Formula optimization and preparation method of composite gel for sodium fluoride drug sustained release

By optimizing the formulation of graphene oxide/polyacrylamide composite hydrogel, using a variety of mathematical modeling and machine learning technologies, the sustained release of sodium fluoride in the oral environment is achieved, and the sodium fluoride slowdown in the existing technology is solved The problems of low release efficiency and overdose use have achieved excellent drug sustained release performance and biocompatibility.

CN119074647AInactive Publication Date: 2024-12-06HANTECH MEDICAL DEVICE CO
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411329650.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-09-24
Publication Date
2024-12-06
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The prior art fails to effectively utilize graphene oxide/polyacrylamide composite hydrogel to achieve sustained release of sodium fluoride in the oral environment, resulting in insufficient bioavailability of fluoride and risk of excessive use.

Method used

By selecting suitable gel components such as graphene oxide, acrylamide, acrylic acid, crosslinking agent, initiator, pH adjuster, natural polymer materials and biofunctional modifier, design multiple sets of initial formulations and use principal component analysis , Box-Behnken experimental design, genetic algorithm and deep neural network model for formulation optimization, establish a second-order polynomial model and multi-branch structure model, optimize graphene oxide concentration, monomer ratio, crosslinking agent dosage and natural polymer material content, Achieve comprehensive optimization of fluoride ion sustained release kinetics, mechanical properties and biocompatibility.

Benefits of technology

Developed a sodium fluoride sustained release composite gel with excellent drug sustained release performance, good biocompatibility and controllable mechanical properties, which reduces the risk of excessive use of fluoride and increases the bioavailability of fluoride. It is suitable for Local administration for oral diseases.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119074647A_ABST
    Figure CN119074647A_ABST
Patent Text Reader

Abstract

The invention provides a formula optimization and preparation method of a composite gel for sodium fluoride drug sustained release, and belongs to the technical field of composite gels for sodium fluoride drug sustained release, and the formula optimization method comprises the following steps: designing a plurality of groups of initial formulas based on literatures and preparatory experiments, the influence of the graphene oxide, the monomer, the cross-linking agent and other components on the slow-release gel performance is emphatically investigated. Main formula parameters are determined through a principal component analysis method, and a second-order polynomial model is established to optimize performance indexes. A deep neural network model is adopted to perform virtual screening in a wider parameter space, and candidate formulas are selected in combination with an analytic hierarchy process and a TOPSIS method. The preparation method comprises the following steps: uniformly dispersing graphene oxide in deionized water, adding acrylic acid and acrylamide monomers, performing ultrasonic dispersion, degassing, adding a cross-linking agent and an initiator, heating in a water bath to obtain polyacrylamide composite hydrogel, and soaking the polyacrylamide composite hydrogel in a sodium fluoride solution to obtain the composite hydrogel loaded with the sodium fluoride medicine.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of composite gels for sustained-release of sodium fluoride drugs, and in particular relates to a formula optimization and a preparation method of a composite gel for sustained-release of sodium fluoride drugs. Background Art

[0002] Oral diseases such as dental caries and periodontitis seriously affect the quality of life of individuals and have become one of the major public health issues worldwide. Oral diseases are mostly caused by infections caused by a variety of microorganisms such as aerobic bacteria and anaerobic bacteria. The use of antibiotics or other antimicrobial drugs is an effective method for treating and controlling these diseases, but their clinical efficacy is often limited due to their insufficient bioavailability, poor absorption by the human body, and poor pharmacokinetic properties.

[0003] The application of fluoride to prevent dental caries is a recognized effective and relatively safe method. Currently, topical application of fluoride is the most widely used measure to prevent dental caries. However, long-term and excessive use of fluoride may cause adverse reactions such as dental fluorosis. Therefore, it is necessary to develop a drug delivery system that can achieve sustained release of sodium fluoride, which can not only improve the utilization efficiency of fluoride, but also reduce the risk of excessive use of fluoride.

[0004] As a drug delivery system with good biocompatibility, hydrogel has a three-dimensional network structure and high water content, which is very suitable for drug loading and sustained release. The adsorption and swelling properties of hydrogel are particularly suitable for regulating the release kinetics of drugs and achieving sustained drug release. At present, many studies have reported on hydrogel-based drug sustained release systems, which have broad application prospects in the fields of oral disease treatment.

[0005] On the other hand, graphene and its derivatives such as graphene oxide have also attracted much attention in the field of drug delivery due to their unique physical and chemical properties. Graphene oxide can form complexes with some drug molecules through π-π interactions, thereby regulating the dissolution and release of drugs. The combination of graphene oxide and hydrogel is expected to further improve the performance and function of the gel.

[0006] However, there is no report on the application of graphene oxide / polyacrylamide composite hydrogel in the sustained release of sodium fluoride. Therefore, it is of great practical significance to develop a sodium fluoride sustained release gel that can be used in the oral environment. Summary of the invention

[0007] In view of this, the present invention provides a formula optimization and preparation method of a composite gel for sustained release of sodium fluoride drugs, which can solve the technical problem in the prior art that graphene oxide / polyacrylamide composite hydrogel is not applied to sustained release of sodium fluoride.

[0008] The present invention is achieved in that:

[0009] The first aspect of the present invention provides a method for optimizing the formulation of a composite gel for sustained release of sodium fluoride drugs, which comprises the following steps:

[0010] S10, selecting gel components suitable for sustained release of sodium fluoride, including: graphene oxide, acrylamide, acrylic acid, a cross-linking agent, an initiator, a pH regulator, a hydrophilic / lipophilic regulator, a natural polymer material, and a biofunctional modifier;

[0011] S20. Design multiple initial formulations based on literature research and preliminary experiments; focus on graphene oxide concentration, acrylamide: acrylic acid monomer ratio and crosslinker dosage; obtain performance indicators of composite gels through experiments, including fluoride ion release kinetics, initial burst effect, pH responsiveness, mechanical strength and biocompatibility;

[0012] S30, using the principal component analysis method, calculating the correlation between each component and the performance index, constructing a correlation matrix and determining the main formula parameters, wherein the main formula parameters represent the graphene oxide concentration, acrylamide: acrylic acid ratio, crosslinking agent dosage, pH regulator concentration and natural polymer material content;

[0013] S40, using Box-Behnken design based on the correlation matrix to establish a second-order polynomial model between each performance index and the main formulation parameters; using stepwise regression method to screen significant factors and optimize the structure of the second-order polynomial model;

[0014] S50, evaluating the accuracy and reliability of the second-order polynomial model through residual analysis, variance analysis and lack of fit test; using a genetic algorithm to perform multi-objective optimization on the multiple groups of initial formulas to find the optimal formula interval that balances various performances;

[0015] S60. Based on the performance indicators of the composite gel obtained from the experiment, a deep neural network model with a multi-branch structure is trained to obtain a formula optimization model;

[0016] S70, using the recipe optimization model to perform virtual screening in a wider recipe parameter space; using a Monte Carlo simulation method to generate a large number of candidate recipes, and using a trained deep neural network model to predict the performance of the candidate recipes;

[0017] S80. Use the analytic hierarchy process and TOPSIS method to comprehensively consider the importance of various performance indicators, and comprehensively score and rank the candidate formulas; select several candidate formulas with the highest scores for experimental verification;

[0018] S90, prepare highly ranked candidate formulations and conduct comprehensive performance tests; compare actual performance with predicted results to evaluate the accuracy of the formulation optimization model; fine-tune and update the formulation optimization model based on the verification results; repeat steps S60-S90 until the final optimized formulation is obtained.

[0019] The fitness function of the genetic algorithm adopts the weighted sum of various performance indicators predicted by the second-order polynomial model.

[0020] The cross-linking agent is N,N-methylenebisacrylamide; the initiator is ammonium persulfate; the natural polymer material includes sodium alginate, chitosan, collagen and hyaluronic acid; and the biofunctional modifier includes RGD peptide, cell adhesion factor and growth factor.

[0021] Among them, the multi-branch structured deep neural network model includes a fluoride ion release branch network, a mechanical property branch network and a biocompatibility branch network.

