VEGFR activated by VEGF in the environmental point in the basic unit is as follows:

[0074]

[0075] in:

[0076] V′ m is the concentration of VEGFR that can be activated by VEGF in the 26 adjacent environmental points in the current basic unit;

[0077] V sink is the concentration of VEGF remaining after binding to the receptor type of VEGFR in the non-basic unit and sedimentation, a fixed constant;

[0078] M tot is the number of basic units contained in the vascular unit to which the current basic unit belongs;

[0079] V max The maximum VEGFR concentration allowed in the basic unit of the vascular unit to which the current basic unit belongs is a fixed constant;

[0080] V m is the VEGFR concentration of the current basic unit, a fixed constant;

[0081] It is the sum of VEGF concentrations in 26 environmental points adjacent to the environmental point where the current basic unit is located;

[0082] The Notch concentration activated by the basic unit is the sum of the Notch concentration of the basic unit and the Dll4 protein concentration of all adjacent basic units. m The minimum value of the sum of the two is as follows:

[0083]

[0084] Among them, N′ m The concentration of Notch activated for the basic unit; N mis the notch concentration of the basic unit; D m is the Dll4 protein concentration in basic units.

[0085] Preferably, at time t, the vascular unit D114 protein concentration is and VEGFR concentration The control equations are as follows:

[0086]

[0087] Wherein, δ is the concentration of Dll4 protein generated by the binding of VEGFR and VEGF in the vascular unit, which is a constant; σ is the concentration of VEGFR generated by the binding of notch and Dll4 protein in the vascular unit, which is a constant; N″ m is the activated and effective notch concentration, the activated notch concentration N′ after T1 time step m ,N″ m and N′ m The value is the same, but it is obtained at time t+T1. Update; V″ m For activated and effective VEGFR, the activated VEGFR concentration V' after the T2 time step m , V″ m and V′ m The value is the same, but it is obtained at time t+T2. renew;

[0088] The concentration of Dll4 protein in vascular unit and VEGFR concentration Divide it evenly into the basic unit to which it belongs, and get the Dll4 protein concentration D of each basic unit m and notch concentration N m .

[0089] Preferably, configuring the vascular unit with a vascular three-dimensional model change behavior rule that controls the vascular unit attribute to change with the basic unit attribute includes:

[0090] The configuration of the behavior rules for simulating pseudopod state changes as the basic unit properties change is as follows:

[0091] At every time step, the actin concentration of the basic unit is calculated according to the concentration of VEGFR activated by VEGF, and the number of basic units constituting each pseudopodia is counted;

[0092] According to the actin concentration of the basic unit and the number of basic units constituting each pseudopod, the behavior rules for simulating the initial generation of pseudopods, the behavior rules for simulating the extension of pseudopods and the behavior rules for simulating the retraction of pseudopods are implemented to simulate the behavior rules of the basic state of the basic unit changing over time;

[0093] The configuration of the behavioral rules that simulate the evolution of pseudopods into vascular units is as follows:

[0094] If the number of basic units constituting pseudopods reaches the elongation limit of pseudopods E max , and the number of time steps in which the basic unit of the pseudopod is not extended reaches the preset number f max Before, the accumulated actin concentration of the basic unit of the pseudopod in the tip state T reaches the minimum actin concentration required to generate vascular units equal to the number of basic units constituting the pseudopod, and vascular units are generated respectively with each basic unit of the current pseudopod as the center, and the basic state of the basic unit generating the vascular unit is set to the stateless N;

[0095] Among them, the number of vascular units generated is equal to the number of basic units constituting pseudopodia, the number of basic units contained in each vascular unit generated is the minimum number of basic units contained in the vascular unit; the minimum actin concentration required to generate vascular units equal to the number of basic units constituting pseudopodia is equal to:

[0096] The minimum number of basic units contained in a vascular unit is multiplied by the number of basic units that make up a pseudopodia and the actin concentration required for extension. ext .

[0097] Preferably, the governing equation for the actin concentration p in the basic unit is as follows:

[0098] p=CV′ m M tot / V max

[0099] Among them, C is the signal pathway activation intensity constant representing the binding of VEGF to the basic unit VEGFR in the environmental point; V' m M is the concentration of VEGFR activated by VEGF in the basic unit; tot V is the number of basic units contained in the vascular unit to which the current basic unit belongs; max The maximum allowed VEGFR concentration of the vascular unit to which the current basic unit belongs, which is a fixed constant;

[0100] The actin generated by the stateless N basic unit is evenly distributed to the base state B basic unit of the pseudopodia within a radius of 20 microns from itself, and the actin is transferred to the tip state T basic unit of the pseudopodia through the stem state S basic unit of the pseudopodia. The amount of actin distributed to the corresponding pseudopodia by the tip state T basic unit of the pseudopodia increases.

[0101] Preferably, the configuration simulates the behavioral rules for initial generation of pseudopods, including:

[0102] If the actin concentration p of the stateless N basic units constituting the main blood vessel reaches the actin concentration value t required for the extension of pseudopods ext When , the basic state of the basic unit changes from the non-state N to the base state B, and the environmental point with the highest VEGF concentration among all the environmental points adjacent to the basic unit that do not accommodate the basic unit is set to accommodate one basic unit, and the basic state is the sharp state T;

[0103] If the VEGF concentrations of all the environmental points adjacent to the basic unit that do not accommodate the basic unit are the same, then a random environmental point is selected to be set as a basic unit that accommodates a sharp state T;

[0104] The configuration simulates the behavioral rules of pseudopod extension, including:

[0105] If the actin concentration p of the basic unit of the pseudopod in the pointed state T reaches the actin concentration value t required for the extension of the pseudopod ext , and the number of basic units constituting pseudopods has not reached the extension limit E max , the number of time steps during which the tip state T basic unit of the pseudopod is not extended is cleared, and a stem state S basic unit is added between the base state B basic unit and the tip state T basic unit of the current pseudopod; otherwise, the number of time steps during which the tip state T basic unit of the pseudopod is not extended is increased by 1; wherein, the adding of a stem state S basic unit between the base state B basic unit and the tip state T basic unit of the current pseudopod is specifically as follows:

[0106] Find the environmental point with the highest VEGF concentration among all environmental points adjacent to the pointed state T basic unit that do not accommodate the stateless N basic unit and the base state B basic unit, and use the environmental point as the extension direction; if the VEGF concentrations of the environmental points of all the stateless N basic units and the base state B basic units adjacent to the pointed state T basic unit are the same, then select one environmental point as the extension direction of the pseudopod;

[0107] Determine whether the environmental point of the found environmental point can accommodate the stem state S basic unit or the tip state T basic unit;

