Coenzyme Q10 with effect of protecting heart and cerebral vessels and preparation method of coenzyme Q10

Abstract

The invention relates to the technical field of drug targeting by using structural data, in particular to coenzyme Q10 with a heart and cerebral vessel protection effect and a preparation method thereof.The coenzyme Q10 is obtained by proportioning multiple types of microcapsules, one type of microcapsule is correspondingly prepared according to one target preparation scheme, and the coenzyme Q10 is prepared according to the target preparation scheme. The microcapsule is prepared from the following raw materials in parts by weight: 45-50 parts of coenzyme Q10; 20 to 30 parts of natural vitamin E; 400 to 450 parts of perilla seed oil or walnut oil; 20 to 30 parts of vitamin C; and 80-120 parts of a forming agent. The microcapsule is prepared from the coenzyme Q10 raw material with the effect of protecting heart and cerebral vessels according to the proportion, and on the basis of historical targeting data and human body release curves of various types of microcapsules, the microcapsule proportion corresponding to the target targeting data is dynamically determined by utilizing the release response characteristics of different types of microcapsules in a human body, so that the target targeting data is obtained. Intelligent personalized customization can be better achieved, multi-stage release behaviors are dynamically adjusted according to different focus parts, and the targeted delivery effect is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of drug targeting using structural data in bioinformatics, and particularly relates to coenzyme Q10 with a cardio-cerebrovascular protective effect and a preparation method thereof. Background Art

[0002] Coenzyme Q10 is a fat-soluble antioxidant and also a protease, and is one of the essential elements in the human body. It can strengthen the metabolism of myocardial and brain cells, protect the biomembrane of cardio-cerebrovascular cells from damage, and improve the stability of heart rate; reduce the damage of harmful free radicals to the inner wall of blood vessels, and play a role in preventing and slowing down atherosclerosis.

[0003] In the prior art, a fixed formula is mostly used to directly prepare coenzyme Q10 preparations. The coenzyme Q10 preparations can only provide a fixed release mode (such as rapid release or sustained release), and it is difficult to dynamically adjust the multi-stage release behavior for different lesion sites (such as coronary arteries and cerebral microvessels), resulting in poor targeted delivery effect. Summary of the Invention

[0004] The embodiment of the present invention provides a preparation method of coenzyme Q10 with a cardio-cerebrovascular protective effect, aiming to solve the problem that the existing coenzyme Q10 preparations can only provide a fixed release mode, it is difficult to dynamically adjust the multi-stage release behavior for different lesion sites, and the targeted delivery effect is poor. The present invention prepares coenzyme Q10 raw materials with a cardio-cerebrovascular protective effect and their ratios into microcapsules. Based on historical targeted data and the human release curves of various types of microcapsules, the release response characteristics of different types of microcapsules in the human body are utilized to dynamically determine the microcapsule ratio corresponding to the target targeted data, which can better achieve intelligent and personalized customization, dynamically adjust the multi-stage release behavior for different lesion sites, and improve the targeted delivery effect of coenzyme Q10.

[0005] In a first aspect, the embodiment of the present invention provides coenzyme Q10 with a cardio-cerebrovascular protective effect, which is obtained by mixing multiple types of microcapsules. One type of microcapsule is prepared by a corresponding preparation method of the target preparation scheme. The microcapsule is prepared from the following raw materials and ratios by weight: Coenzyme Q10: 45 - 50 parts; Natural vitamin E: 20 - 30 parts; Perilla seed oil or walnut oil: 400 - 450 parts; Vitamin C: 20 - 30 parts; Molding agent: 80 - 120 parts.

[0006] Further, the dosage form of the coenzyme Q10 with a cardio-cerebrovascular protective effect includes any one of granule, powder, and capsule.

[0007] In a second aspect, an embodiment of the present invention provides a method for preparing coenzyme Q10 with a cardiovascular and cerebrovascular protective effect, which is used to prepare coenzyme Q10 with a cardiovascular and cerebrovascular protective effect as described in any one of the embodiments of the present invention. The preparation method includes: Within the raw material ratio range, multiple types of microcapsules are prepared according to multiple target preparation schemes, and one type of microcapsule is prepared corresponding to one of the target preparation schemes; According to historical targeting data and the human release curves of each type of microcapsule, the microcapsule ratio corresponding to the target targeting data is determined; According to the microcapsule ratio, the coenzyme Q10 with a cardiovascular and cerebrovascular protective effect corresponding to the target dosage form is prepared.

[0008] Furthermore, before the step of preparing multiple types of microcapsules according to multiple target preparation schemes within the raw material ratio range, the preparation method further includes: Within the raw material ratio range, all candidate types of microcapsules are prepared through candidate preparation schemes; Under the human digestive environment, the candidate human release curves corresponding to each candidate type are determined, and one candidate type corresponds to one candidate human release curve; Based on the candidate human release curves, the target preparation scheme is determined among all the preparation schemes, and among all the candidate human release curves, the human release curve corresponding to the target preparation scheme is determined.

[0009] Furthermore, the step of determining the candidate human release curves corresponding to each candidate type under the human digestive environment specifically includes: Under the simulated human digestive environment, the digestion simulation of all candidate types of microcapsules is carried out to obtain the simulated release data corresponding to all candidate types of microcapsules; According to the simulated release data, the first candidate human release curves corresponding to each candidate type are determined, and one candidate type corresponds to one first candidate human release curve; Based on the first candidate human release curves corresponding to each candidate type, the candidate human release curves corresponding to each candidate type are determined.

[0010] Furthermore, the step of determining the candidate human release curves corresponding to each candidate type under the human digestive environment specifically includes: Under the clinical trial environment, the clinical trials of all candidate types of microcapsules are carried out to obtain the clinical release data corresponding to all candidate types of microcapsules; Based on the clinical release data, determine the second candidate human release curve corresponding to each of the candidate types, with one second candidate human release curve corresponding to one candidate type; Based on the second candidate human release curve, determine the candidate human release curve corresponding to each of the candidate types; Alternatively, based on the first candidate human release curve and the second candidate human release curve, determine the candidate human release curve corresponding to each of the candidate types.

[0011] Further, the step of determining the candidate human release curve corresponding to each of the candidate types based on the first candidate human release curve and the second candidate human release curve specifically includes: Calculate the first correlation degree between the first candidate human release curve and the second candidate human release curve of the same candidate type; If the first correlation degree is greater than or equal to the first correlation degree threshold, then fit the first candidate human release curve and the second candidate human release curve to obtain the candidate human release curve corresponding to the candidate type; If the first correlation degree is less than the first correlation degree threshold, then keep the second candidate human release curve unchanged as the benchmark, and perform fitting iteration on the fitting curve between the first candidate human release curve and the second candidate human release curve and the second candidate human release curve. When the fitting iteration reaches the number of iterations, determine the fitting curve obtained in the last fitting iteration as the candidate human release curve corresponding to the candidate type, and the number of iterations is determined according to the difference between the first correlation degree and the first correlation degree threshold.