[0022] Furthermore, the fluoride ion release branch network is used to predict the fluoride ion release kinetics and initial burst effect, the input is the formulation parameters and pH value, the output is the release curve parameters and the initial release amount, and the structure is composed of a three-layer convolutional neural network and two fully connected layers;

[0023] The mechanical property branch network is used to predict the mechanical strength and elastic modulus of the gel, the input is the formula parameters and the cross-linking density, the output is the compressive strength and Young's modulus, and the structure is a feedforward neural network consisting of four fully connected layers;

[0024] The biocompatibility branch network is used to predict the cytotoxicity and biodegradability of the gel. The input is the formulation parameters and the type of biofunctional modifiers. The output is the cell survival rate and degradation half-life. The structure is a recurrent neural network composed of long short-term memory layers and fully connected layers.

[0025] The second-order polynomial model is specifically expressed as follows:

[0026] ;

[0027] In the formula, It is the comprehensive performance index of composite gel; is the graphene oxide concentration; is the ratio of acrylamide to acrylic acid; is the amount of cross-linking agent; is the pH regulator concentration; It is the content of natural polymer materials; is the intercept term; is the coefficient of the first-order term; is the interaction term coefficient; is the coefficient of the quadratic term; is the pH responsiveness coefficient; is the mechanical strength influence coefficient; is the biocompatibility adjustment coefficient; is the error term; Represents two independent variables and The multiplication of and The difference is only in the subscript.

[0028] In the formula, the parameter acquisition method and calculation steps are as follows:

[0029] 1. arrive : Obtained through experimental design. The Box-Behnken design method was used to design factor levels and conduct composite gel preparation experiments.

[0030] 2. , , , : Obtained by fitting the experimental data using the least squares method.

[0031] 3. : Obtained by measuring the change in swelling rate of the gel under different pH conditions. The calculation formula is:

[0032] ;

[0033] in, is the swelling rate change, is the pH change, is the number of measurements.

[0034] 4. : represents the rate of change of fluoride ion concentration in the gel over time, determined experimentally. The fluoride ion concentration in the dissolution solution is measured at different time points using an ion selective electrode, and then the rate of change of the concentration over time is calculated.

[0035] 5. : Obtained through compression test. The calculation formula is:

[0036] ;

[0037] in, is the applied force, is the cross-sectional area of ​​the gel, is the shape variable, is the initial length.

[0038] 6. : represents a complex power term, which is used to simulate the viscoelastic behavior of the gel. and Obtained through Dynamic Mechanical Analysis (DMA) testing, is the fitting parameter.

[0039] 7. : Obtained through cytotoxicity experiments. The calculation formula is:

[0040] ;

[0041] in, For the The cell viability at each concentration was As negative control, As a positive control, is the test concentration number, here it is 5.

[0042] 8. : represents an alternating series used to simulate the nonlinear changes in gel properties.

[0043] 9. : Error term, assumed to be normally distributed ,in Estimated by residual analysis.

[0044] This model comprehensively considers various performance indicators of the gel, including drug release kinetics, pH responsiveness, mechanical strength and biocompatibility, and can more comprehensively describe the relationship between the performance of the composite gel and the formulation parameters.

[0045] Specifically, step S10 includes: obtaining from existing material literature research and preliminary experiments, selecting 8 types of components including graphene oxide, acrylamide, acrylic acid, crosslinking agent, initiator, pH regulator, hydrophilic / lipophilic regulator, natural polymer material and biofunctional modifier as gel components suitable for sodium fluoride sustained release. Among them, graphene oxide, as one of the main components of the gel matrix, can form a complex with sodium fluoride to control drug sustained release; acrylamide and acrylic acid are the other two main monomer components of the gel matrix, and their molar ratio is an important parameter for regulating gel performance; crosslinking agents, initiators, etc. participate in the crosslinking polymerization reaction of the gel, which has an important influence on the final gel performance; pH regulators and hydrophilic / lipophilic regulators are used to adjust the environmental responsiveness of the gel; natural polymer materials and biofunctional modifiers give the gel good biocompatibility and bioactivity.

[0046] Wherein, the step S20 includes: designing multiple groups of initial formulas according to literature research and preliminary experiments, focusing on the three parameters of graphene oxide concentration of 1-5% by weight, molar ratio of acrylamide to acrylic acid between 1:1 and 3:1, and amount of cross-linking agent N,N-methylenebisacrylamide between 0.01-0.03 mol%. The performance indicators such as fluoride ion release kinetics, pH responsiveness, mechanical strength and biocompatibility of the obtained composite gel are obtained through experimental testing. These experimental data provide a basis for subsequent formula optimization.

[0047] Wherein, the step S30 includes: using principal component analysis to calculate the correlation matrix between each formulation parameter and the performance index, and determining the five main parameters affecting the performance of the composite gel, namely, graphene oxide concentration, acrylamide: acrylic acid ratio, crosslinking agent dosage, pH regulator concentration and natural polymer material content. These main parameters will become the focus of subsequent optimization.

[0048] Wherein, the step S40 includes: based on the Box-Behnken experimental design, a second-order polynomial regression model between each performance index and the main formulation parameters is established. The least squares method is used to fit the experimental data to obtain the regression coefficient, and the stepwise regression method is used to screen out the significant factors and optimize the model structure. The established second-order polynomial model can quantitatively describe the relationship between the composite gel performance and the formulation parameters, providing a mathematical tool for subsequent intelligent optimization.

[0049] Wherein, the step S50 includes: comprehensively evaluating the accuracy and reliability of the established second-order polynomial model by residual analysis, variance analysis and lack of fit test. If these tests are passed, it can be considered that the model can better describe the relationship between the performance of the composite gel and the formulation parameters. Next, the initial formulation will be optimized by a genetic algorithm to find the optimal formulation interval that balances various performances.

[0050] Wherein, the step S60 includes: based on the composite gel performance data obtained from the experiment, training a deep neural network model with a multi-branch structure, including a fluoride ion release branch network, a mechanical property branch network and a biocompatibility branch network. This multi-branch structure can better capture the complex relationship between various performance indicators, and the obtained deep neural network model is the formula optimization model.

[0051] Wherein, the step S70 includes: using the deep neural network model trained in step S60 to perform virtual screening in a wider formulation parameter space. A large number of candidate formulations are generated by Monte Carlo simulation, and their performance indicators are predicted by the deep neural network model. Then, the candidate formulations are comprehensively scored and ranked by the analytic hierarchy process and TOPSIS method, and the highest-scoring ones are selected for experimental verification. Such a virtual screening process can greatly improve the efficiency of optimization.

[0052] A second aspect of the present invention provides a method for preparing a composite gel for sustained release of sodium fluoride drugs, wherein the method for preparing the composite gel for sustained release of sodium fluoride drugs after optimizing the formula of the composite gel for sustained release of sodium fluoride drugs described above comprises the following steps:

[0053] uniformly dispersing graphene oxide in deionized water to obtain a graphene oxide dispersion;

[0054] Adding acrylic acid and acrylamide monomers to the graphene oxide dispersion, and performing ultrasonic dispersion to mix the solution uniformly to obtain a gel solution;

[0055] Degassing the gel solution;

[0056] Adding a crosslinking agent and an initiator to the degassed gel solution, and heating the solution in a water bath to obtain a polyacrylamide composite hydrogel;

[0057] The polyacrylamide composite hydrogel is immersed in a sodium fluoride solution to obtain a composite hydrogel loaded with sodium fluoride medicine.

[0058] Furthermore, the concentration of the graphene oxide is 1-5 wt%.

[0059] Furthermore, the molar ratio of the acrylamide monomer to the acrylic acid monomer is 1:1 to 3:1.

[0060] Furthermore, the crosslinking agent is N,N-methylenebisacrylamide, and the molar mass of the crosslinking agent is 0.01 to 0.03 grams per mole; the initiator is ammonium persulfate, and the molar mass of the initiator is 0.005 to 0.01 grams per mole; and the water bath temperature is 40 to 60 degrees Celsius.