[0108] If the found environmental point does not accommodate, the found environmental point is set to accommodate a pointed state T basic unit, and the environmental point where the pseudopod's pointed state T basic unit is located is set to accommodate the stem state S basic unit; wherein the initial values ​​of the VEGFR concentration and Dll4 protein concentration of the pointed state T basic unit of the found environmental point are the pseudopod's pointed state T basic unit, and the initial value of the actin concentration is the pseudopod's pointed state T basic unit. The actin concentration of the pseudopod's pointed state T basic unit decreases by one actin concentration value t required for extension. ext ;

[0109] If the found environmental point accommodates the stem state S basic unit or the tip state T basic unit, the basic state of the tip state T basic unit of the pseudopod is changed to the stem state S; the actin of the tip state T basic unit of the pseudopod is transferred to the tip state T basic unit of the pseudopod to which the found environmental point accommodates the stem state S basic unit or the tip state T basic unit belongs;

[0110] The configuration simulates the behavior rules of pseudopod retreat, including:

[0111] If the number of time steps in which the pseudopod tip state T basic unit is not extended is greater than or equal to the preset number f max , and the actin concentration p of the pseudopod in the tip state T basic unit does not reach the actin concentration value t required for extension ext , delete a basic unit contained in pseudopods; wherein, delete a basic unit contained in pseudopods, as follows:

[0112] Divide all the actin in the T basic unit of the pseudopod tip equally to all the basic units connected to the T basic unit of the pseudopod tip;

[0113] Determine whether the basic state of the basic unit connected to the tip state T basic unit of the pseudopodia is the base state B; if it is the base state B, delete the tip state T basic unit of the pseudopodia, and change the basic state of the base state B basic unit connected to the tip state T basic unit of the pseudopodia to the stateless N; if it is not the base state B, that is, the stem state S, delete the tip state T basic unit of the pseudopodia, and change the basic state of the stem state S basic unit connected to the tip state T basic unit of the pseudopodia to the tip state T.

[0114] In a second aspect, an embodiment of the present invention further provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the method for simulating a diabetic retinopathy process based on an agent model as described in the first aspect is implemented.

[0115] The embodiment of the present invention provides a method and electronic device for simulating diabetic retinopathy process based on an agent model, which has the following effective technical effects:

[0116] (1) Compared with traditional mathematical models that can only describe simple biological processes and cannot fully consider the complex characteristics of the disease, the agent-based diabetic retinopathy model can simulate key events in the biological process, such as the effects of blood sugar, oxygen diffusion, angiogenesis, and the effects of growth factors, and is therefore more biologically reasonable;

[0117] (2) Compared with traditional static models that can only provide the disease status at a certain point in time and are unable to reveal the dynamic changes of the disease, the agent-based diabetic retinopathy model can simulate the dynamic development process of diabetic retinopathy over time, thereby better understanding the pathogenesis and progression of the disease;

[0118] (3) Compared with traditional models that often only provide average results and are difficult to make personalized predictions based on the characteristics of different patients, the agent-based diabetic retinopathy model can perform personalized simulations and predictions based on individual differences of patients, thereby providing clinicians with personalized treatment plans and decision support;

[0119] (4) Compared with the traditional drug development process, which often requires a lot of time and resources, the agent-based diabetic retinopathy model can be used for virtual drug screening and evaluation to discover potential therapeutic targets and drugs, thereby accelerating the process of new drug development and clinical transformation;

[0120] (5) Compared with traditional experiments that often require a large amount of experiments and experience accumulation, the agent-based diabetic retinopathy model can provide theoretical guidance for experimental design and data interpretation, guide experimental design and data collection, and thus better verify the accuracy and reliability of the model;

[0121] In summary, compared with traditional methods, the agent-based model for building diabetic retinopathy models has higher biological plausibility, dynamic simulation capabilities, personalized medical applications, drug screening and evaluation capabilities, as well as advantages in theoretical guidance and experimental design. These advantages make the agent-based model a powerful tool for studying and treating diabetic retinopathy.

[0122] Other features and advantages of the present invention will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the present invention.

[0123] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0124] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0125] Figure 1 A schematic diagram of a flow chart of a diabetic retinopathy process simulation based on an agent model provided in an embodiment of the present invention;

[0126] Figure 2 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0127] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0128] An agent-based model (ABM) is a computational model used to simulate a system consisting of multiple independent individuals (agents) that behave according to certain rules and interactions, and generate the dynamic behavior and evolution of the overall system. Agent-based models are a powerful modeling tool that can handle diversity, heterogeneity, and nonlinear interactions in complex systems. It provides an effective way to simulate complex phenomena and system behaviors, helping researchers understand and predict the complexity in the real world. The following are some important concepts and features of agent-based models:

[0129] 1. Agents:

[0130] Agents are the basic units in agent-based models, which can be individuals, organizations, robots, molecules, etc. Agents have autonomy and local perception and decision-making capabilities, and interact and behave according to their internal state and external environment.

[0131] 2. Space and Environment:

[0132] Agent-based models usually involve the movement and interaction of agents in a space or environment. The space can be a two-dimensional or three-dimensional grid space, or an abstract state space. The environment can include static or dynamic elements that affect agent behavior and evolution.

[0133] 3. Interaction and Communication:

[0134] Agents exchange information and interact with each other through defined interaction methods and communication protocols. These interactions can be direct physical contact or virtual interactions achieved through information transmission and network connections.

[0135] 4. Behavior Rules:

[0136] Each agent performs actions according to the set behavioral rules and decision-making mechanisms. The behavioral rules can be based on simple if-then logic or complex decision-making algorithms.

[0137] 5. Self-organization and Emergence:

[0138] Agent-based models induce system-level self-organization and new global behaviors through local interactions and behavioral rules between agents. This self-organization and evolutionary nature is one of the important features of agent-based models.

[0139] In diabetic patients, hyperglycemia leads to damage to retinal microvessels and increased hypoxia, which triggers the following series of biological responses:

[0140] High blood sugar triggers inflammation and cell damage: Long-term high blood sugar can induce abnormal metabolic products in cells, triggering an inflammatory response, leading to retinal tissue cell damage, blood vessel blockage and contraction, and affecting blood supply;

[0141] Exacerbated hypoxia: As high blood sugar affects the normal function of microvessels, retinal tissue hypoxia is further aggravated;

[0142] Release of angiogenic factors: Hypoxia stimulates retinal tissue to release angiogenic factors, such as vascular endothelial growth factor (VEGF), which promotes the growth of new blood vessels;

[0143] Increased angiogenesis: Under the action of angiogenic factors, the retinal tissue begins to abnormally proliferate new blood vessels. These new blood vessels are unstable and prone to rupture and bleeding, seriously threatening vision.

[0144] In summary, in order to simulate the complex phenomena and system behaviors in the process of diabetic retinopathy, it helps researchers understand and predict the complexity of the process of diabetic retinopathy, such as Figure 1 As shown, the embodiment of the present application provides a method for simulating diabetic retinopathy process based on an agent model, the method comprising the following steps:

[0145] Step S100: construct a three-dimensional discrete coordinate system (x, y, z), and construct a three-dimensional cube in the three-dimensional discrete coordinate system for simulating a retinal vascular environment.