[0012] Further, the step of determining the target preparation plan among all the preparation plans and determining the human release curve corresponding to the target preparation plan among all the candidate human release curves based on the candidate human release curve specifically includes: Extract the N-dimensional features of the candidate human release curve, where N is an integer greater than 1; Cluster all the candidate human release curves according to the N-dimensional features to obtain N clustering clusters, and each clustering cluster includes at least one candidate human release curve; For each clustering cluster, determine the preparation plans corresponding to the M candidate human release curves closest to the cluster center as candidate preparation plans, where M is an integer greater than or equal to 1; Remove duplicates from all the candidate preparation plans, and determine the de-duplicated candidate preparation plans as the target preparation plan, and determine the candidate human release curve corresponding to the target preparation plan as the human release curve corresponding to the target preparation plan.

[0013] Further, the steps of determining the microcapsule ratio corresponding to the target targeting data according to the historical targeting data and the human release curves of each type of the microcapsules specifically include: Based on the historical targeting data and the human release curves of each type of the microcapsules, construct a first mapping model between the targeting data and the microcapsule ratio, where the first mapping model includes a mapping path between the targeting data and the microcapsule ratio, and each mapping path corresponds to a mapping weight; According to the collected real-time process parameters, determine the real-time dynamic path adjustment parameters, and adjust the mapping weights corresponding to each of the mapping paths in the first mapping model according to the real-time dynamic path adjustment parameters to obtain a second mapping model; According to the second mapping model, determine the microcapsule ratio corresponding to the target targeting data.

[0014] Further, the steps of constructing a first mapping model between the targeting data and the microcapsule ratio based on the historical targeting data and the human release curves of each type of the microcapsules specifically include: Based on the human release curves of each type of the microcapsules, determine the comprehensive human release curves corresponding to different microcapsule ratios; Determine the coenzyme Q10 absorption curve required for each of the historical targeting data; Calculate a second correlation degree between the comprehensive human release curve and the coenzyme Q10 absorption curve; Associate the comprehensive human release curves with the coenzyme Q10 absorption curves whose second correlation degree is greater than or equal to a second correlation degree threshold to obtain an association relationship between the comprehensive human release curve and the coenzyme Q10 absorption curve; Based on the association relationship between the comprehensive human release curve and the coenzyme Q10 absorption curve, determine the mapping path between the microcapsule ratio and the historical targeting data, and determine the mapping weight of the corresponding mapping path according to the second correlation degree to obtain a first mapping model between the microcapsule ratio and the historical targeting data.

[0015] In the embodiments of the present invention, coenzyme Q10 with the function of protecting the cardiovascular and cerebrovascular system is obtained by mixing multiple types of microcapsules. One type of microcapsule is prepared corresponding to one of the target preparation methods. The microcapsule is prepared from the following raw materials and their ratios by weight: coenzyme Q10: 45 - 50 parts; natural vitamin E: 20 - 30 parts; perilla seed oil or walnut oil: 400 - 450 parts; vitamin C: 20 - 30 parts; molding agent: 80 - 120 parts. By preparing the raw materials and ratios of coenzyme Q10 with the function of protecting the cardiovascular and cerebrovascular system into microcapsules, based on historical targeting data and the human release curves of each type of microcapsule, the release response characteristics of different types of microcapsules in the human body are utilized to dynamically determine the microcapsule ratio corresponding to the target targeting data, which can better achieve intelligent and personalized customization, dynamically adjust the multi-stage release behavior for different lesion sites, and improve the targeted delivery effect of coenzyme Q10. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0017] Figure 1 is a flowchart of a method for preparing coenzyme Q10 with the function of protecting the cardiovascular and cerebrovascular system provided by the embodiments of the present invention; Figure 2 is a flowchart of a method for obtaining the human release curves of each type of microcapsule in the embodiments of the present invention; Figure 3 is a specific flowchart of step S50 in the embodiments of the present invention; Figure 4 is another specific flowchart of step S50 in the embodiments of the present invention; Figure 5 is a specific flowchart of step S57 in the embodiments of the present invention; Figure 6 is a specific flowchart of step S60 in the embodiments of the present invention; Figure 7 is a specific flowchart of step S20 in the embodiments of the present invention; Figure 8 is a specific flowchart of step S21 in the embodiments of the present invention. DETAILED IMPLEMENTATION METHODS

[0018] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0019] An embodiment of the present invention provides coenzyme Q10 with a function of protecting the cardiovascular and cerebrovascular system, which is obtained by mixing multiple types of microcapsules. One type of microcapsule is prepared by a corresponding target preparation scheme. The microcapsule is prepared from the following raw materials and ratios by weight: Coenzyme Q10: 45 - 50 parts; Natural vitamin E: 20 - 30 parts; Perilla seed oil or walnut oil: 400 - 450 parts; Vitamin C: 20 - 30 parts; Forming agent: 80 - 120 parts.

[0020] Among them, coenzyme Q10 is a fat-soluble antioxidant and also a protease, and is one of the important elements indispensable to the human body. It can strengthen the metabolism of myocardial and brain cells, protect the biomembrane of cardiovascular and cerebrovascular cells from damage, and improve the stability of the heart rate; reduce the damage of harmful free radicals to the inner wall of blood vessels, and play a role in preventing and alleviating atherosclerosis.

[0021] Coenzyme Q10 achieves the function of protecting the cardiovascular and cerebrovascular system in the following aspects: Improving heart failure: Coenzyme Q10 is a key component of the mitochondrial electron transport chain and participates in the generation of ATP (adenosine triphosphate). The energy metabolism of myocardial cells in heart failure patients is damaged. Coenzyme Q10 can enhance the energy production of myocardial cells, improve heart function, and enhance the systolic and diastolic functions of the heart by improving the energy supply of myocardial cells.

[0022] Improving angina pectoris: Supplementing coenzyme Q10 can improve the hypoxia tolerance of myocardial cells, provide energy for the myocardium, relieve ischemic myocardial symptoms, and effectively relieve exertional angina pectoris, etc.

[0023] Dilating blood vessels: Coenzyme Q10 can promote the synthesis and release of nitric oxide (NO). NO is a vasodilator produced by endothelial cells, which can promote the relaxation of vascular smooth muscle, reduce vascular resistance, dilate blood vessels, and increase blood flow; this is beneficial to improving the blood supply and nutritional status of ischemic tissues.

[0024] Lowering blood pressure: Associated with the blood vessel dilation effect, supplementing coenzyme Q10 can significantly reduce the blood pressure level of hypertensive patients, have a good adjuvant antihypertensive effect, and prevent kidney damage.