[0061] Compared with the prior art, the composite gel provided by the present invention for the formulation optimization and preparation method of the composite gel for the sustained release of sodium fluoride drugs uses graphene oxide as the gel matrix, forms a three-dimensional network structure with acrylamide-acrylic acid copolymer, and is loaded with sodium fluoride drugs. Compared with the prior art, the composite gel has the following significant advantages:

[0062] 1. Good fluoride ion sustained release performance. Graphene oxide can form The complex helps to control the slow release of the drug and avoid adverse reactions caused by local high concentrations. At the same time, the swelling and pH responsiveness of acrylamide-acrylic acid copolymers are also conducive to the sustained release of sodium fluoride.

[0063] 2. Excellent biocompatibility. The high water content and hydrophilicity of the hydrogel, as well as the added natural polymer materials such as sodium alginate and chitosan, give the composite gel good biocompatibility. Biofunctional modification further enhances the cell affinity and biological activity of the gel.

[0064] 3. Controllable mechanical properties. By adjusting the monomer ratio of acrylamide and acrylic acid, as well as the amount of chemical crosslinking agent, the mechanical strength and elasticity of the gel can be effectively controlled to meet the requirements of use in the oral environment.

[0065] 4. Operable preparation process. This method adopts common conventional chemical reaction steps such as polymerization and impregnation. The preparation process is simple and controllable, and is easy for industrial production.

[0066] In summary, the graphene oxide / polyacrylamide composite hydrogel developed in the present invention not only has excellent drug sustained-release performance, good biocompatibility and controllable mechanical properties, but also has a relatively simple preparation process, providing an innovative solution for local administration of oral diseases, and solving the technical problem in the prior art that graphene oxide / polyacrylamide composite hydrogel has not been applied to the sustained-release of sodium fluoride. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Figure 1 The present invention provides a flow chart of a method for optimizing the formulation of a composite gel for sustained release of sodium fluoride drugs in the first aspect. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution 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.

[0069] like Figure 1 As shown, it is a flow chart of a method for optimizing the formula of a composite gel for sustained release of sodium fluoride drugs provided by the first aspect of the present invention. The method comprises the following steps:

[0070] S10, selecting gel components suitable for sustained release of sodium fluoride, including: graphene oxide, acrylamide, acrylic acid, a cross-linking agent, an initiator, a pH regulator, a hydrophilic / lipophilic regulator, a natural polymer material, and a biofunctional modifier;

[0071] S20. Design multiple initial formulations based on literature research and preliminary experiments; focus on graphene oxide concentration, acrylamide: acrylic acid monomer ratio and crosslinker dosage; obtain performance indicators of composite gels through experiments, including fluoride ion release kinetics, initial burst effect, pH responsiveness, mechanical strength and biocompatibility;

[0072] S30, using the principal component analysis method, calculating the correlation between each component and the performance index, constructing a correlation matrix and determining the main formula parameters, the main formula parameters represent the graphene oxide concentration, acrylamide: acrylic acid ratio, cross-linking agent dosage, pH regulator concentration and natural polymer material content;

[0073] S40. Based on the correlation matrix, a second-order polynomial model between each performance index and the main formulation parameters was established using Box-Behnken design; the stepwise regression method was used to screen significant factors and optimize the structure of the second-order polynomial model;

[0074] S50. Evaluate the accuracy and reliability of the second-order polynomial model through residual analysis, variance analysis and lack of fit test; use genetic algorithm to perform multi-objective optimization on multiple groups of initial formulas to find the optimal formula range that balances various properties;

[0075] S60. Based on the performance indicators of the composite gel obtained from the experiment, a deep neural network model with a multi-branch structure is trained to obtain a formula optimization model;

[0076] S70. Use the recipe optimization model to perform virtual screening in a wider recipe parameter space; use the Monte Carlo simulation method to generate a large number of candidate recipes, and use the trained deep neural network model to predict the performance of the candidate recipes;

[0077] S80. Use the analytic hierarchy process and TOPSIS method to comprehensively consider the importance of various performance indicators, and comprehensively score and rank the candidate formulas; select several candidate formulas with the highest scores for experimental verification;

[0078] S90. Prepare highly ranked candidate formulations and conduct comprehensive performance tests; compare actual performance with predicted results to evaluate the accuracy of the formulation optimization model; fine-tune and update the formulation optimization model based on the verification results; repeat steps S60-S90 until the final optimized formulation is obtained.

[0079] The specific implementation methods of the above steps are described in detail below:

[0080] Step S10: Selecting a gel component suitable for sustained release of sodium fluoride

[0081] In this step, you first need to select a gel component suitable for sustained release of sodium fluoride. Based on literature research and existing preliminary experimental experience, the following types of components can be selected:

[0082] 1. Graphene oxide: As one of the main components of the gel matrix, it can form a complex with sodium fluoride through π-π interaction, thereby controlling the sustained release behavior of the drug. The concentration of graphene oxide is an important formulation parameter.

[0083] 2. Acrylamide and acrylic acid: As the other two main monomer components of the gel matrix, the molar ratio between acrylamide and acrylic acid is another important formulation parameter that can regulate the mechanical properties and pH responsiveness of the gel.

[0084] 3. Cross-linking agent: N,N-methylenebisacrylamide is a commonly used chemical cross-linking agent that can copolymerize with acrylamide and acrylic acid to form a three-dimensional network structure. The amount of cross-linking agent is also an important formulation parameter.

[0085] 4. Initiator: Ammonium persulfate is a commonly used free radical initiator that can promote the polymerization of acrylamide and acrylic acid monomers. The amount of initiator needs to be controlled within an appropriate range.

[0086] 5. pH regulator: Some weak acid or weak alkaline substances can be selected as pH regulators, such as citric acid, sodium dihydrogen phosphate, etc., to adjust the pH responsiveness of the gel. The concentration of the pH regulator is also an important parameter.

[0087] 6. Hydrophilic / lipophilic regulator: Some hydrophilic or lipophilic small molecules or high molecular substances, such as polyethylene glycol, polypropylene glycol, etc., can be added to adjust the hydrophilicity and hydrophobicity of the gel.

[0088] 7. Natural polymer materials: such as sodium alginate, chitosan, collagen and hyaluronic acid, can give the gel good biocompatibility and biological functionality.

[0089] 8. Biofunctional modifiers: such as RGD peptides, cell adhesion factors and growth factors, can further enhance the bioactivity and cell affinity of the gel.

[0090] The above 8 types of components are all optional gel formula ingredients in this method. By properly selecting these components and optimizing the proportions, a composite gel system suitable for the sustained release of sodium fluoride can be designed.

[0091] Step S20: Design multiple groups of initial formulations and conduct performance evaluation

[0092] In this step, based on the gel components selected in the above step S10, multiple groups of initial formulas are designed through literature research and preliminary experiments. The focus is on the following three aspects:

[0093] 1. Concentration of graphene oxide: usually the range of 1-5 wt% is selected for exploration.

[0094] 2. The monomer ratio of acrylamide and acrylic acid: usually a molar ratio range of 1:1 to 3:1 is selected for exploration.

[0095] 3. The amount of cross-linking agent used: usually the range of 0.01-0.03 mol% is selected for exploration.

[0096] For the designed initial formula, various performance indicators of the composite gel need to be obtained through experimental means, including:

[0097] 1. Release kinetics of fluoride ions: including release curve parameters and initial burst effect.

[0098] 2. pH responsiveness: obtained by measuring the change in gel swelling rate under different pH conditions.

[0099] 3. Mechanical strength: including compressive strength and Young's modulus.

[0100] 4. Biocompatibility: evaluated through cytotoxicity and biodegradability experiments.

[0101] The measurement results of these performance indicators will provide important basis for subsequent formula optimization.

[0102] Step S30: Constructing a correlation matrix between recipe parameters and performance indicators

[0103] In this step, the principal component analysis (PCA) method will be used to calculate the correlation between each recipe parameter and the performance index. The specific steps are as follows:

[0104] 1. Organize the performance index data obtained from the experiment into a matrix form .

[0105] 2. Arrange the designed formula parameter data into a matrix form .

[0106] 3. Calculation and The correlation matrix between :

[0107] ;

[0108] in, is the number of samples.