[0146] Step S200, define the basic unit MemAgent, vascular unit EcAgent and environmental points including attributes, configure the simulated environment behavior rules for the environmental points to control the update of the attributes over time, configure the signal pathway antagonism behavior rules for the basic units to control the changes of the basic unit attributes with the environmental point attributes, and configure the vascular three-dimensional model change behavior rules for the vascular units to control the changes of the vascular unit attributes with the basic unit attributes.

[0147] Among them, the basic unit is defined as an agent representing the physical object of the epithelial cell part; the vascular unit is defined as the smallest physical unit of the retinal vascular composed of the basic units; the environmental point is defined as the agent of the physical object of the retina where the retinal blood vessels to be simulated are located, that is, the agent of the simulated environment; the definition content of the basic unit MemAgent, the vascular unit EcAgent and the environmental point all include attributes, which are the parameters that can affect the concentration of vascular endothelial growth factor and vascular endothelial growth factor receptor or show the growth of blood vessels in the corresponding entities of the basic unit MemAgent, the vascular unit EcAgent and the environmental point. The basic unit, the environmental unit and the vascular unit update the value, state or 3D model of the attribute according to the configured behavior rules.

[0148] In this embodiment, the basic unit MemAgent definition includes but is not limited to the following:

[0149] The basic unit MemAgent type represents the agent of the vascular epithelial cell part;

[0150] The properties of the basic unit MemAgent include VEGFR (vascular endothelial growth factor receptor) concentration, activated VEGFR concentration, effective VEGFR concentration in activated VEGFR, Dll4 protein concentration, notch concentration, activated notch concentration, effective notch concentration in activated notch, and the basic state of the basic unit MemAgent. The basic state of the basic unit MemAgent includes:

[0151] Stateless N: The basic unit MemAgent, which constitutes the main unit of blood vessels;

[0152] Base state B: connects the stateless N basic unit MemAgent with the stalk state S basic unit MemAgent of the pseudopod;

[0153] Stalk state S: a part of the pseudopod, connecting the basic unit MemAgent of the base state B and the basic unit MemAgent of the tip state T (tip cell);

[0154] Tip state T: The tip cell is responsible for selecting the extension direction. When the pseudopod is just generated, the initial state only has the tip state T basic unit. After the pseudopod extends forward, it is composed of the stem state S basic unit MemAgent and the tip state T basic unit.

[0155] The definition of vascular unit EcAgent includes but is not limited to the following:

[0156] Vascular unit type, the smallest physical unit of a vascular vessel composed of basic units;

[0157] The properties of the vascular unit EcAgent include the number of basic unit MemAgents (the initial value is obtained according to the discretized retinal vascular model), the maximum and minimum numbers of basic unit MemAgents, VEGFR concentration, activated VEGFR concentration, effective VEGFR concentration in activated VEGFR, Dll4 protein concentration, notch concentration, activated notch concentration, and effective notch concentration in activated notch.

[0158] The definition of environmental points includes but is not limited to the following:

[0159] The type of environmental point is the proxy of the physical object in the retina where the retinal blood vessels to be simulated are located;

[0160] Environmental point attributes, including oxygen concentration, VEGF (vascular endothelial growth factor) concentration, and the patient's blood glucose concentration.

[0161] Step S300, start simulation for a certain patient, perform cube discretization on the three-dimensional model of the patient's retinal blood vessels constructed in a three-dimensional cube, and set the obtained cube grid of a preset size as an environmental point.

[0162] In the embodiment of the present application, step S300 includes the following steps:

[0163] A three-dimensional model of the patient's retinal blood vessels is constructed in a three-dimensional cube; wherein the preset three-dimensional retinal blood vessel model can be directly called as the three-dimensional retinal blood vessel model of the current patient (i.e., a three-dimensional model approximating the distal end of the retinal blood vessels, where the distal end refers to the growth end). In order to be close to the patient's current retinal blood vessel state, the patient's retinal blood vessel three-dimensional model can also be constructed based on the patient's current retinal blood vessel image.

[0164] The three-dimensional cube containing the three-dimensional model of retinal blood vessels is discretized into cube grids of a preset size (such as a 0.5-micron cube grid), and the obtained cube grid of the preset size is set as the environmental point. At this time, the three-dimensional model of retinal blood vessels wrapped in the three-dimensional cube is also discretized into 0.5-micron cube grids.

[0165] Step S400: Assume that an environmental point with a three-dimensional model of retinal blood vessels accommodates a basic unit MemAgent, configure a row of basic units MemAgent connected in a ring as a blood vessel unit EcAgent, and record the number of basic units contained in each blood vessel unit.

[0166] Step S500, respectively initialize and set the properties of the basic unit MemAgent, the properties of the vascular unit EcAgent and the properties of the environmental points; wherein, the properties of the environmental points include blood glucose concentration, which is set according to the patient's examination results in order to be closer to the actual situation of the patient and ensure that the simulation results are closer to the actual development; the basic state of the basic unit is initialized to stateless N.

[0167] In step S600, the basic units update their respective properties according to the signal pathway antagonism behavior rules, the environmental points according to the simulated environment behavior rules, and the vascular units according to the vascular three-dimensional model change behavior rules. The shape of the patient's retinal vascular three-dimensional model changes in real time according to the updated properties, that is, the property changes of the vascular units will be reflected in the shape changes of the retinal vascular three-dimensional model, thereby realizing the simulation of the retinopathy process of diabetic patients.

[0168] The embodiment of the present application not only comprehensively considers the complex characteristics of the disease, but also simulates the effects of key events in the process of retinal vascular lesions (such as blood sugar influence, oxygen diffusion, angiogenesis and growth factors) on retinal vascular lesions by respectively representing the physical epithelial cell part and the physical environment of the retina where the blood vessels to be simulated are located. Therefore, it is more biologically reasonable, and can also dynamically simulate the real-time development and changes of diabetic retinopathy, so as to better understand the pathogenesis and progression of the disease. In addition, the initial value of the attribute to be simulated in this embodiment is based on the patient's own situation, especially the environmental point agent of the simulated environment, and personalized simulation and prediction are performed according to the individual differences of the patient, effectively ensuring that the simulation results are the same as the actual lesion results.

[0169] In the embodiment of the present application, since the attributes of the environmental point include oxygen concentration, VEGF concentration and blood glucose concentration, the simulated environmental behavior rules for controlling the attributes to be updated over time for the environmental point configuration (i.e., the configuration of the environmental point state control equation) include:

[0170] (1) Configure the oxygen concentration control equation for each environmental point. The oxygen concentration control equation can simulate the transportation and consumption of oxygen in each environmental point. The equation is as follows:

[0171]

[0172] in:

[0173] C ois the oxygen concentration at each environmental point in the retina at time t, and the initial value is obtained from the patient's most recent test result;

[0174] D o is the diffusion coefficient of oxygen in the retina, a fixed constant;

[0175] F(C o ) is the oxygen consumption rate equation for each environmental point, specifically:

[0176]

[0177] A is the normal oxygen consumption rate coefficient at each environmental point, a fixed constant;

[0178] C lim The minimum oxygen concentration required for normal physiological activities of each basic unit is a fixed constant. The basic unit below this constant value is in an oxygen-deficient state.