[0025] Anti-inflammatory and anticoagulant: Coenzyme Q10 can inhibit platelet aggregation, reduce blood viscosity, and alleviate the inflammatory response of blood vessel walls, thereby playing an anti-inflammatory and anticoagulant role, preventing blood vessel blockage and thrombus formation, and reducing the risk of cardiovascular and cerebrovascular events. However, traditional coenzyme Q10 preparations have the following problems: Coenzyme Q10 has high liposolubility and low absorption rate in the human digestive system; it is easily degraded by light, heat, and oxidation; it is difficult to precisely control its release time and targeted site in the body.

[0026] Generally speaking, coenzyme Q10 is a substance naturally present in human cells and is widely used in dietary supplements. Its safety has been studied and clinically verified for many years and is suitable for most people to use.

[0027] However, vitamin E, as a liposoluble antioxidant, synergistically scavenges free radicals with coenzyme Q10 and delays oxidative degradation.

[0028] Perilla seed oil or walnut oil contains ω-3 polyunsaturated fatty acids, which, as emulsifiers, improve the dispersion degree of coenzyme Q10 and enhance cell membrane permeability.

[0029] Vitamin C, a water-soluble antioxidant, forms an oxidation-reduction cycle with vitamin E to maintain a reducing environment in the system.

[0030] The forming agents include gelatin, hydroxypropyl methylcellulose (HPMC), or sodium alginate, which are used for microencapsulation to control the release rate.

[0031] In a possible embodiment, the microcapsule is a single-wall structure. The preparation process of the single-wall structure microcapsule is as follows: Dissolve coenzyme Q10 and vitamin E in perilla seed oil and stir in a 50°C water bath until completely dissolved; Add vitamin C and disperse it with a homogenizer (10000 rpm, 5 minutes) to form a uniform oil phase; Dissolve HPMC in pure water (concentration 8%), mix it with the oil phase, and form an emulsion by high-pressure homogenization (30 MPa); Prepare microcapsules by spray drying (inlet air temperature 160°C, outlet air temperature 80°C), with a particle size distribution of 50 - 150 μm.

[0032] In a possible embodiment, the microcapsule is a multi-wall structure. Taking the double-wall structure as an example, the preparation process of the multi-wall structure microcapsule is as follows: Dissolve coenzyme Q10 and natural vitamin E in perilla seed oil and stir at 50°C until transparent and homogeneous.

[0033] Dissolve vitamin C in pure water, add gelatin, and dissolve it at 50°C to form an aqueous phase.

[0034] Mix the oil phase and the water phase at a volume ratio of 1:4 and homogenize at high speed (10,000 rpm, 5 minutes) to form a W / O primary emulsion.

[0035] Drop the primary emulsion into cold water at about 4 °C, and the gelatin quickly gels to form an inner wall material (particle size 100 - 150 μm), that is, the inner microcapsule.

[0036] The gelatin gel temperature at 25 - 30 °C can ensure the embedding integrity and avoid the leakage of coenzyme Q10.

[0037] Mix hydroxypropyl methylcellulose and vitamin C solution, and adjust the pH to 6.5 to obtain a secondary aqueous phase.

[0038] Disperse the inner microcapsules in the secondary aqueous phase and homogenize at low speed (rotation speed 3,000 rpm, time 2 minutes) to form a W / O / W multiple emulsion.

[0039] The inlet air temperature is 160 °C and the outlet air temperature is 70 °C. Hydroxypropyl methylcellulose forms a dense outer layer to wrap the inner microcapsules. A double-walled microcapsule is formed.

[0040] In the embodiments of the present invention, the above microcapsules can be of single-wall structure or multi-wall structure, and different types of microcapsules can also be distinguished by single-wall structure or multi-wall structure.

[0041] In the embodiments of the present invention, coenzyme Q10 with the function of protecting the cardiovascular and cerebrovascular system is obtained by mixing multiple types of microcapsules. One type of microcapsule is prepared by a corresponding target preparation scheme. The microcapsules are prepared from the following raw materials and ratios by weight: coenzyme Q10: 45 - 50 parts; natural vitamin E: 20 - 30 parts; perilla seed oil or walnut oil: 400 - 450 parts; vitamin C: 20 - 30 parts; shaping agent: 80 - 120 parts. By preparing the raw materials and ratios of coenzyme Q10 with the function of protecting the cardiovascular and cerebrovascular system into microcapsules, based on the historical targeting data and the human release curves of various types of microcapsules, the release response characteristics of different types of microcapsules in the human body are utilized to dynamically determine the microcapsule ratio corresponding to the target targeting data, which can better achieve intelligent and personalized customization, dynamically adjust the multi-stage release behavior for different lesion sites, and improve the targeted delivery effect of coenzyme Q10.

[0042] Furthermore, the dosage form of coenzyme Q10 with the function of protecting the cardiovascular and cerebrovascular system includes any one of granule, powder, and capsule.

[0043] In the embodiments of the present invention, the diameter of the microcapsules is generally 1 - 500 μm, and the thickness of the wall is 0.5 - 150 μm. Based on the microcapsules, coenzyme Q10 with the function of protecting the cardiovascular and cerebrovascular system can be prepared into any one of granule, powder, and capsule.

[0044] Based on the microcapsules, coenzyme Q10, which has the function of protecting cardiovascular and cerebrovascular vessels, is prepared into granules. The microcapsules can be combined with auxiliary materials to make particles with a certain particle size and fluidity, which is convenient for taking in divided doses.

[0045] Specifically, the particle size of the microcapsules is screened (usually controlled to be 50-200μm), and the agglomerated or broken particles are removed to ensure uniformity. The corresponding type and amount of microcapsules are selected according to the microcapsule ratio, and the microcapsules are mixed according to the microcapsule ratio.

[0046] Add 5-10% hydroxypropyl methylcellulose (HPMC) aqueous solution or polyvinyl pyrrolidone (PVP) ethanol solution to enhance particle forming properties.

[0047] Add microcrystalline cellulose (MCC) or lactose (20-30%) to adjust the particle hardness and disintegration rate.

[0048] Addition of sodium carboxymethyl starch (CMS-Na, 3-5%) promotes rapid dispersion of particles.

[0049] The mixed microcapsules and auxiliary materials are put into a high-speed shear granulator and sprayed with a binder solution (temperature ≤ 40°C) to form a soft material.

[0050] The granules were sieved (16-20 mesh) and the pressure was controlled at 0.2-0.5 MPa to avoid excessive squeezing that may cause the microcapsules to rupture.

[0051] Fluidized bed drying (40-45°C, humidity <15%) to a moisture content of ≤3%.

[0052] The whole granules are film-coated and treated with Opadry for moisture protection or taste masking.

[0053] It should be noted that the microcapsule breakage rate (≤5%) needs to be monitored during the granulation process, and the integrity needs to be verified by scanning electron microscopy (SEM).

[0054] Particle size range: 0.5-1.5mm, repose angle ≤35°, ensuring fluidity.

[0055] Based on microcapsules, coenzyme Q10, which has the function of protecting cardiovascular and cerebrovascular vessels, is prepared into powder. The microcapsule powder can be directly packaged to achieve rapid dissolution or oral administration, which is suitable for flexible dosage adjustment.