[0109] 4. Correlation Matrix Perform eigenvalue decomposition to obtain the principal component weight vector .

[0110] 5. According to the principal component weight vector Determine the main formulation parameters, usually selecting the parameters corresponding to the first five main components. These main parameters include: graphene oxide concentration; acrylamide: acrylic acid monomer ratio; crosslinking agent dosage; pH adjuster concentration; natural polymer material content; these main formulation parameters will be the focus of subsequent optimization.

[0111] Step S40: Establishing a second-order polynomial model of recipe parameters and performance indicators

[0112] In this step, the Box-Behnken experimental design method will be used to establish a second-order polynomial regression model between various performance indicators and main formulation parameters. The specific steps are as follows:

[0113] 1. According to the five main formula parameters determined in step S30, determine the factor levels of the Box-Behnken test. Usually, three levels -1, 0, and +1 are selected.

[0114] 2. According to the Box-Behnken test design matrix, prepare the corresponding composite gel samples and test various performance indicators.

[0115] 3. Use the least squares method to fit the experimental data and establish various performance indicators With the main recipe parameters A second-order polynomial model between:

[0116] ;

[0117] in, is the regression coefficient, is a random error term.

[0118] 4. Use stepwise regression to screen out significant factors and optimize the structure of the second-order polynomial model.

[0119] The second-order polynomial model established by this method can quantitatively describe the relationship between various performance indicators of the composite gel and the main formulation parameters. Subsequent optimization will rely heavily on this model.

[0120] Step S50: Evaluating the accuracy and reliability of the second-order polynomial model

[0121] In this step, the accuracy and reliability of the established second-order polynomial model will be evaluated using the following methods:

[0122] 1. Residual Analysis: Calculate the residual between the model prediction value and the experimental value, check whether it obeys the normal distribution, and verify whether the linear assumption of the model is valid.

[0123] 2. Analysis of variance (ANOVA): Perform significance test on the regression model and evaluate the significance level of each regression coefficient.

[0124] 3. Lack-of-Fit Test: Test whether the model fits well, that is, whether there are systematic errors in the model.

[0125] Through the above analysis, the accuracy and reliability of the established second-order polynomial model can be comprehensively evaluated. If the model passes these tests, it can be considered that it can better describe the relationship between formulation parameters and performance indicators.

[0126] Next, we will use the genetic algorithm to perform multi-objective optimization on these initial formulas to find the optimal formula interval that balances various performances. The fitness function of the genetic algorithm uses the weighted sum of various performance indicators predicted by the second-order polynomial model.

[0127] Step S60: Training a deep neural network model for formula optimization

[0128] In this step, a multi-branch deep neural network model is trained based on the composite gel performance index data obtained from the experiment to predict the performance of the composite gel. Specifically, it includes the following three branch networks:

[0129] 1. Fluoride ion release branch network:

[0130] Input: recipe parameters and pH value;

[0131] Output: release curve parameters and initial release amount;

[0132] Network structure: It consists of three layers of convolutional neural network and two layers of fully connected layers;

[0133] 2. Mechanical properties branch network:

[0134] Input: formulation parameters and cross-link density;

[0135] Output: compressive strength and Young's modulus;

[0136] Network structure: A feed-forward neural network consisting of four fully connected layers;

[0137] 3. Biocompatibility branch network:

[0138] Input: formulation parameters and biofunctional modifier type;

[0139] Output: Cell viability and degradation half-life;

[0140] Network structure: A recurrent neural network consisting of a long short-term memory (LSTM) layer and a fully connected layer;

[0141] This multi-branch structure can better capture the complex relationship between the various performance indicators of the composite gel. After training, the obtained deep neural network model is the formula optimization model, which can be used to guide subsequent virtual screening in a wider space.

[0142] Step S70: Virtual screening based on deep neural network model

[0143] In this step, the deep neural network model trained in step S60 is used to perform virtual screening in a wider formulation parameter space to find the optimal composite gel formulation. The specific steps are as follows:

[0144] 1. Using the Monte Carlo simulation method, a large number of candidate recipes are generated within the range of recipe parameters.

[0145] 2. Input these candidate recipes into the trained deep neural network model to predict their performance indicators.

[0146] 3. The Analytic Hierarchy Process (AHP) and TOPSIS method (Technique for Order of Preference by Similarity to Ideal Solution) were used to comprehensively consider the importance of various performance indicators and comprehensively score and rank the candidate formulations.

[0147] 4. Select several candidate formulations with the highest scores for experimental verification.

[0148] Such a virtual screening process can search for the optimal solution in a wider formulation space, greatly improving the efficiency of optimization. Subsequently, only a few preferred candidate formulations need to be experimentally verified to obtain the final optimized composite gel formulation.

[0149] In summary, the formula optimization method in this invention integrates a variety of advanced mathematical modeling and machine learning techniques, including principal component analysis, second-order polynomial regression, genetic algorithm, deep neural network, etc. Through the organic combination of these methods, the intrinsic relationship between the various performance indicators of the composite gel and the formula parameters can be fully explored to guide the optimal composite gel formula. This method has strong universality and can be applied not only to sodium fluoride sustained-release gel, but also to the development of other drug sustained-release systems.

[0150] Step S80: Comprehensively evaluate candidate formulations and conduct experimental verification

[0151] In this step, the analytic hierarchy process (AHP) and TOPSIS method will be used to comprehensively evaluate and rank the candidate formulas screened in the previous step. The specific steps are as follows:

[0152] 1. Determine the weight vector of performance indicators :

[0153] Through expert scoring, the pairwise comparison matrix of the relative importance of each performance indicator is determined .

[0154] The indicator weight vector is calculated using the eigenvalue method .

[0155] 2. Use the TOPSIS method to calculate the comprehensive score of each candidate formula :

[0156] Building a standardized decision matrix ; Determine the ideal solution and negative ideal solution ; Calculate the distance of each candidate formula to the ideal solution and the negative ideal solution ; Calculate the comprehensive score based on the distance .

[0157] 3. Based on the comprehensive score Sort the candidate recipes and select the ones with the highest scores for experimental verification.

[0158] Through the AHP and TOPSIS methods, we can find a balance between multiple performance indicators and obtain the candidate formula with the best comprehensive performance. In the future, we only need to conduct experimental verification on these highly ranked candidate formulas to obtain the final optimized composite gel formula.

[0159] Step S90: Experimental verification and model update

[0160] In this step, highly ranked candidate formulations will be prepared and subjected to comprehensive performance testing to evaluate the predictive accuracy of the established formulation optimization model. The specific steps are as follows:

[0161] 1. According to the result of step S80, several candidate formulations with the highest comprehensive scores are selected for experimental preparation.

[0162] 2. The prepared gel samples were tested for their fluoride ion release kinetics, pH responsiveness, mechanical properties and biocompatibility.

[0163] 3. Compare the actual test results with the prediction results of the deep neural network model to evaluate the prediction accuracy of the model. If there is a large error, the model needs to be further optimized and updated.

[0164] 4. Based on the experimental verification results, fine-tune the network structure, hyperparameters, etc. of the deep neural network model, and repeat steps S60-S90 until the final optimized composite gel formula is obtained.

[0165] Through the iterative process of experimental verification and model updating, the prediction accuracy of the formula optimization model can be further improved, and ultimately the optimal composite gel formula that meets the requirements of various performance indicators can be obtained.

[0166] In summary, the composite gel formula optimization method proposed in the patent includes the following key steps:

[0167] 1. Select gel components suitable for sustained release of sodium fluoride, including graphene oxide, acrylamide, acrylic acid, cross-linking agent, etc.

[0168] 2. Design multiple sets of initial formulas, focus on parameters such as graphene oxide concentration, monomer ratio and cross-linker dosage, and test various performance indicators through experiments.

[0169] 3. Use principal component analysis to construct the correlation matrix between formula parameters and performance indicators and determine the main optimization parameters.

[0170] 4. Based on the Box-Behnken experimental design, a second-order polynomial regression model was established to describe the relationship between formulation parameters and performance.

[0171] 5. Use genetic algorithms to perform multi-objective optimization on the initial formula and find the optimal formula range that balances various performances.