[0179] G(ρ) is the amount of oxygen transported from the surrounding blood vessels at the current environmental point at time t. Vascular transport is the source of oxygen at the environmental point. It is obtained based on the environmental point where each basic unit MemAgent contained in the vascular unit EcAgent is located. The average blood glucose concentration of the current vascular unit (within the T3 time period) is not higher than G within the time T3. lim When the average blood sugar concentration is higher than G lim Afterwards, inflammatory reactions and oxidative stress will occur in the blood vessels, causing the blood vessels to shrink and become blocked, affecting the delivery and utilization of oxygen. Therefore, the calculation equation of G(ρ) is as follows:

[0180]

[0181] α o is the oxygen transport rate constant in blood vessels;

[0182] G avg is the average blood glucose concentration during T3;

[0183] ρ(x, y, z, t) is the capillary density of the current environmental point at time t, which is obtained according to the number of basic units accommodated in the environmental points around the environmental point. For example, the current environmental point is the ratio of the number of basic unit MemAgents existing in a cube with a central side length of N times the side length of the cube grid to the total number of environmental points (the cube grid is 0.5 microns, N is 5, the cube has a side length of 2.5 microns, and the total number of environmental points is 125).

[0184] (2) The VEGF concentration control equation for the environmental point is configured as follows:

[0185]

[0186] in:

[0187] C v is the VEGF concentration at each environmental point at time t;

[0188] γ is the degradation rate of VEGF at the environmental site, a fixed constant;

[0189] λ is the VEGF production rate at the environmental site, a fixed constant;

[0190] D v is the VEGF diffusion constant at the environmental site;

[0191] S(C0) is the control equation for the VEGF concentration generated by oxygen concentration. If the oxygen concentration at the environmental point is C lim There is no VEGF generated due to hypoxia. If the oxygen concentration at the environmental point is less than C lim VEGF will be produced due to lack of oxygen, so the equation is as follows:

[0192]

[0193] (3) Configure the control equation for blood glucose concentration at the environmental point and input it according to the patient's specific situation:

[0194]

[0195] in:

[0196] C g is the blood glucose concentration at the current environmental point at time t;

[0197] C g (t) is the equation of blood glucose concentration variation over time at the environmental point, and the initial value is obtained by patient testing;

[0198] F(t) is the increase in blood glucose concentration over time after food intake;

[0199] I(t) is the equation of insulin concentration changing with time, and the initial value of insulin concentration is obtained from the patient's test;

[0200] H(t) is the equation for the change of glucagon concentration over time, and the initial value is obtained from the patient test;

[0201] k1 is the influence coefficient of insulin concentration on blood glucose concentration, which is a fixed constant;

[0202] k2 is the coefficient of influence of glucagon concentration on blood glucose concentration, which is a fixed constant;

[0203] L(t) is the equation for the change of blood glucose concentration consumed by liver effect over time.

[0204] In the embodiment of the present application, configuring a signal pathway antagonism behavior rule for a basic unit to control the change of basic unit attributes with environmental point attributes includes configuring a simulation of the signal pathway antagonism between the basic unit and the environmental point and configuring a simulation of the signal pathway antagonism between the basic units, which are as follows:

[0205] (A) Configure the signaling pathway antagonistic behavior rules of VEGF binding to the basic unit VEGFR in environmental sites.

[0206] (1) VEGF in the environmental point binds to VEGFR in the current basic unit, VEGFR is activated, and after T1 time, Dll4 protein is produced in the current basic unit.

[0207] In this embodiment, the control equation for the concentration of VEGFR activated by VEGF in the environmental point in each basic unit is as follows:

[0208]

[0209] in:

[0210] V′ m is the concentration of VEGFR that can be activated by VEGF in the 26 adjacent environmental points in the current basic unit;

[0211] V sink The concentration of VEGF remaining after binding to other receptors (receptor types other than VEGFR in the basic unit) is a fixed constant;

[0212] M tot The number of basic unit MemAgents contained in the vascular unit ECAgent to which the current basic unit MemAgent belongs;

[0213] V max The maximum value of VEGFR concentration allowed in the basic unit MemAgent contained in the vascular unit ECAgent to which the current basic unit MemAgent belongs is a fixed constant;

[0214] V m is the VEGFR concentration of the current basic unit, a fixed constant;

[0215] It is the sum of the VEGF concentrations in the 26 environmental points adjacent to the environmental point where the current basic unit is located.

[0216] (2) Because the current basic unit produces Dll4 protein, the Dll4 protein concentration in the current basic unit increases. After binding with notch in the adjacent basic unit, the Dll4 protein concentration in the basic unit returns to the initial concentration after R1 time.

[0217] (ii) Configure the signaling pathway antagonistic behavior rules of the Dll4 protein in the basic unit binding to the basic unit notch.

[0218] (1) The Dll4 protein in the current basic unit binds to the notch in the adjacent basic unit (there are 26 environmental points adjacent to the current basic unit. The adjacent basic unit here refers to the environmental point that accommodates the basic unit among these 26 adjacent basic units, which may not belong to the vascular unit to which the current basic unit belongs). Notch is activated, and after T2 time, VEGFR is generated in the current basic unit.

[0219] (2) Because VEGFR is generated in the current basic unit, the VEGFR concentration of the current basic unit increases. After combining with VEGF in the 26 environmental points adjacent to the environmental point where the current basic unit is located, the VEGFR concentration in the basic unit returns to the initial concentration after R2 time.

[0220] In this model, the dynamic changes of Dll4 protein in the basic unit are mainly studied. It is assumed that the notch concentration in the basic unit is a constant, that is, the generation and consumption of notch in the basic unit are balanced. Notch in the basic unit is activated by Dll4 protein in the adjacent basic unit, that is, notch in the basic unit is activated by binding to Dll4 protein in the adjacent basic unit that does not belong to the same basic unit. The concentration of notch activated in each basic unit is the sum of the notch concentration in the basic unit and the Dll4 protein concentration of all adjacent basic units. m The minimum value of the sum of the two is as follows:

[0221]

[0222] Among them, N′ m The concentration of Notch activated for the basic unit; N m is the notch concentration of the basic unit; D m is the Dll4 protein concentration in basic units.