[0056] Specifically, 1-2% silicon dioxide (SiO2) or magnesium stearate is added to prevent the microcapsules from absorbing moisture and agglomerating.

[0057] Adding stevioside or mint flavor (0.5-1%) improves palatability.

[0058] Use a three-dimensional motion mixer to mix the microcapsules and excipients at a low speed of 20-30 rpm for 15-20 minutes to avoid frictional heat generation that damages the microcapsule structure.

[0059] The content of coenzyme Q10 was detected by HPLC, with RSD ≤ 5%.

[0060] Fill into aluminum-plastic composite film bags according to dosage (e.g. 500 mg / bag), seal with nitrogen, and store away from light. The packaging material must have high barrier properties (water vapor transmission rate ≤ 0.5g / m²·day).

[0061] It should be noted that the microcapsule powder needs to be tested by a laser particle size analyzer, with D90 ≤ 150μm, to ensure that there is no gritty feeling when taken.

[0062] Based on the microcapsules, coenzyme Q10, which has the function of protecting cardiovascular and cerebrovascular vessels, is prepared into capsules, and the microcapsules are filled into the capsule shell to achieve precise dosage and targeted release.

[0063] Specifically, 0.5-1% sodium stearyl fumarate is added to improve the filling fluidity of the microcapsules.

[0064] Ion wind eliminates static electricity on the surface of microcapsules to prevent adsorption loss.

[0065] The capsules are filled by a fully automatic capsule filling machine (such as Bosch GKF 2000), and the dosage disk specifications are adjusted according to the microcapsule bulk density. The filling speed is 2000-3000 capsules / minute; the filling accuracy error is ±3% (controlled by the weight difference method).

[0066] The capsule shell is made of gelatin or hydroxypropyl methylcellulose (HPMC), which dissolves quickly in gastric juice.

[0067] Alternatively, the capsule shell is coated with cellulose acetate phthalate (CAP) for release at intestinal pH (≥5.5).

[0068] Polish to remove residual powder on the surface and package in aluminum-plastic blister (temperature 25℃, humidity below 40%).

[0069] After accelerated testing at 40℃ / 75%RH for 6 months, the deviation of the microcapsule release curve was ≤10%.

[0070] It should be noted that capsules need to pass a dissolution test (such as the paddle method in the Chinese Pharmacopoeia) to verify whether the cumulative release rate in different release stages (0-2h, 2-6h) meets the preset release curve.

[0071] Verify the sealing of the capsule shell, such as the release of enteric-coated capsules in simulated gastric fluid within 2 hours is ≤10%.

[0072] By the above method, under the premise of retaining the targeted release function of the microcapsules, the industrial production requirements of granule, powder and capsule preparations can be flexibly adapted, while ensuring the stability and clinical effectiveness of the final coenzyme Q10 product.

[0073] As Figure 1 shown, Figure 1 FIG. is a flowchart of a method for preparing coenzyme Q10 with a cardio-cerebrovascular protective effect provided by an embodiment of the present invention, and is used to prepare coenzyme Q10 with a cardio-cerebrovascular protective effect as described in any one of the embodiments of the present invention. The method for preparing coenzyme Q10 with a cardio-cerebrovascular protective effect includes the steps of: S10. Within the range of the raw material ratio, a plurality of types of microcapsules are prepared according to a plurality of target preparation schemes.

[0074] In the embodiment of the present invention, the above raw materials and the range of the raw material ratio are as follows: Coenzyme Q10: 45-50 parts; Natural vitamin E: 20-30 parts; Perilla seed oil or walnut oil: 400-450 parts; Vitamin C: 20-30 parts; Forming agent: 80-120 parts.

[0075] One type of microcapsule is prepared corresponding to one of the target preparation schemes. The above target preparation scheme is a representative preparation scheme. Specifically, the human release curve corresponding to the type of microcapsule prepared by the above target preparation scheme is representative.

[0076] The different target preparation schemes have different ratios and / or different microcapsule structures. Both the different ratios and the microcapsule structures will affect the human release curve of the microcapsules. Therefore, different types of microcapsules have different human release curves.

[0077] S20. According to the historical targeting data and the human release curves of each type of microcapsule, determine the microcapsule ratio corresponding to the target targeting data.

[0078] In this In the embodiment of the invention, modeling can be performed according to the historical targeting data and the human release curves of each type of microcapsule, and the mapping between the targeting data and the microcapsule ratio is realized through modeling. Therefore, after the target targeting data is determined, according to the mapping between the targeting data and the microcapsule ratio, the microcapsule ratio corresponding to the target targeting data is determined.

[0079] The above historical targeted data can be collected through clinical trials or downloaded from medical databases. The above targeted data may include users' physiological parameters, disease status, medication history, and genetic information. The above targeted data can be structured data obtained through electronic health records (EHRs), wearable devices, questionnaires, etc.

[0080] Physiological parameters may include age, gender, weight, blood pressure, blood lipid levels, liver and kidney functions, etc.

[0081] Disease status may include angina pectoris, heart failure, degree of atherosclerosis, such as plaque area.

[0082] Medication history may include currently used cardiovascular drugs and dosages, such as beta blockers, statins, etc.

[0083] Genetic information may include gene polymorphisms related to coenzyme Q10 metabolism, such as COQ2 gene mutations.

[0084] The above human release curve is used to record the relationship between the release rate of the microcapsules and time.

[0085] Among them, the above microcapsule formulation includes the amounts of various types of microcapsules. Through the above microcapsule formulation, the corresponding amounts of each type of microcapsule can be weighed to prepare the corresponding coenzyme Q10 preparation.

[0086] S30. According to the microcapsule formulation, prepare coenzyme Q10 with a cardio-cerebrovascular protective effect corresponding to the target dosage form.

[0087] In the embodiments of the present invention, the above target dosage form can be any one of granules, powders, and capsules. After determining the microcapsule formulation and the target dosage form, the corresponding types and amounts of microcapsules can be weighed according to the microcapsule formulation and prepared according to the preparation method of the target dosage form. The preparation methods of granules, powders, and capsules have been provided in the above embodiments and will not be elaborated here. It should be noted that the microcapsules can be regarded as powders and prepared into the corresponding granules, powders, and capsules by using other existing preparation methods. At the same time, the preparation method of the present invention can also achieve comprehensive coverage and intelligent screening of the microcapsule preparation scheme, providing core technical support for the large-scale production of personalized preparations.

[0088] In the embodiments of the present invention, within the range of the raw material ratio, multiple types of microcapsules are prepared according to multiple target preparation schemes, and one type of microcapsule is prepared correspondingly by one target preparation scheme; according to the historical targeting data and the human release curves of each type of microcapsule, the microcapsule ratio corresponding to the target targeting data is determined; according to the microcapsule ratio, a coenzyme Q10 with a cardiovascular protection effect corresponding to the target dosage form is prepared. The coenzyme Q10 with a cardiovascular protection effect in the present invention has coenzyme Q10 as its core component. Based on the historical targeting data and the human release curves of each type of microcapsule, the release response characteristics of different types of microcapsules in the human body are utilized to dynamically determine the microcapsule ratio corresponding to the target targeting data, which can better achieve intelligent and personalized customization, dynamically adjust the multi-stage release behavior for different lesion sites, and improve the targeted delivery effect of coenzyme Q10.