[0172] 6. Train a multi-branch deep neural network model as a recipe optimization model to perform virtual screening in a wider parameter space.

[0173] 7. Use AHP and TOPSIS methods to comprehensively evaluate and rank the candidate formulas, and select the preferred formula for experimental verification.

[0174] 8. Based on the experimental verification results, fine-tune and update the deep neural network model, and repeat the optimization iteration until the final formula is obtained.

[0175] This method makes full use of a variety of advanced mathematical modeling and machine learning techniques to achieve intelligent optimization of the composite gel formula, providing an effective technical path for the development of sodium fluoride sustained-release gel with excellent performance.

[0176] In order to better understand and implement the formula optimization method of the composite gel for sustained release of sodium fluoride drugs provided in the first aspect of the present invention, an embodiment 1 of the above-mentioned formula optimization method is provided below. The specific steps of this embodiment 1 are described as follows:

[0177] Step S10: Selecting a gel component suitable for sustained release of sodium fluoride

[0178] In this step, it is necessary to select gel components suitable for sustained release of sodium fluoride. Based on literature research and preliminary experiments, the following key components can be selected:

[0179] 1. Graphene oxide : As one of the main components of the gel matrix, its concentration It is an important formulation parameter, usually selected Graphene oxide can be - It interacts with sodium fluoride to form a complex, thereby controlling the sustained release behavior of the drug.

[0180] 2. Acrylamide and acrylic : As the other two main monomer components of the gel matrix. The molar ratio between them is another important formulation parameter, usually selected The copolymerization of acrylamide and acrylic acid can regulate the mechanical properties and pH responsiveness of the gel.

[0181] 3. Cross-linking agent: N,N-methylenebisacrylamide (MBAA) is a commonly used chemical cross-linking agent that can copolymerize with acrylamide and acrylic acid to form a three-dimensional network structure. The molar percentage of the cross-linking agent It is also an important formulation parameter, usually selected .

[0182] 4. Initiator: Ammonium persulfate (APS) is a commonly used free radical initiator that can promote the polymerization reaction of acrylamide and acrylic acid monomers. The molar percentage of the initiator Need to be controlled within an appropriate range, usually choose .

[0183] 5. pH adjuster: You can choose some weak acid or weak alkaline substances, such as citric acid , sodium dihydrogen phosphate etc. to regulate the pH responsiveness of the gel. The concentration of the pH regulator It is also an important parameter.

[0184] 6. Hydrophilic / lipophilic regulators: Some hydrophilic or lipophilic small molecules or high molecular substances, such as polyethylene glycol (PEG), polypropylene glycol (PPG), etc., can be added to adjust the hydrophilicity and hydrophobicity of the gel.

[0185] 7. Natural polymer materials: such as sodium alginate, chitosan, collagen and hyaluronic acid, can give the gel good biocompatibility and biofunctionality. Their mass fraction are parameters that need to be considered.

[0186] 8. Biofunctional modifiers: such as RGD peptide, cell adhesion factors and growth factors, can further enhance the bioactivity and cell affinity of the gel. Their addition amount It is also an important factor.

[0187] The above 8 types of components are all optional gel formula ingredients in this method. By properly selecting these components and optimizing the ratio, a composite gel system suitable for the sustained release of sodium fluoride can be designed.

[0188] Step S20: Design multiple groups of initial formulations and conduct performance evaluation

[0189] In this step, based on the gel components selected in the above step S10, multiple groups of initial formulas are designed through literature research and preliminary experiments. The focus is on the following three aspects:

[0190] 1. Graphene oxide concentration : Usually choose to explore the scope.

[0191] 2. Monomer ratio of acrylamide and acrylic acid : Usually choose to explore the scope.

[0192] 3. Dosage of cross-linking agent MBAA : Usually choose to explore the scope.

[0193] For the designed initial formula, various performance indicators of the composite gel need to be obtained through experimental means, including:

[0194] 1. Fluoride ion release kinetics :The rate of change of fluoride ion concentration over time ; Initial burst effect ;

[0195] 2. pH responsiveness :Swelling rate With pH Rate of change ;

[0196] 3. Mechanical properties, including: compressive strength ; Young's modulus ;

[0197] 4. Biocompatibility, including: cell viability ;Biodegradation half-life ;

[0198] The measurement results of these performance indicators will provide important basis for subsequent formula optimization.

[0199] Step S30: Constructing a correlation matrix between recipe parameters and performance indicators

[0200] In this step, the principal component analysis (PCA) method will be used to calculate the correlation between each formula parameter and performance index. The specific steps are as follows:

[0201] 1. Organize the performance index data obtained from the experiment into a matrix form .

[0202] 2. Arrange the designed formula parameter data into a matrix form ,in , .

[0203] 3. Calculation and The correlation matrix between : ;

[0204] in, is the number of samples.

[0205] 4. Correlation Matrix Perform eigenvalue decomposition to obtain the principal component weight vector .

[0206] 5. According to the principal component weight vector Determine 5 main formulation parameters, including: graphene oxide concentration ; Acrylamide: acrylic acid ratio ; Dosage of cross-linking agent MBAA ; pH adjuster concentration ; Total content of natural polymer materials ; These main formulation parameters will be focused on in the subsequent optimization process.

[0207] Step S40: Establishing a second-order polynomial model of formulation parameters and performance indicators

[0208] In this step, the Box-Behnken experimental design method will be used to establish various performance indicators. With the main recipe parameters The second-order polynomial regression model between . The specific steps are as follows:

[0209] 1. Based on the five main formulation parameters determined in step S30, design the factor levels of the Box-Behnken test. Usually, Three levels.

[0210] 2. According to the Box-Behnken test design matrix, prepare the corresponding composite gel samples and test various performance indicators .

[0211] 3. Use the least squares method to fit the experimental data and establish a second-order polynomial model:

[0212]

[0213] in, is the regression coefficient, is a random error term. .

[0214] 4. Use stepwise regression method to screen out significant factors and optimize the structure of the second-order polynomial model.

[0215] The second-order polynomial model established by this method can quantitatively describe the relationship between various performance indicators of the composite gel and the main formulation parameters. Subsequent optimization will rely heavily on this model.

[0216] Step S50: Evaluate the accuracy and reliability of the second-order polynomial model

[0217] In this step, the accuracy and reliability of the established second-order polynomial model will be evaluated using the following methods:

[0218] 1. Residual Analysis: Calculate the model prediction value With experimental value The residual between , check whether it follows a normal distribution , verify whether the linear assumption of the model is valid.

[0219] 2. Analysis of variance (ANOVA): Perform significance test on the regression model and evaluate the significance level of each regression coefficient.

[0220] 3. Lack-of-Fit Test: Test whether the model fits well, that is, whether there are systematic errors in the model.

[0221] Through the above analysis, the accuracy and reliability of the established second-order polynomial model can be comprehensively evaluated. If the model passes these tests, it can be considered that it can better describe the relationship between formulation parameters and performance indicators.

[0222] Afterwards, a genetic algorithm will be used to perform multi-objective optimization on these initial formulas to find the optimal formula range that balances various performances. The fitness function of the genetic algorithm uses the weighted sum of various performance indicators predicted by the second-order polynomial model:

[0223] ;

[0224] in, They are , is the indicator weight vector.

[0225] Step S60: Training a deep neural network model for formula optimization

[0226] In this step, a multi-branch deep neural network model is trained based on the composite gel performance index data obtained from the experiment to predict the performance of the composite gel. Specifically, it includes the following three branch networks:

[0227] 1. Fluoride ion release branch network:

[0228] Input: Recipe parameters and pH ;

[0229] Output: Release curve parameters and initial release ;

[0230] Network structure: It consists of three layers of convolutional neural network and two layers of fully connected layers;

[0231] 2. Mechanical properties branch network:

[0232] Input: Recipe parameters and crosslink density ;

[0233] Output: Compression Strength and Young's modulus ;

[0234] Network structure: A feed-forward neural network consisting of four fully connected layers;

[0235] 3. Biocompatibility branch network:

[0236] Input: Recipe parameters and biofunctional modifier types ;

[0237] Output: Cell viability and degradation half-life ;

[0238] Network structure: A recurrent neural network consisting of a long short-term memory (LSTM) layer and a fully connected layer;

[0239] This multi-branch structure can better capture the complex relationship between the various performance indicators of the composite gel. After training, the obtained deep neural network model is the formula optimization model, which can be used to guide the subsequent virtual screening of a wider space.