[0223] VEGFR recovers its initial concentration after R2 time, and the increase of VEGFR inhibits VEGFR activity; in a vascular unit EcAgent, after a delay of time T1 and time T2, a new Dll4 protein concentration and VEGFR concentration (i.e., the sum of the Dll4 protein concentration of the basic units contained in the vascular unit EcAgent and the sum of the VEGFR concentration of the basic units contained) are obtained. At time t, the Dll4 protein concentration of the vascular unit EcAgent is and VEGFR concentration The control equations are as follows:

[0224]

[0225] Wherein, δ is the concentration of Dll4 protein generated by the binding of VEGFR and VEGF in the vascular unit EcAgent, which is a constant; σ is the concentration of VEGFR generated by the binding of notch and Dll4 protein in the vascular unit EcAgent, which is a constant; N″ m is the activated and effective notch concentration, the activated notch concentration N′ after T1 time step m , that is, N″ m and N′ m The value is the same, but it is not obtained until t+T1. Update; V″ m For activated and effective VEGFR, the activated VEGFR concentration V' after the T2 time step m , that is, V″ m and V′ m The value is the same, but it is not obtained until t+T2. renew.

[0226] Dll4 protein concentration and VEGFR concentration is the total concentration of EcAgent in the vascular unit, and the Dll4 protein concentration is and VEGFR concentration Averagely distribute the MemAgent to the basic unit to which it belongs, and obtain the Dll4 protein concentration D of each basic unit m and notch concentration N m .

[0227] In the embodiment of the present application, the shape of the retinal blood vessel three-dimensional model changes according to the change of the blood vessel unit attribute, and the change of the blood vessel unit attribute changes according to the change of the basic unit attribute. The related attributes of the basic unit attribute that can cause the change of the blood vessel shape are mainly manifested as the basic state of the basic unit. The basic state change of the basic unit MemAgent is actually the state change of whether the basic unit MemAgent that mainly constitutes the blood vessel has pseudopodia, and the state change of pseudopodia includes the initial generation of pseudopodia, the extension of pseudopodia, and the retreat of pseudopodia. Therefore, the blood vessel three-dimensional model change behavior rules that control the change of blood vessel unit attributes with the basic unit attributes are configured for the blood vessel unit, including:

[0228] (A) Configuration of behavioral rules to simulate pseudopod state changes as basic unit properties change.

[0229] In the embodiment of the present application, the configuration of the behavior rule of simulating the state change of pseudopods with the change of basic unit properties includes:

[0230] At every time step, the actin concentration of the basic unit is calculated according to the concentration of VEGFR activated by VEGF, and the number of basic unit MemAgents constituting each pseudopodia is counted.

[0231] The increase in actin concentration is related to the VEGFR concentration, so the governing equation for actin concentration p in the basic unit MemAgent is as follows:

[0232] p=CV′ m M tot / V max

[0233] Among them, C is the signal pathway activation intensity constant representing the binding of VEGF to the basic unit VEGFR in the environmental point; V′ m is the concentration of VEGFR activated by VEGF in the basic unit MemAgent; M tot V is the number of basic units contained in the vascular unit ECAgent to which the current basic unit MemAgent belongs; max The maximum allowed VEGFR concentration of the basic unit MemAgent contained in the vascular unit ECAgent to which the current basic unit MemAgent belongs is a fixed constant.

[0234] In the embodiment of the present application, each basic unit produces actin regardless of the basic state, and the actin concentration in vascular cells tends to the pseudopodia that have already grown. When a basic unit MemAgent without state N obtains actin, it is not used, but all is evenly distributed to the basic units of base state B within a radius of 20 microns from itself, and then the actin is transferred to the corresponding basic unit of the stem state S of the pseudopodia through the basic unit of base state B, and finally to the basic unit of the tip state T of the pseudopodia. Here, the transfer is represented in the simulation program as the actin value of the basic unit of the tip state T of the pseudopodia plus the actin distributed to the pseudopodia without state N.

[0235] According to the actin concentration of the basic unit and the number of basic unit MemAgents that constitute each pseudopod, the behavior rules for simulating the initial generation of pseudopods, the behavior rules for simulating the extension of pseudopods, and the behavior rules for simulating the retraction of pseudopods are configured to implement the behavior rules for simulating the basic state of the basic unit MemAgent changing over time. Among them:

[0236] (1) Configure behavioral rules to simulate the initial generation of pseudopods.

[0237] If the actin concentration p of the stateless N basic unit MemAgent constituting the main vessel reaches the actin concentration value t required for the extension of pseudopods ext(a fixed constant), the basic state of the basic unit MemAgent changes from no state N to base state B, and the environmental point with the highest VEGF concentration among all the environmental points adjacent to the basic unit that do not accommodate basic units is set to accommodate one basic unit, and the basic state is the pointed state T (because the pseudopod has only the tip cell when it is initially generated, that is, only the basic unit MemAgent with the pointed state T). If the VEGF concentration of all the environmental points adjacent to the basic unit that do not accommodate basic units is the same, then a random environmental point is selected from them to be set as a basic unit that accommodates a pointed state T. Among them, the actin, Dll4, and VEGFR of the pointed state T basic unit of the initially generated pseudopod are all set to the initial values.

[0238] (2) Configure behavioral rules to simulate pseudopod extension.

[0239] The basic unit MemAgent in the tip state T uses actin to extend pseudopods. In each time step, the pseudopods only extend one basic unit in the stem state S. For each basic unit in the stem state S, the actin concentration of the basic unit MemAgent in the tip state T is reduced by one actin concentration value t required for extension. ext So the behavior rules for simulating pseudopod extension are configured as:

[0240] If the pseudopod is in the pointed state T, the actin concentration p of the basic unit MemAgent reaches the actin concentration value t required for pseudopod extension ext , and the number of basic units MemAgent constituting pseudopods (the sum of the number of basic units in the tip state T and the number of basic units in the stem state S of pseudopods) has not reached the extension limit E max (a preset fixed constant, when the number of basic unit MemAgents constituting the pseudopodia reaches the extension limit, there is no extension capability), the pseudopodia extends, the number of time steps during which the tip state T basic unit of the pseudopodia has not been extended is reset to zero, and a stem state S basic unit is added between the base state B basic unit and the tip state T basic unit of the current pseudopodia; otherwise, the pseudopodia does not extend, and the number of time steps during which the tip state T basic unit of the pseudopodia has not been extended is increased by 1.