[0089] Further, as Figure 2 shown, before step S10, the preparation method further includes: S40. Within the range of the raw material ratio, all candidate types of microcapsules are prepared by candidate preparation schemes.

[0090] S50. In the human digestive environment, the candidate human release curves corresponding to each candidate type are determined.

[0091] Among them, one candidate type corresponds to one candidate human release curve.

[0092] S60. Based on the candidate human release curves, the target preparation scheme is determined among all the preparation schemes, and the human release curve corresponding to the target preparation scheme is determined among all the candidate human release curves.

[0093] Specifically, within the range of the raw material ratio in the above embodiment, the existing preparation schemes for preparing coenzyme Q10 with a cardiovascular protection effect are obtained, and the above existing preparation schemes are classified and screened according to the candidate types of microcapsules to obtain the candidate preparation schemes corresponding to each candidate type.

[0094] The above human digestive environment can be a simulated human digestive environment or a clinically significant human digestive environment. The simulated human digestive environment can be artificial gastric juice with pH 1.2 and intestinal juice with pH 6.8. In the simulated human digestive environment, the dissolution curves of different candidate types of microcapsules are tested. The dissolution curve includes the relationship between the dissolution rate and time. The dissolution of the microcapsule is the release of the core material coenzyme Q10. Therefore, the dissolution curve can be determined as the human release curve of the candidate type of microcapsule, and the candidate human release curves corresponding to each candidate type can be obtained.

[0095] For a type, tests can be conducted multiple times in a simulated human digestive environment, and the average dissolution curve can be taken as the candidate human release curve corresponding to this candidate type. In this way, the candidate human release curve can be made more accurate.

[0096] Among all the candidate human release curves, a representative candidate human release curve can be selected empirically. For example, a candidate human release curve with a peak release rate meeting the empirical conditions can be selected, and the preparation scheme corresponding to the selected candidate human release curve can be determined as the target preparation scheme. After determining the target preparation scheme, among all the candidate human release curves, the candidate human release curve corresponding to the target preparation scheme can be determined as the human release curve corresponding to the target preparation scheme. Of course, in a possible embodiment, it is also possible to first determine, among all the candidate human release curves, the candidate human release curve selected empirically as the target human release curve, and then determine the corresponding target preparation scheme according to the target human release curve, and the target human release curve can then be correspondingly determined as the human release curve corresponding to the target preparation scheme.

[0097] In the embodiments of the present invention, through all the candidate microcapsule types within the raw material ratio range in the above embodiments, combined with multi-level verification of simulation and clinical data, a target preparation scheme that meets the targeted requirements can be screened out, which can solve the problems of incomplete coverage of fixed formulas and low screening efficiency in traditional process optimization.

[0098] Further, as Figure 3 shown, step S50 specifically includes: S51. Under a simulated human digestive environment, conduct digestion simulation on all candidate types of microcapsules to obtain the simulation release data corresponding to all candidate types of microcapsules; S52. According to the simulation release data, determine the first candidate human release curve corresponding to each candidate type.

[0099] Among them, one candidate type corresponds to one first candidate human release curve; S53. Based on the first candidate human release curve corresponding to each candidate type, determine the candidate human release curve corresponding to each candidate type.

[0100] Specifically, the simulated human digestive environment can be a multi-chamber continuous simulation, which includes a coherent gastric simulation chamber, an intestinal simulation chamber, and a colon simulation chamber. Gastric simulation chamber: pH 1.2, containing pepsin (0.1% w / v), temperature 37°C, magnetic stirring (50 rpm) to simulate peristalsis; Intestinal simulation chamber: pH 6.8, containing pancreatin (0.5% w / v) and bile salts (0.3% w / v), continuously introducing CO2 to maintain an anaerobic environment; Colon simulation chamber: pH 7.4, adding probiotics (such as Bifidobacterium) to simulate microbial metabolism. The microcapsule suspension is transferred between the gastric simulation chamber, the intestinal simulation chamber, and the colon simulation chamber according to the digestion time axis (stomach 2h → intestine 4h → colon 6h) through a peristaltic pump.

[0101] The concentration of coenzyme Q10 in each chamber is detected in real time by ultraviolet-visible spectroscopy (UV-Vis) (detection wavelength 275 nm); the change in the particle size of the microcapsules is monitored by an optical fiber sensor to identify rupture or erosion events. Release data is recorded at preset intervals (such as every 5 minutes). After one simulation of digestion is completed, simulated release data is obtained, and a time-release curve can be generated based on the simulated release data.

[0102] After obtaining the time-release curve, a fitting model can be selected based on the structural characteristics of the microcapsules. The Higuchi model (diffusion-controlled release) is selected for single-layer microcapsules; the Korsmeyer-Peppas model (non-Fickian diffusion) is selected for multi-layer microcapsules. For a certain candidate type, the time-release curves obtained from multiple simulations are fitted by the selected fitting model to obtain the first human release curve corresponding to the candidate type.

[0103] After obtaining the first human release curve corresponding to the candidate type, the first human release curve can be directly determined as the human release curve corresponding to the candidate type.

[0104] In the embodiments of the present invention, during the selection process of candidate microcapsules, the release characteristics of candidate microcapsules are accurately measured by simulating the human digestive environment, and high-fidelity release curves are generated by combining data fitting techniques, providing a reliable basis for targeted ratio decision-making.

[0105] Further, as Figure 4 shown, step S50 specifically includes: S54. In a clinical trial environment, conduct clinical trials on microcapsules of all candidate types to obtain the clinical release data corresponding to the microcapsules of all candidate types.

[0106] S55. According to the clinical release data, determine the second candidate human release curve corresponding to each candidate type.

[0107] Among them, one candidate type corresponds to one of the second candidate human body release curves.

[0108] S56. Based on the second candidate human body release curves, determine the candidate human body release curves corresponding to each candidate type.

[0109] S57. Based on the first candidate human body release curve and the second candidate human body release curves, determine the candidate human body release curves corresponding to each candidate type.

[0110] Specifically, in a clinical trial environment, collect clinical trial data. Group the subjects according to the candidate types. Specifically, include healthy volunteers and target patients (such as coronary heart disease, stroke) in the trial at a ratio of 1:1 (n≥50 cases); sample at time points of 0.5 h, 1 h, 2 h, 4 h, 6 h, 8 h, and 12 h, detect the plasma coenzyme Q10 concentration of the subjects by HPLC, and analyze the metabolites to verify the targeted release (such as the fecal recovery rate of the colon-targeted preparation ≥80%), so as to obtain the clinical release data corresponding to the microcapsules of all candidate types.