[0240] Step S70: Virtual screening based on deep neural network model

[0241] In this step, the deep neural network model trained in step S60 is used to perform virtual screening in a wider formulation parameter space to find the optimal composite gel formulation. The specific steps are as follows:

[0242] 1. Using the Monte Carlo simulation method, a large number of candidate recipes are randomly generated within the range of recipe parameters. .

[0243] 2. Input these candidate recipes into the trained deep neural network model to predict their performance indicators .

[0244] 3. Use the Analytic Hierarchy Process (AHP) to determine the weight vector of each performance indicator .

[0245] 4. Use the TOPSIS method (Technique for Order of Preference by Similarity to Ideal Solution) to calculate the comprehensive score of each candidate formula :

[0246] Construct a standardized decision matrix:

[0247] ;

[0248] Determine the ideal solution:

[0249] ;

[0250] Determine the negative ideal solution:

[0251] ;

[0252] Compute the Euclidean distance of each candidate formulation to the ideal solution and the negative ideal solution:

[0253] and ;

[0254] Calculate the comprehensive score based on the distance ;

[0255] 5. Based on the comprehensive score Sort the candidate recipes and select the ones with the highest scores for experimental verification.

[0256] Through the AHP and TOPSIS methods, we can find a balance between multiple performance indicators and obtain the candidate formula with the best comprehensive performance. In the future, we only need to conduct experimental verification on these highly ranked candidate formulas to obtain the final optimized composite gel formula.

[0257] Step S80: Experimental verification and model update

[0258] In this step, highly ranked candidate formulations will be prepared and subjected to comprehensive performance testing to evaluate the predictive accuracy of the established formulation optimization model. The specific steps are as follows:

[0259] 1. According to the result of step S70, select several candidate formulas with the highest comprehensive scores Prepare for the experiment.

[0260] 2. Test the fluoride ion release kinetics of these prepared gel samples , pH responsiveness , Mechanical properties and biocompatibility And other indicators.

[0261] 3. Compare the actual test results with the prediction results of the deep neural network model Compare and evaluate the prediction accuracy of the model. If there is a large error, the model needs to be further optimized and updated.

[0262] 4. Based on the experimental verification results, fine-tune the network structure, hyperparameters, etc. of the deep neural network model, and repeat steps S60-S80 until the final optimized composite gel formula is obtained.

[0263] Through the iterative process of experimental verification and model updating, the prediction accuracy of the formula optimization model can be further improved, and ultimately the optimal composite gel formula that meets the requirements of various performance indicators can be obtained.

[0264] In general, this embodiment 1 specifically includes the following key steps:

[0265] 1. Select 8 types of components including graphene oxide, acrylamide, acrylic acid, cross-linking agent, etc. as the components of the gel system.

[0266] 2. Design multiple initial formulas, focusing on graphene oxide concentration , acrylamide to acrylic acid ratio , Dosage of cross-linking agent MBAA And other parameters, and test various performance indicators.

[0267] 3. Use principal component analysis to construct the correlation matrix between formulation parameters and performance indicators , determine the five main optimization parameters.

[0268] 4. Based on the Box-Behnken experimental design, a second-order polynomial regression model was established.

[0269] 5. Use genetic algorithm for multi-objective optimization, the objective function is the weighted sum of various performance indicators predicted by the second-order polynomial model .

[0270] 6. Train a multi-branch deep neural network model as a recipe optimization model to perform virtual screening in a wider parameter space.

[0271] 7. Use AHP and TOPSIS methods to comprehensively evaluate and rank the candidate formulas, and select the preferred formula for experimental verification.

[0272] 8. Based on the experimental verification results, fine-tune and update the deep neural network model, and iterate and optimize until the final formula is obtained.

[0273] A second aspect of the present invention provides a method for preparing a composite gel for sustained release of sodium fluoride drugs, wherein the method for preparing the composite gel for sustained release of sodium fluoride drugs after optimizing the formula of the composite gel for sustained release of sodium fluoride drugs described above comprises the following steps:

[0274] Step 1. 1 to 5 mg / mL of graphene oxide is uniformly dispersed in deionized water to prepare a graphene oxide dispersion.

[0275] Step 2. Add acrylamide and acrylic acid monomers to the graphene oxide dispersion in a molar ratio of 1:1 to 3:1, and disperse under ultrasonication for 20 minutes to mix the solution evenly.

[0276] Step 3: The obtained gel solution is degassed by passing nitrogen gas for 3 to 5 minutes.

[0277] Step 4. Pour the degassed gel solution into a mold, add 0.01 to 0.03 mol of N,N-methylenebisacrylamide crosslinker and 0.005 to 0.01 g of ammonium persulfate initiator, and then heat the reaction in a water bath at 40 to 60 degrees Celsius to obtain a polyacrylamide composite hydrogel.

[0278] Step 5. Soak the obtained composite hydrogel in a 0.1 mol / L sodium fluoride standard solution for 24 hours to obtain a composite gel loaded with sodium fluoride drugs.

[0279] The preparation method uses graphene oxide as a gel matrix to form a composite hydrogel with acrylamide-acrylic acid copolymer.

[0280] Furthermore, the concentration of graphene oxide in step 1 is 1-5 mg / ml.

[0281] Furthermore, in step 2, the molar ratio of acrylamide monomer to acrylic acid monomer is 1:1 to 3:1.

[0282] Furthermore, the nitrogen introduction time in step 3 is 3 to 5 minutes.

[0283] Furthermore, in step 4, the molar mass of the cross-linking agent is 0.01 to 0.03 mol grams per mol, the molar mass of the initiator is 0.005 to 0.01 grams per mol, and the water bath temperature is 40 degrees Celsius to 60 degrees Celsius.

[0284] Furthermore, the concentration of the sodium fluoride standard solution in step 5 is 0.1 mol / L.

[0285] The preparation method uses graphene oxide as a gel matrix to form a composite hydrogel with acrylamide-acrylic acid copolymer, and loads sodium fluoride drugs through the leaching method. By optimizing various process parameters, such as graphene oxide content, monomer ratio, crosslinking agent and initiator dosage, a sodium fluoride drug sustained-release composite gel with excellent performance can be prepared.

[0286] Specifically, the principle of the present invention is:

[0287] 1. Select appropriate gel components. Based on literature research and preliminary experiments, eight types of compounds, including graphene oxide, acrylamide, acrylic acid, crosslinkers, initiators, pH regulators, hydrophilic / lipophilic regulators, natural polymer materials, and biofunctional modifiers, were selected as components of the gel. Among them, graphene oxide is the main component of the gel matrix and can form a complex with fluoride ions to control the sustained release behavior of the drug; acrylamide and acrylic acid copolymers give the gel good swelling properties and pH responsiveness, which is conducive to the sustained and slow release of sodium fluoride; other ingredients such as crosslinkers and natural polymer materials further improve the mechanical properties and biocompatibility of the gel. By rationally matching these components, a composite hydrogel that meets the sustained release requirements of sodium fluoride can be designed.

[0288] 2. Use multivariate optimization method to optimize the formulation parameters. First, the principal component analysis method was used to determine the five main parameters that affect the gel performance, including graphene oxide concentration, monomer ratio of acrylamide to acrylic acid, crosslinker dosage, pH regulator concentration and natural polymer material content. Then, the Box-Behnken experimental design was used to establish a second-order polynomial regression model between various performance indicators and these main parameters, and the accuracy and reliability of the model were evaluated by residual analysis, variance analysis and other methods. Next, a genetic algorithm was used to perform multi-objective optimization on the initial formula, and the performance indicators predicted by the second-order polynomial model were used as the objective function to find the optimal formula range that balances various performances.