[0241] In the embodiment of the present application, a stem state S basic unit is added between the base state B basic unit and the tip state T basic unit of the current pseudopod, and the specific steps are as follows:

[0242] First, determine the extension direction, and find the environmental point with the highest VEGF concentration among all the environmental points adjacent to the pointed state T basic unit MemAgent that do not accommodate the stateless N basic unit and the base state B basic unit, and use this environmental point as the extension direction; if the VEGF concentrations of the environmental points of all the stateless N basic units and the base state B basic units adjacent to the pointed state T basic unit MemAgent are the same, then randomly select one environmental point among them as the extension direction of the pseudopod;

[0243] Determine whether the environmental point of the found environmental point can accommodate the stem state S basic unit or the tip state T basic unit;

[0244] If the found environmental point does not accommodate, the environmental point is set to accommodate a pointed state T basic unit as the pointed state T basic unit after the pseudopod is extended, and the basic state of the pointed state T basic unit MemAgent before the pseudopod is extended is changed to the stem state S; wherein the VEGFR concentration and Dll4 protein concentration values ​​of the pointed state T basic unit MemAgent before the pseudopod is extended are the pointed state T basic unit MemAgent before the extension, and the actin concentration value of the pointed state T basic unit MemAgent before the extension is the actin concentration value t that is reduced by one actin concentration required for extension. ext The actin concentration, Dll4 protein concentration, and VEGFR concentration of the stem state S basic unit obtained by changing the basic state of the pre-extended tip state T basic unit are set as initial values.

[0245] If the found environmental point contains a stem state S basic unit or a tip state T basic unit, it means that the current pseudopod encounters other pseudopods in the extension direction. At this time, the current pseudopod fuses with other pseudopods, that is, the basic state of the tip state T basic unit before the pseudopod extends is changed to the stem state S, and it shares a tip cell (tip state T basic unit) with the encountered pseudopod, and the tip state T basic unit of the encountered other pseudopods is the tip state T basic unit after the pseudopod extends; and the actin of the tip state T basic unit before the pseudopod extends is transferred to the tip state T basic unit after the pseudopod extends. The actin of the tip state T basic unit after the pseudopod extends is the sum of the actin of the tip state T basic unit after the pseudopod extends and the tip state T basic unit of the encountered other pseudopods. The concentrations of Dll4 protein and VEGFR remain unchanged and are also the values ​​of the tip state T basic units of the encountered other pseudopods. The actin concentration, Dll4 protein concentration, and VEGFR concentration of the stem state S basic unit obtained by changing the basic state of the tip state T basic unit before extension are set as the initial values.

[0246] (3) Configure behavioral rules to simulate pseudopod retraction.

[0247] If the number of time steps in which the pseudopod tip state T basic unit is not extended is greater than or equal to the preset number fmax , that is, the basic unit MemAgent of the sharp state T is in the preset number f max The pseudopod fails to extend within time steps, and the actin concentration p of the basic unit in the pointed state T does not reach the actin concentration value t required for extension. ext , the pseudopod will shrink and delete a basic unit contained in the pseudopod.

[0248] The embodiment of the present application deletes a basic unit contained in a pseudopod, including the following steps:

[0249] All the actin in the T basic unit of the pseudopod in the pointed state was equally divided into all the basic units connected to the T basic unit of the pseudopod in the pointed state, and the actin in the T basic unit of the pseudopod in the pointed state was reversely transferred, while the Dll4 protein concentration and VEGFR concentration were not transferred.

[0250] Determine whether the basic state of the basic unit connected to the tip state T basic unit of the pseudopodia is the base state B; if it is the base state B, delete the tip state T basic unit of the pseudopodia, and change the basic state of the base state B basic unit connected to the tip state T basic unit of the pseudopodia to the stateless N; if it is not the base state B, that is, the stem state S, delete the tip state T basic unit of the pseudopodia, and change the basic state of the stem state S basic unit connected to the tip state T basic unit of the pseudopodia to the tip state T.

[0251] (B) Configuration of behavioral rules that mimic the evolution of pseudopods into vascular units.

[0252] In the embodiment of the present application, the configuration of the behavior rules for simulating pseudopods evolving into vascular units includes:

[0253] If the number of MemAgents constituting the basic unit of pseudopods reaches the extension limit of pseudopods E max , and the number of time steps in which the basic unit of the pseudopod is not extended reaches the preset number f max Before, the accumulated actin concentration of the basic unit in the tip state T of the pseudopod reaches the minimum actin concentration required to generate vascular units equal to the number of basic unit MemAgents constituting the pseudopod, and vascular units are generated with each basic unit of the current pseudopod as the center, and the pseudopod disappears; among which, the number of vascular units generated is equal to the number of basic unit MemAgents constituting the pseudopod, and the number of basic units contained in each generated vascular unit is the minimum number of basic units contained in the vascular unit; the minimum actin concentration required to generate vascular units equal to the number of basic unit MemAgents constituting the pseudopod is equal to:

[0254] The minimum number of basic units contained in a vascular unit is multiplied by the number of basic units MemAgent that make up a pseudopodia and multiplied by the actin concentration value t required for extension. ext .

[0255] and Figure 1 Corresponding to the embodiment shown, the embodiment of the present invention further provides an electronic device, see Figure 2 Specifically, it includes: one or more processors and memory, taking a processor 200 and a memory 100 as an example. The processor 200 and the memory 100 may be connected via a bus or other means, such as taking the connection via a bus as an example.

[0256] The memory 100 is a non-transitory computer-readable storage medium that can be used to store non-transitory software programs and non-transitory computer executable programs, such as a method for simulating a diabetic retinopathy process based on an agent model in an embodiment of the present application. The processor 200 implements the method for simulating a diabetic retinopathy process based on an agent model in an embodiment of the present application by running the non-transitory software program and instructions stored in the memory 100.

[0257] The memory 100 may include a program storage area and a data storage area, wherein the program storage area may store an operating device, an application required for at least one function; the data storage area may store data required for executing a method for simulating a diabetic retinopathy process based on an agent model in the above-mentioned embodiment, etc. In addition, the memory 100 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0258] The non-transient software program and instructions required to implement the method for simulating the diabetic retinopathy process based on the agent model in the above embodiment are stored in the memory, and when executed by one or more processors, the method for simulating the diabetic retinopathy process based on the agent model in the above embodiment is executed, for example, the above described method is executed. Figure 1 The method comprises steps S100 to S600.

[0259] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for simulating diabetic retinopathy process based on an agent model, characterized in that: The following steps are involved: Constructing a three-dimensional discrete coordinate system, and constructing a three-dimensional cube in the three-dimensional discrete coordinate system for simulating a retinal vascular environment; The definition content includes the basic unit of attributes, the vascular unit and the environmental point, and configures the simulated environment behavior rules for controlling the update of the attributes over time for the environmental point, configures the signal pathway antagonism behavior rules for controlling the basic unit attributes to change with the environmental point attributes for the basic unit, and configures the vascular three-dimensional model change behavior rules for controlling the vascular unit attributes to change with the basic unit attributes for the vascular unit; wherein the basic unit is defined as an agent representing a part of the epithelial cell; the vascular unit is defined as the smallest unit of the retinal vascular entity composed of the basic units; the environmental point is defined as an agent of the retinal environment entity where the retinal blood vessels to be simulated are located; Performing cube discretization on the three-dimensional model of the patient's retinal blood vessels constructed in the three-dimensional cube, and setting the obtained cube grid of a preset size as the environmental point position; Assuming that an environmental point with a three-dimensional model of retinal blood vessels accommodates one of the basic units, a row of the basic units connected in a ring is configured as a blood vessel unit, and the number of basic units contained in each blood vessel unit is recorded; Initializing and setting the attributes of the basic unit, the attributes of the blood vessel unit and the attributes of the environmental point respectively; wherein the attributes of the environmental point include the blood glucose concentration of the patient; The basic unit updates their respective attributes according to the signal pathway antagonism behavior rules, the environmental points according to the simulated environment behavior rules, and the vascular unit according to the vascular three-dimensional model change behavior rules, and the shape of the patient's retinal vascular three-dimensional model changes in real time according to the updated attributes.