[0111] It should be noted that the collection of the clinical release data in the embodiments of the present invention needs to comply with the relevant laws and regulations of the National Health Commission and the industry rules and regulations, the clinical conditions need to meet the relevant regulations, and the subjects need to consent to the clinical trial and the collection of user data. That is, the clinical release data in the embodiments of the present invention is legally obtained.

[0112] The clinical release data may include the plasma coenzyme Q10 concentration corresponding to each collection time, as well as various status indicators of the subjects. Based on physiological pharmacokinetics (PBPK), construct a clinical blood drug concentration-time curve. The clinical blood drug concentration-time curve can be determined as the second human body release curve corresponding to the candidate type.

[0113] In a possible embodiment, after obtaining the second human body release curve corresponding to the candidate type, the second human body release curve can be directly determined as the human body release curve corresponding to the candidate type.

[0114] In another possible embodiment, after obtaining the first human release curve corresponding to the candidate type and the second human release curve corresponding to the candidate type, the first human release curve and the second human release curve of the same candidate type can be weighted and fitted to obtain the human release curve corresponding to the candidate type. The weight between the first human release curve and the second human release curve can be determined according to the correlation between the first human release curve and the second human release curve. The greater the correlation, the closer the weight between the first human release curve and the second human release curve is to 0.5:0.5. The smaller the correlation, the smaller the value of a in the weight a:1 - a between the first human release curve and the second human release curve, where a is a decimal less than 0.5 and greater than 0.

[0115] In the embodiments of the present invention, further combined with clinically measured data, a high-fidelity human release curve is generated through multi-source data calibration to improve the accuracy of the human release curve, providing an accurate basis for the targeted ratio of the embodiments of the present invention.

[0116] Furthermore, as Figure 5 shown, step S57 specifically includes: S571. Calculate the first correlation between the first candidate human release curve and the second candidate human release curve of the same candidate type.

[0117] S572. If the first correlation is greater than or equal to the first correlation threshold, fit the first candidate human release curve and the second candidate human release curve to obtain the candidate human release curve corresponding to the candidate type.

[0118] S573. If the first correlation is less than the first correlation threshold, keep the second candidate human release curve unchanged, and fit and iterate the fitting curve between the first candidate human release curve and the second candidate human release curve with the second candidate human release curve. When the fitting iteration reaches the number of iterations, determine the fitting curve obtained in the last fitting iteration as the candidate human release curve corresponding to the candidate type.

[0119] Among them, the number of iterations is determined according to the difference between the first correlation and the first correlation threshold.

[0120] Specifically, the first correlation between the first candidate human release curve and the second candidate human release curve may be the Pearson correlation coefficient or the curve similarity. The first correlation threshold can be set according to empirical values, such as set to 0.85.

[0121] The first correlation degree is greater than or equal to the first correlation degree threshold, indicating that the release curve of the first candidate human body is similar to the release curve of the second candidate human body, that is, the in vitro simulation is similar to the clinical trial data. At this time, the release curve of the first candidate human body and the release curve of the second candidate human body can be directly fitted to obtain the release curve of the candidate human body corresponding to the candidate type. Of course, one of the release curve of the first candidate human body and the release curve of the second candidate human body can also be directly selected as the release curve of the candidate human body corresponding to the candidate type. The least squares method can be used to fit the release curve of the first candidate human body and the release curve of the second candidate human body into the release curve of the candidate human body. The specific fitting formula is as follows: ; where, y ( t ) is the release curve of the candidate human body after fitting, α is the weight coefficient, α can be a decimal value between 0 and 1, for example, it can be 0.5, is the release curve of the first candidate human body, is the release curve of the second candidate human body.

[0122] If the first correlation degree is less than the first correlation degree threshold, it indicates that the release curve of the first candidate human body is quite different from the release curve of the second candidate human body, that is, the in vitro simulation is not similar to the clinical trial data. At this time, taking the release curve of the second candidate human body as the benchmark and the release curve of the first candidate human body as an auxiliary reference, an initial fitting curve is generated, and then the initial fitting curve and the release curve of the second candidate human body are fitted. In the subsequent iteration process, the fitting curves obtained are all fitted with the release curve of the second candidate human body in the next round of iteration.

[0123] Specifically, the iterative fitting can be performed through the following formula: ; where, represents the fitting curve obtained in the T th iteration, represents the fitting curve obtained in the th iteration.

[0124] The number of iterations is determined according to the difference between the first correlation degree and the first correlation degree threshold. Specifically, the number of iterations is [5*(Th1 - R)] + , where Th1 is the first correlation degree threshold, R is the first correlation degree, and [ ] + represents rounding up to the nearest integer.

[0125] In the embodiments of the present invention, on the basis of simulated-clinical data fusion, the problem of curve correction in the scenario of low correlation between in vitro simulation and clinical trial data is further solved, and the robustness of release curve prediction is improved through iterative fitting technology.

[0126] Further, as Figure 6 shown, step S60 specifically includes: S61. Extract the N-dimensional features of the candidate human release curves, where N is an integer greater than 1.

[0127] S62. Cluster all the candidate human release curves according to the N-dimensional features to obtain N clusters.

[0128] Wherein, each cluster includes at least one candidate human release curve.

[0129] S63. For each cluster, determine the preparation schemes corresponding to the M candidate human release curves closest to the cluster center as the candidate preparation schemes, where M is an integer greater than or equal to 1.

[0130] S64. Remove duplicates from all the candidate preparation schemes, determine the de-duplicated candidate preparation schemes as the target preparation schemes, and determine the candidate human release curves corresponding to the target preparation schemes as the human release curves corresponding to the target preparation schemes.

[0131] Specifically, the above N-dimensional features may include time-related features, dynamic change features, statistical features, etc. Among them, the time-related features may include T25, T50, T80 (25%, 50%, 80% release times), etc. The dynamic change features may include the slope of the gastric release stage (k1), the slope of the intestinal release stage (k2), the colon-targeted release amount (Qc), etc. The statistical features may include the area under the curve (AUC), the peak volatility rate (PFR), etc. The Z-score normalization can be used to eliminate the dimensional differences between the N-dimensional features.

[0132] The DBSCAN density clustering can be used to cluster all the candidate human release curves according to the N-dimensional features to obtain N clusters. In a possible embodiment, the feature weights can be dynamically assigned according to the targeting requirements (for example, the weight of Qc is increased to 70% in the colon-targeting requirement).

[0133] After obtaining N clusters, for each cluster, calculate the cluster center in the weighted feature space, calculate the Euclidean distances between all the candidate human release curves in the cluster and the cluster center, and select the preparation schemes corresponding to the top M candidate human release curves with the smallest distances (M = 1-3) as the candidate preparation schemes.