[0289] In addition, the present invention also trains a multi-branch deep neural network model as a formula optimization tool. The model includes three branch networks of fluoride ion release, mechanical properties and biocompatibility, which can more accurately predict the key properties of the composite gel. In a wider parameter space, Monte Carlo simulation is used to generate a large number of candidate formulas, and the trained deep neural network model is used for rapid prediction. The candidate formulas are then comprehensively evaluated and ranked in combination with the hierarchical analysis method and the TOPSIS method, so as to find the optimal formula that meets various performance requirements.

[0290] Finally, through the iterative process of experimental verification and model updating, the prediction accuracy of the formula optimization model was further improved, ensuring that the final optimized formula had excellent fluoride ion sustained-release properties, good biocompatibility, and controllable mechanical strength, and could exert ideal therapeutic effects in the oral environment.

[0291] In general, the composite gel preparation method proposed in the present invention makes full use of the synergistic effect of materials such as graphene oxide and acrylamide-acrylic acid copolymer, and realizes intelligent optimization of the formula through mathematical modeling and machine learning technology, thereby developing a composite gel system that integrates drug sustained release, biocompatibility and mechanical strength, providing an innovative solution for local drug delivery of oral diseases.

[0292] In order to further better understand and implement the present invention, Example 2 of a specific application scenario of the above-mentioned formula optimization method is provided below: This Example 2 adopts the formula optimization method proposed in the present invention to obtain a sodium fluoride sustained-release composite gel with excellent performance.

[0293] First, for the 8 types of gel components described in the present invention, the following preliminary selection and setting were carried out:

[0294] 1. Graphene oxide : Graphene oxide with a higher degree of oxidation was selected and the concentration was explored in the range of 1-5 wt%.

[0295] 2. Acrylamide and acrylic : Use analytically pure monomers, monomer molar ratio Optimize between 1:1 and 3:1.

[0296] 3. Cross-linking agent: N,N-methylenebisacrylamide (MBAA) is used, the dosage is The concentration range was 1-3 mol%.

[0297] 4. Initiator: Use ammonium persulfate (APS), dosage Controlled at 0.5-1 mol%.

[0298] 5. pH adjuster: choose citric acid and sodium dihydrogen phosphate mixture, concentration The optimization was performed in the range of 0.01-0.1 mol / L.

[0299] 6. Hydrophilic / lipophilic regulator: A mixture of polyethylene glycol (PEG) and polypropylene glycol (PPG) is selected.

[0300] 7. Natural polymer materials: Sodium alginate, chitosan, collagen and hyaluronic acid, with a total mass fraction of The range of 5-20 wt% was explored.

[0301] 8. Biofunctional modifiers: RGD peptide, cell adhesion factor and growth factor, dosage The optimization was performed in the range of 0.1-1wt%.

[0302] Next, based on the above preliminary selection, 16 groups of composite gel samples with different formulations were designed, and the corresponding formulation parameters are shown in Table 1. These initial formulations mainly focus on exploring the effects of key factors such as graphene oxide concentration, monomer ratio, and cross-linking agent dosage on gel properties.

[0303] Table 1 Recipe parameters

[0304]

[0305] The 16 groups of initial formula composite gel samples were comprehensively tested for performance, including fluoride ion release kinetics, pH responsiveness, mechanical properties, and biocompatibility. The test results are shown in Table 2.

[0306] Table 2 Test results

[0307]

[0308] Based on the above experimental data, the principal component analysis method was used to calculate the correlation matrix between each formulation parameter and performance index. According to the correlation matrix, the concentration of graphene oxide , the ratio of acrylamide to acrylic acid , the amount of cross-linking agent MBAA , pH adjuster concentration And the total content of natural polymer materials These are the five main parameters that affect the performance of the composite gel. These parameters will be the focus of subsequent optimization.

[0309] Next, based on the Box-Behnken experimental design, a second-order polynomial regression model was established between each performance index and the five main parameters. The accuracy and reliability of the model were verified through residual analysis, variance analysis and lack of fit test. For example, its regression model can be expressed as:

[0310] ;

[0311] in, is a random error term, which follows a normal distribution .

[0312] Using this second-order polynomial model, a genetic algorithm was used to optimize the initial formula for multiple objectives. The objective function is the weighted sum of various performance indicators:

[0313] ;

[0314] After multiple rounds of iterative optimization, the formula range with the best comprehensive performance was finally obtained, specifically:

[0315] ;

[0316] ;

[0317] ;

[0318] ;

[0319] ;

[0320] ;

[0321] In order to further improve the optimization efficiency, a multi-branch deep neural network model was trained as a formula optimization tool. The model includes three branch networks: fluoride ion release, mechanical properties, and biocompatibility, which can more accurately predict the key performance indicators of the composite gel.

[0322] Based on this deep neural network model, a large number of virtual screenings were carried out in a wider parameter space. First, the Monte Carlo simulation method was used to generate 10,000 random candidate formulations. Then, the trained deep neural network model was used to quickly predict the performance indicators of these candidate formulations. Finally, the hierarchical analysis method and TOPSIS method were used to comprehensively consider the relative importance of various performance indicators and rank and evaluate these candidate formulations.

[0323] After this series of optimization steps, three candidate formulations with the best comprehensive performance were finally selected for experimental verification. The specific parameters are as follows:

[0324] Recipe 1: , , , , , ;

[0325] Recipe 2: , , , , , ;

[0326] Recipe 3: , , , , , ;

[0327] The three optimal formulas were prepared into gel samples and subjected to comprehensive performance tests. The experimental results showed that the performance indicators of these optimized formulas, such as fluoride ion release kinetics, pH responsiveness, mechanical strength and biocompatibility, were better than those of the 16 initially designed formulas. At the same time, the experimental data also matched well with the prediction results of the deep neural network model, verifying the accuracy of the optimization model.

[0328] Through the above-mentioned multi-step optimization method, an optimized composite gel formula that meets various performance requirements was finally obtained, providing an effective technical path for the development of sodium fluoride sustained-release gel with excellent comprehensive performance.

[0329] Example 3, a method for preparing a composite hydrogel for sustained release of sodium fluoride drugs: In this example 3, based on the optimal formula determined in example 2, a composite hydrogel for sustained release of sodium fluoride drugs was prepared by the following steps:

[0330] 1. Disperse 3.8 g of graphene oxide in 100 mL of deionized water and ultrasonically disperse for 20 min to obtain a uniform graphene oxide dispersion.

[0331] 2. Add 8.6 g of acrylamide monomer and 3.7 g of acrylic acid monomer to the above graphene oxide dispersion at a molar ratio of 2.3:1. Place the mixture in an ultrasonic cleaner under magnetic stirring and perform ultrasonic treatment for 20 min to obtain a uniform gel precursor solution.

[0332] 3. Pass nitrogen into the prepared gel precursor solution and degas it for 4 minutes.

[0333] 4. Slowly pour the degassed gel solution into a polytetrafluoroethylene mold, add 2.2 g N,N-methylenebisacrylamide (MBAA) crosslinker and 0.007 g ammonium persulfate (APS) initiator. Then put the mold containing the gel solution into a 50°C water bath and react at constant temperature for 2 h to obtain a polyacrylamide composite hydrogel.

[0334] 5. Soak the prepared polyacrylamide-graphene oxide composite hydrogel in 0.1 mol / L sodium fluoride solution and let it stand for 24 hours to allow it to fully adsorb and load the sodium fluoride drug to obtain the final sodium fluoride sustained-release composite gel.

[0335] In the above preparation method, the amount of graphene oxide is 3.8 wt%, the molar ratio of acrylamide to acrylic acid is 2.3:1, the amount of crosslinker MBAA is 2.2 g (about 2.2 mol%), and the amount of initiator APS is 0.007 g (about 0.7 mol%). The polyacrylamide-graphene oxide composite hydrogel is prepared by reacting at 50 °C for 2 h using a water bath method, and then loaded with sodium fluoride drug by immersion method.

[0336] The key features of this preparation method are as follows:

[0337] 1. Using graphene oxide as a gel matrix, it can form a - The complex is beneficial for controlling the slow release of sodium fluoride.