2. The method according to claim 1, characterized in that The basic unit definition includes the following: The type of the basic unit is a proxy representing a physical part of a vascular epithelial cell; The properties of the basic unit include VEGFR concentration, activated VEGFR concentration, effective VEGFR concentration in activated VEGFR, Dll4 protein concentration, notch concentration, activated notch concentration, effective notch concentration in activated notch, and basic state of the basic unit; The vascular unit definition includes the following: The type of the blood vessel unit, the smallest physical unit of the blood vessel composed of basic units; The properties of the vascular unit include the number of basic units, the maximum number and the minimum number of basic units, VEGFR concentration, activated VEGFR concentration, effective VEGFR concentration in activated VEGFR, Dll4 protein concentration, notch concentration, activated notch concentration, and effective notch concentration in activated notch; The environmental point definition includes the following contents: The type of the environmental point, the proxy of the physical object of the retina environment where the retinal blood vessels to be simulated are located; The properties of the environmental points include oxygen concentration, VEGF concentration, and blood glucose concentration of the patient; The basic status includes: Stateless N: The basic unit that constitutes the main unit of blood vessels; Base state B: connecting the stateless N basic unit and the stem state S basic unit of the pseudopod; Stalk state S: a part of the pseudopod, a basic unit connecting the basic unit of the base state B and the basic unit of the tip state T; Tip state T: The tip cell is responsible for selecting the extension direction. When the pseudopodia are just generated, there are only the tip state T basic units. After the pseudopodia are extended, they are composed of the stem state S basic units and the tip state T basic units.

3. The method according to claim 2, characterized in that The simulated environment behavior rules for updating the control attributes of the environment point configuration over time include: Configure the oxygen concentration control equation for the environmental point, the equation is as follows: in: is the oxygen concentration at each environmental point in the retina at time t, and the initial value is obtained from the patient's most recent test result; is the diffusion coefficient of oxygen in the retina, a fixed constant; The oxygen consumption rate equation for each environmental point is: A is the normal oxygen consumption rate coefficient at each environmental point, a fixed constant; The minimum oxygen concentration required for normal physiological activities of each basic unit is a fixed constant. The basic unit below this constant value is in an oxygen-deficient state. is the amount of oxygen transported from the peripheral blood vessels in the current environmental point at time t, obtained according to the environmental point where each basic unit contained in the vascular unit is located, and the calculation equation is as follows: is the oxygen transport rate constant in blood vessels; is the average blood glucose concentration during T3; is the capillary density of the current environmental point at time t, which is obtained according to the number of basic units accommodated in the environmental points around the environmental point, specifically, the ratio of the number of basic units existing in a cube with a side length of N times the side length of the cube grid at the center of the current environmental point to the total number of environmental points; Configure the VEGF concentration control equation for the environmental point, the equation is as follows: in: is the VEGF concentration at each environmental point at time t; is the degradation rate of VEGF at the environmental site, a fixed constant; is the VEGF production rate at the environmental site, a fixed constant; is the VEGF diffusion constant at the environmental site; The governing equation for the VEGF concentration generated by oxygen concentration is as follows: Configure the control equation for blood glucose concentration for the environmental point and input according to the patient's specific situation: in: is the blood glucose concentration at the current environmental point at time t; is the equation of the change of blood glucose concentration at the environmental point over time, and the initial value is obtained by patient testing; It is the increase in blood sugar concentration over time after food intake; is the equation of insulin concentration changing with time, and the initial value of insulin concentration is obtained from the patient's test; is the equation of glucagon concentration changing with time, and the initial value is obtained from the patient test; k1 is the influence coefficient of insulin concentration on blood glucose concentration, which is a fixed constant; k2 is the coefficient of influence of glucagon concentration on blood glucose concentration, which is a fixed constant; The equation for the change in blood glucose concentration over time for liver effect consumption.

4. The method according to claim 2, characterized in that: The signal pathway antagonistic behavior rule configured for the basic unit to control the basic unit attribute changing with the environmental point attribute includes: Configuring the signal pathway antagonistic behavior rules of VEGF binding to the basic unit VEGFR in the environmental point includes: (1) VEGF in the environmental point binds to the basic unit VEGFR, VEGFR is activated, and after T1 time, Dll4 protein is generated in the current basic unit; (2) Because Dll4 protein is produced in the basic unit, the concentration of Dll4 protein increases. After binding with notch in the adjacent basic unit, the concentration of Dll4 protein in the basic unit returns to the initial concentration after R1 time; Configuring the signaling pathway antagonistic behavior rules of the Dll4 protein in the basic unit binding to the basic unit notch, including: (1) The Dll4 protein in the basic unit binds to the notch in the adjacent basic unit, and the notch is activated. After T2 time, VEGFR is generated in the current basic unit; wherein the adjacent basic unit is the environmental point that accommodates the basic unit among the 26 environmental points connected to the basic unit; (2) Because the basic unit produces VEGFR, the VEGFR concentration increases. After combining with VEGF in the 26 environmental points adjacent to the environmental point where the basic unit is located, the VEGFR concentration in the basic unit returns to the initial concentration after R2 time.

5. The method according to claim 4, characterized in that The control equation for the concentration of VEGFR activated by VEGF in the environmental point in the basic unit is as follows: in: is the concentration of VEGFR activated by VEGF in the 26 adjacent environmental points of the current basic unit; The concentration of VEGF remaining after binding to the receptor type of VEGFR in the non-basic unit and sedimentation is a fixed constant; is the number of basic units contained in the vascular unit to which the current basic unit belongs; The maximum VEGFR concentration allowed in the basic unit of the vascular unit to which the current basic unit belongs is a fixed constant; is the VEGFR concentration of the current basic unit, a fixed constant; It is the sum of VEGF concentrations in 26 environmental points adjacent to the environmental point where the current basic unit is located; The Notch concentration activated by the basic unit is the sum of the Notch concentration of the basic unit and the Dll4 protein concentration of all adjacent basic units. The minimum value of the sum of the two is as follows: in, The concentration of notch activated for the basic unit; is the notch concentration of the basic unit; is the Dll4 protein concentration in basic units.