[0134] Deduplicate all candidate preparation schemes, and determine the deduplicated candidate preparation schemes as the target preparation schemes. After obtaining the target preparation schemes, determine the human release curves corresponding to the target preparation schemes as the human release curves corresponding to the target preparation schemes.

[0135] In the embodiments of the present invention, based on candidate type microcapsules, through multi-dimensional feature extraction and clustering analysis, the target preparation schemes that meet the targeting requirements are efficiently screened, improving the low screening efficiency and poor screening accuracy.

[0136] Further, as Figure 7 shown, step S20 specifically includes: S21. Based on historical targeting data and the human release curves of each type of microcapsules, construct a first mapping model between the targeting data and the microcapsule ratio. The first mapping model includes the mapping paths between the targeting data and the microcapsule ratio, and each mapping path corresponds to a mapping weight.

[0137] S22. According to the collected real-time process parameters, determine the real-time dynamic path adjustment parameters, and adjust the mapping weights corresponding to each mapping path in the first mapping model according to the real-time dynamic path adjustment parameters to obtain a second mapping model.

[0138] S23. According to the second mapping model, determine the microcapsule ratio corresponding to the target targeting data.

[0139] Specifically, classify the targeting requirements from the historical targeting data, and use a graph neural network (GNN) to construct an A-C-B three-layer heterogeneous graph spectrum with the targeting requirements as one layer of nodes (A-layer nodes), the microcapsule ratio as one layer of nodes (B-layer nodes), and the release curve as one layer of nodes (C-layer nodes). The C-layer nodes are intermediate nodes for connecting the A-layer nodes and the B-layer nodes. In the three-layer heterogeneous graph spectrum, the nodes in each layer will be connected to the nodes in the adjacent layer to form a mapping path. For example, the mapping path A i -C j -B k represents a path formed by connecting the i-th node in the A-layer nodes, the j-th node in the C-layer nodes, and the k-th node in the B-layer nodes.

[0140] Among them, the mapping path corresponds to a mapping weight, and the mapping weight can be related to the association frequency or matching success rate between the historical targeting data and the microcapsule ratio. For example, if the proportion of the number of associations between a certain targeting requirement and a specific microcapsule ratio analyzed from the historical targeting data to the total number of associations between this targeting requirement and all microcapsule ratios is 90%, or the clinical matching success rate of a certain targeting requirement and a specific microcapsule ratio reaches 90%, then the initial weight is set to 0.9.

[0141] The above mapping weights can also be determined according to the curve correlation between the targeting requirements and the microcapsule ratio.

[0142] In a possible embodiment, artificial intelligence technology, such as through a medical large language model (MedGPT), can be used to analyze historical targeting data and the human release curves of various types of microcapsules, so as to construct a mapping model between the targeting data and the microcapsule ratio. The mapping model records the mapping relationship and mapping weights between the targeting data and the microcapsule ratio, and can map the corresponding microcapsule ratio according to the input targeting data, or map the corresponding targeting data (indications) according to the input microcapsule ratio.

[0143] In a possible embodiment, after constructing the first mapping model between the targeting data and the microcapsule ratio, the target targeting data can be used as input to obtain the microcapsule ratio corresponding to the target targeting data.

[0144] To better achieve intelligent and personalized precise customization, the real-time process parameters of each ratio of microcapsules during the production process can also be monitored in real time. For the microcapsules corresponding to a target preparation plan, the process deviation value Δ between the real-time process parameters of the microcapsules and the target process parameters can be determined, and the mapping weights corresponding to each mapping path in the first mapping model can be adjusted according to the process deviation value Δ. w new = w old ⋅ e −γΔ , w old is the mapping weight corresponding to all mapping paths including the microcapsule in the first mapping model, w new is the mapping weight in the second mapping model, γ is the attenuation coefficient, which can be set to 0.1 by default. Among them, the above process parameters can include parameters such as flow rate, temperature, rotation speed, etc. in each process step. The process deviation value Δ can be the absolute value of the difference between the calculated parameters in the same process step (the absolute value of the difference between the real-time process parameters and the target process parameters), and then the absolute value of the difference between the parameters is weighted and summed according to the preset parameter weights to obtain the step deviation value corresponding to each process step. The step deviation values are weighted and summed according to the step weights corresponding to the process steps to obtain the process deviation value Δ. The parameter weights of the above parameters and the parameter weights of each process step can be set according to experience.

[0145] After constructing the second mapping model between the targeting data and the microcapsule ratio, the target targeting data can be used as input to obtain the microcapsule ratio corresponding to the target targeting data.

[0146] In the embodiments of the present invention, based on the preparation and data analysis process of coenzyme Q10 with the effect of protecting the cardiovascular and cerebrovascular system in the above embodiments, by constructing a dynamic association model between the target data and the microcapsule ratio, the intelligent mapping from clinical needs to formulation is realized, and the low efficiency problem of the traditional experience-driven ratio design is solved.

[0147] Further, as Figure 8 shown, step S21 specifically includes: S211. Based on the human release curves of each type of microcapsule, determine the comprehensive human release curves corresponding to different microcapsule ratios.

[0148] S212. Determine the coenzyme Q10 absorption curves required for each historical target data.

[0149] S213. Calculate the second correlation degree between the comprehensive human release curve and the coenzyme Q10 absorption curve.

[0150] S214. Associate the comprehensive human release curves with the coenzyme Q10 absorption curves whose second correlation degree is greater than or equal to the second correlation degree threshold, and obtain the association relationship between the comprehensive human release curve and the coenzyme Q10 absorption curve.

[0151] S215. Based on the association relationship between the comprehensive human release curve and the coenzyme Q10 absorption curve, determine the mapping path between the microcapsule ratio and the historical target data, and determine the mapping weight of the corresponding mapping path according to the second correlation degree, so as to obtain the first mapping model between the microcapsule ratio and the historical target data.

[0152] Specifically, some representative microcapsule ratios can be prepared in advance to obtain all microcapsule ratios, and based on the human release curves corresponding to each type of microcapsule, generate the comprehensive human release curves corresponding to different microcapsule ratios. Align the time axes and perform weighted averaging on multiple batches of curves under the same ratio to generate the comprehensive human release curve. The formula is: ; where y Comprehensive( t ) is the comprehensive human release curve; w i represents the data source weight, that is, the weight of the i-th human release curve, which is set according to the reliability of the data source. For example, the clinical data weight = 0.7, and the in vitro data = 0.3; y i ( t ) is the i-th human release curve.

[0153] Convert the historical target data into the coenzyme Q10 absorption curve, such as "coronary heart disease requires maintaining an effective concentration for 12h", and define the key parameter: onset time (Tonset ), duration of maintenance (T maintain ), minimum effective concentration at the target site (C min ).

[0154] The dynamic time warping (DTW) algorithm can be used to calculate the similarity score between the comprehensive human release curve and the coenzyme Q10 absorption curve (DTW distance ≤ 10 is highly correlated), and the DTW distance can be determined as the second correlation degree; extract curve features, such as peak overlap rate, slope consistency, etc., and construct a random forest classifier to determine the association level (strong / medium / weak).