[0338] 2. The 2.3:1 molar ratio of acrylamide and acrylic acid monomers gives the gel excellent pH responsiveness and swelling properties, which helps to achieve sustained release of sodium fluoride.

[0339] 3. MBAA chemical crosslinking and APS free radical initiation, reacting at 50 °C for 2 h, can prepare a composite hydrogel with good mechanical strength.

[0340] 4. The prepared hydrogel was loaded with sodium fluoride drug using the immersion method, and the adsorption and swelling properties of the gel were utilized to achieve effective loading of the drug.

[0341] Through the above preparation method, a sodium fluoride sustained-release system based on graphene oxide / polyacrylamide composite hydrogel was successfully prepared. The composite gel has excellent fluoride ion sustained-release performance, good biocompatibility and controllable mechanical strength, and can meet the application requirements in the oral environment.

[0342] In order to verify the performance of the prepared composite gel, a series of tests and characterizations were carried out:

[0343] 1. Fluoride ion release kinetics: The ion selective electrode method was used to measure the cumulative release of fluoride ions over time. The results showed that the composite gel can achieve sustained and slow release of fluoride ions, with about 85% released within 24 hours, which meets the ideal sustained release kinetic characteristics.

[0344] 2. pH responsiveness: The swelling experiment was used to measure the changes in the swelling rate of the gel under different pH conditions. The results showed that the composite gel had good pH responsiveness. In the pH range of 5.0-8.0, the swelling rate increased linearly with the increase of pH. It is about 15% / pH unit, which is beneficial for pH regulation in the oral environment.

[0345] 3. Mechanical properties: Through compression test, the compression strength of the composite gel was measured to be 0.72 MPa and the Young's modulus was 19.4 MPa. These mechanical parameters meet the requirements for use in the oral environment and can withstand normal chewing pressure.

[0346] 4. Biocompatibility: The biocompatibility of the composite gel was evaluated by MTT cytotoxicity test and in vitro degradation test. The results showed that the gel had low toxicity to human dental pulp fibroblasts, with a cell survival rate of more than 90%; in simulated saliva solution, the gel could be slowly degraded with a half-life of about 15 days, which met the biosafety requirements in the oral environment.

[0347] In summary, this Example 3 successfully prepared a sodium fluoride sustained-release system based on graphene oxide / polyacrylamide composite hydrogel. While the composite gel has the characteristics of excellent drug sustained-release performance, good biocompatibility and controllable mechanical strength, the preparation process is also relatively simple, providing an innovative solution for local drug delivery of oral diseases.

[0348] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for optimizing the formulation of a composite gel for sustained release of sodium fluoride drugs, characterized in that: The following steps are involved: S10, selecting gel components suitable for sustained release of sodium fluoride, including: graphene oxide, acrylamide, acrylic acid, a cross-linking agent, an initiator, a pH regulator, a hydrophilic / lipophilic regulator, a natural polymer material, and a biofunctional modifier; S20. Design multiple initial formulations based on literature research and preliminary experiments; focus on graphene oxide concentration, acrylamide: acrylic acid monomer ratio and crosslinker dosage; obtain performance indicators of composite gels through experiments, including fluoride ion release kinetics, initial burst effect, pH responsiveness, mechanical strength and biocompatibility; S30, using the principal component analysis method, calculating the correlation between each component and the performance index, constructing a correlation matrix and determining the main formula parameters, wherein the main formula parameters represent the graphene oxide concentration, acrylamide: acrylic acid ratio, crosslinking agent dosage, pH regulator concentration and natural polymer material content; S40, using Box-Behnken design based on the correlation matrix to establish a second-order polynomial model between each performance index and the main formulation parameters; using stepwise regression method to screen significant factors and optimize the structure of the second-order polynomial model; S50, evaluating the accuracy and reliability of the second-order polynomial model through residual analysis, variance analysis and lack of fit test; using a genetic algorithm to perform multi-objective optimization on the multiple groups of initial formulas to find the optimal formula interval that balances various performances; S60. Based on the performance indicators of the composite gel obtained from the experiment, a deep neural network model with a multi-branch structure is trained to obtain a formula optimization model; S70, using the recipe optimization model to perform virtual screening in a wider recipe parameter space; using a Monte Carlo simulation method to generate a large number of candidate recipes, and using a trained deep neural network model to predict the performance of the candidate recipes; S80. Use the analytic hierarchy process and TOPSIS method to comprehensively consider the importance of various performance indicators, and comprehensively score and rank the candidate formulas; select several candidate formulas with the highest scores for experimental verification; S90, prepare highly ranked candidate formulations and conduct comprehensive performance tests; compare actual performance with predicted results to evaluate the accuracy of the formulation optimization model; fine-tune and update the formulation optimization model based on the verification results; repeat steps S60-S90 until the final optimized formulation is obtained.

2. The method for optimizing the formula of a composite gel for sustained release of sodium fluoride drugs according to claim 1, characterized in that: The fitness function of the genetic algorithm adopts the weighted sum of various performance indicators predicted by the second-order polynomial model.

3. The method for optimizing the formula of a composite gel for sustained release of sodium fluoride drugs according to claim 1, characterized in that: The cross-linking agent is N,N-methylenebisacrylamide; the initiator is ammonium persulfate; the natural polymer material includes sodium alginate, chitosan, collagen and hyaluronic acid; the biofunctional modifier includes RGD peptide, cell adhesion factor and growth factor.

4. The method for optimizing the formula of a composite gel for sustained release of sodium fluoride drugs according to claim 1, characterized in that: The multi-branch structured deep neural network model includes a fluoride ion release branch network, a mechanical property branch network and a biocompatibility branch network.

5. The method for optimizing the formula of a composite gel for sustained release of sodium fluoride drugs according to claim 4, characterized in that: The fluoride ion release branch network is used to predict the fluoride ion release kinetics and initial burst effect. The input is the formulation parameters and pH value, and the output is the release curve parameters and the initial release amount. The structure consists of a three-layer convolutional neural network and two fully connected layers. The mechanical property branch network is used to predict the mechanical strength and elastic modulus of the gel, the input is the formula parameters and the cross-linking density, the output is the compressive strength and Young's modulus, and the structure is a feedforward neural network consisting of four fully connected layers; The biocompatibility branch network is used to predict the cytotoxicity and biodegradability of the gel. The input is the formulation parameters and the type of biofunctional modifiers. The output is the cell survival rate and degradation half-life. The structure is a recurrent neural network composed of long short-term memory layers and fully connected layers.

6. A method for preparing a composite gel for sustained release of sodium fluoride drugs, characterized in that: The preparation of a composite gel for sustained release of sodium fluoride drugs after optimizing the formula of the composite gel for sustained release of sodium fluoride drugs according to any one of claims 1 to 5 comprises the following steps: uniformly dispersing graphene oxide in deionized water to obtain a graphene oxide dispersion; Adding acrylic acid and acrylamide monomers to the graphene oxide dispersion, and performing ultrasonic dispersion to mix the solution uniformly to obtain a gel solution; Degassing the gel solution; Adding a crosslinking agent and an initiator to the degassed gel solution, and heating the solution in a water bath to obtain a polyacrylamide composite hydrogel; The polyacrylamide composite hydrogel is immersed in a sodium fluoride solution to obtain a composite hydrogel loaded with sodium fluoride medicine.

7. The method for preparing a composite gel for sustained release of sodium fluoride drugs according to claim 6, characterized in that: The concentration of the graphene oxide is 1-5 wt %.

8. The method for preparing a composite gel for sustained release of sodium fluoride drugs according to claim 7, characterized in that: The molar ratio of the acrylamide monomer to the acrylic acid monomer is 1:1 to 3:

1.

9. The method for preparing a composite gel for sustained release of sodium fluoride drugs according to claim 8, characterized in that: The nitrogen gas introduction time is 3 to 5 minutes.

10. The method for preparing a composite gel for sustained release of sodium fluoride drugs according to claim 9, characterized in that: The water bath temperature is 40°C to 60°C.

Citation Information

Cited By

  • Method for optimizing formula of medicament for preventing and treating powdery mildew

    CN119864122A

  • Preparation method of insecticide synergistic responsive release delivery system

    CN120748553A