6. The method according to claim 4, characterized in that At time t, the Dll4 protein concentration in the vascular unit and VEGFR concentration The control equations are as follows: in, is the concentration of Dll4 protein generated by the binding of VEGFR and VEGF in the vascular unit, which is a constant; is the concentration of VEGFR generated by the binding of notch and Dll4 proteins in the vascular unit, which is a constant; is the activated and effective notch concentration, the activated notch concentration after T1 time step , and The value is the same, but it is obtained at time t+T1. renew; For activated and effective VEGFR, the concentration of activated VEGFR after the T2 time step , and The value is the same, but it is obtained at time t+T2. renew; The concentration of Dll4 protein in vascular unit and VEGFR concentration Divide it evenly into the basic unit to which it belongs, and get the Dll4 protein concentration of each basic unit and notch concentration .

7. The method according to claim 2, characterized in that The blood vessel three-dimensional model change behavior rule configured for the blood vessel unit to control the blood vessel unit attribute to change with the basic unit attribute includes: The configuration of the behavior rules for simulating pseudopod state changes as the basic unit properties change is as follows: At every time step, the actin concentration of the basic unit is calculated according to the concentration of VEGFR activated by VEGF, and the number of basic units constituting each pseudopodia is counted; According to the actin concentration of the basic unit and the number of basic units constituting each pseudopod, the behavior rules for simulating the initial generation of pseudopods, the behavior rules for simulating the extension of pseudopods and the behavior rules for simulating the retraction of pseudopods are implemented to simulate the behavior rules of the basic state of the basic unit changing over time; The configuration of the behavioral rules that simulate the evolution of pseudopods into vascular units is as follows: If the number of basic units constituting pseudopods reaches the limit of pseudopod extension , and the number of time steps in which the basic unit T of the pseudopod is not extended reaches the preset number Before, the accumulated actin concentration of the basic unit of the pseudopod in the tip state T reaches the minimum actin concentration required to generate vascular units equal to the number of basic units constituting the pseudopod, and vascular units are generated respectively with each basic unit of the pseudopod as the center, and the basic state of the basic unit generating the vascular unit is set to the stateless N; Among them, the number of vascular units generated is equal to the number of basic units constituting pseudopodia, the number of basic units contained in each vascular unit generated is the minimum number of basic units contained in the vascular unit; the minimum actin concentration required to generate vascular units equal to the number of basic units constituting pseudopodia is equal to: The minimum number of basic units contained in the vascular unit is multiplied by the number of basic units constituting pseudopods and the actin concentration required for extension. .

8. The method according to claim 7, characterized in that The actin concentration in the basic unit The control equation is as follows: Among them, C is the signal pathway activation intensity constant representing the binding of VEGF to the basic unit VEGFR in the environmental site; is the concentration of VEGFR activated by VEGF in the basic unit; The number of basic units contained in the vascular unit to which the current basic unit belongs; The maximum allowed VEGFR concentration of the vascular unit to which the current basic unit belongs, which is a fixed constant; The actin generated by the stateless N basic unit is evenly distributed to the base state B basic unit of the pseudopodia within a radius of 20 microns from itself, and the actin is transferred to the tip state T basic unit of the pseudopodia through the stem state S basic unit of the pseudopodia. The amount of actin distributed to the corresponding pseudopodia by the tip state T basic unit of the pseudopodia increases.

9. The method according to claim 7, characterized in that: The configuration simulates the behavioral rules for the initial generation of pseudopods, including: If the actin concentration of the stateless N basic unit constituting the main blood vessel Actin concentration required to reach pseudopod extension When , the basic state of the basic unit changes from the non-state N to the base state B, and the environmental point with the highest VEGF concentration among all the environmental points adjacent to the basic unit that do not accommodate the basic unit is set to accommodate one basic unit, and the basic state is the sharp state T; If the VEGF concentrations of all the environmental points adjacent to the basic unit that do not accommodate the basic unit are the same, then a random environmental point is selected to be set as a basic unit that accommodates a sharp state T; The configuration simulates the behavioral rules of pseudopod extension, including: If the pseudopod is in the pointed state, the actin concentration of the basic unit T Actin concentration required to reach pseudopod extension , and the number of basic units constituting pseudopods has not reached the extension limit , the number of time steps during which the tip state T basic unit of the pseudopod is not extended is cleared, and a stem state S basic unit is added between the base state B basic unit and the tip state T basic unit of the current pseudopod; otherwise, the number of time steps during which the tip state T basic unit of the pseudopod is not extended is increased by 1; wherein, the adding of a stem state S basic unit between the base state B basic unit and the tip state T basic unit of the current pseudopod is specifically as follows: Find the environmental point with the highest VEGF concentration among all environmental points adjacent to the pointed state T basic unit that do not accommodate the stateless N basic unit and the base state B basic unit, and use the environmental point as the extension direction; if the VEGF concentrations of the environmental points of all the stateless N basic units and the base state B basic units adjacent to the pointed state T basic unit are the same, then select one environmental point as the extension direction of the pseudopod; Determine whether the environmental point of the found environmental point can accommodate the stem state S basic unit or the tip state T basic unit; If the environmental point found does not accommodate, the environmental point found is set to accommodate a pointed state T basic unit, and the environmental point where the pseudopod's pointed state T basic unit is located is set to accommodate the stem state S basic unit; wherein the VEGFR concentration and Dll4 protein concentration of the pointed state T basic unit of the environmental point found are the pseudopod's pointed state T basic unit, and the actin concentration is the actin concentration of the pseudopod's pointed state T basic unit, which is reduced by one actin concentration value required for extension. ; If the found environmental point accommodates the stem state S basic unit or the tip state T basic unit, the basic state of the tip state T basic unit of the pseudopod is changed to the stem state S; the actin of the tip state T basic unit of the pseudopod is transferred to the tip state T basic unit of the pseudopod to which the found environmental point accommodates the stem state S basic unit or the tip state T basic unit belongs; The configuration simulates the behavior rules of pseudopod retreat, including: If the number of time steps in which the pseudopod tip state T basic unit is not extended is greater than or equal to the preset number , and the actin concentration p of the pseudopod's tip state T basic unit does not reach the actin concentration required for extension , delete a basic unit contained in pseudopods; wherein, delete a basic unit contained in pseudopods, as follows: Divide all the actin in the T basic unit of the pseudopod tip equally to all the basic units connected to the T basic unit of the pseudopod tip; Determine whether the basic state of the basic unit connected to the tip state T basic unit of the pseudopodia is the base state B; if it is the base state B, delete the tip state T basic unit of the pseudopodia, and change the basic state of the base state B basic unit connected to the tip state T basic unit of the pseudopodia to the stateless N; if it is not the base state B, that is, the stem state S, delete the tip state T basic unit of the pseudopodia, and change the basic state of the stem state S basic unit connected to the tip state T basic unit of the pseudopodia to the tip state T.

10. An electronic device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements any one of the methods described in claims 1 to 9 when executing the computer program.

Images