[0155] A graph neural network (GNN) can be adopted. Taking the target data as one layer of nodes (layer A nodes), the microcapsule ratio as one layer of nodes (layer B nodes), the comprehensive human release curve as one layer of nodes (layer C nodes), and the coenzyme Q10 absorption curve as one layer of nodes (layer D nodes), an A-D-C-B four-layer heterogeneous graph spectrum is constructed. The layer D nodes and layer C nodes are intermediate nodes. The layer D nodes connect the layer A nodes and layer C nodes, and the layer C nodes connect the layer D nodes and layer B nodes. In the four-layer heterogeneous graph spectrum, the nodes in each layer will be connected to the nodes in the adjacent layer to form a mapping path, such as the mapping path A i -D h -C j -B k represents the path formed by connecting the i-th node in the layer A nodes, the h-th node in the layer D nodes, the j-th node in the layer C nodes, and the k-th node in the layer B nodes.

[0156] Among them, the mapping path corresponds to a mapping weight. The mapping weight can be determined by the second correlation degree between the h-th node in the layer D nodes and the j-th node in the layer C nodes. The layer D nodes represent the coenzyme Q10 absorption curve, and the layer C nodes represent the comprehensive human release curve. The weight between the target data and the coenzyme Q10 absorption curve can be 1 / K, where K is the number of coenzyme Q10 absorption curves required for the target data. There is a one-to-one correspondence between the microcapsule ratio and the comprehensive human release curve, and the weight can be 1. Therefore, the product of the second correlation degree and 1 / K can be directly used as the mapping weight corresponding to the mapping path.

[0157] In a possible embodiment, through the generation of an initial mapping table, the high-correlation (DTW ≤ 10 and classified as "strong") ratio-target pairs (i.e., the mapping relationship between the microcapsule ratio and the historical target data) can be stored in the mapping database. New ratio-target pairs can be added to the mapping database subsequently, so that the mapping model can be simplified.

[0158] In a possible embodiment, through an online learning mechanism, after a new targeting requirement is input, if there is no matching record in the database, the microcapsule formulation recommendation model is activated to generate a new solution; the verified new solution is added to the mapping table, and the similarity calculation rule is updated, such as adjusting the DTW weight.

[0159] In a possible embodiment, if the comprehensive curve of a certain formulation is visually similar to the target curve but the DTW score is abnormally high, the Manhattan distance is activated for rechecking to exclude the interference of time-axis offset; the training set and the test set are divided by the hold-out method to ensure that the accuracy of the mapping model is ≥90%.

[0160] In the embodiments of the present invention, a more accurate mapping model can be constructed through multi-dimensional correlation analysis of the comprehensive release curve and the absorption curve, and a high-precision formulation recommendation rule can be dynamically generated.

[0161] The preparation method of coenzyme Q10 with the effect of protecting the cardiovascular and cerebrovascular system provided by the embodiments of the present invention can be based on the preparation and data analysis process of coenzyme Q10 with the effect of protecting the cardiovascular and cerebrovascular system in the above embodiments, and by quantifying the correlation between the release curve and the clinical requirements, an intelligent mapping model of "targeting requirement - microcapsule formulation" is dynamically constructed to solve the problems of strong subjectivity and poor adaptability in traditional formulation design.

[0162] The above-disclosed are only the preferred embodiments of the present invention, and of course, the scope of the rights of the present invention cannot be limited thereby. Therefore, equivalent changes made according to the claims of the present invention still fall within the scope covered by the present invention.

Claims

1. A coenzyme Q10 with the function of protecting the cardiovascular and cerebrovascular system, characterized in that, It is obtained by mixing multiple types of microcapsules. One type of microcapsule is prepared according to a corresponding target preparation scheme. The microcapsule is prepared from the following raw materials and their ratios by weight: Coenzyme Q10: 45 - 50 parts; Natural vitamin E: 20 - 30 parts; Perilla seed oil or walnut oil: 400 - 450 parts; Vitamin C: 20 - 30 parts; Forming agent: 80 - 120 parts.

2. The coenzyme Q10 with the function of protecting cardiovascular and cerebrovascular systems according to claim 1, characterized in that, The dosage forms of coenzyme Q10 with the effect of protecting the cardiovascular and cerebrovascular system include any one of granule, powder, and capsule.

3. A preparation method of coenzyme Q10 with the function of protecting cardiovascular and cerebrovascular systems, characterized in that, For preparing coenzyme Q10 with the effect of protecting the cardiovascular and cerebrovascular system as described in claim 1 or 2, the preparation method includes: Within the ratio range of the raw materials, multiple types of microcapsules are prepared according to multiple target preparation schemes. One type of microcapsule is prepared corresponding to one of the target preparation schemes; According to the historical targeting data and the human release curves of each type of microcapsule, the microcapsule ratio corresponding to the target targeting data is determined; According to the microcapsule ratio, coenzyme Q10 with the effect of protecting the cardiovascular and cerebrovascular system corresponding to the target dosage form is prepared.

4. The preparation method of coenzyme Q10 with the function of protecting cardiovascular and cerebrovascular systems as claimed in claim 3, wherein Before the step of preparing multiple types of microcapsules according to multiple target preparation schemes within the ratio range of the raw materials, the preparation method further includes: Within the ratio range of the raw materials, all candidate types of microcapsules are prepared by candidate preparation schemes; Under the human digestive environment, the candidate human release curves corresponding to each candidate type are determined. One candidate type corresponds to one candidate human release curve; Based on the candidate human release curves, the target preparation scheme is determined among all the preparation schemes, and among all the candidate human release curves, the human release curve corresponding to the target preparation scheme is determined.

5. The preparation method of coenzyme Q10 with the function of protecting cardiovascular and cerebrovascular systems according to claim 4, characterized in that, The step of determining the candidate human release curves corresponding to each candidate type under the human digestive environment specifically includes: Under the simulated human digestive environment, digestion simulation is carried out on all candidate types of microcapsules to obtain the simulated release data corresponding to all candidate types of microcapsules; According to the simulated release data, the first candidate human release curves corresponding to each candidate type are determined. One candidate type corresponds to one first candidate human release curve; Based on the first candidate human release curves corresponding to each candidate type, the candidate human release curves corresponding to each candidate type are determined.

6. The preparation method of coenzyme Q10 with the function of protecting cardiovascular and cerebrovascular system as claimed in claim 5, characterized in that, The step of determining the candidate human release curves corresponding to each candidate type under the human digestive environment specifically includes: Under the clinical trial environment, clinical trials are carried out on all candidate types of microcapsules to obtain the clinical release data corresponding to all candidate types of microcapsules; According to the clinical release data, the second candidate human release curves corresponding to each candidate type are determined. One candidate type corresponds to one second candidate human release curve; Based on the second candidate human release curves, the candidate human release curves corresponding to each candidate type are determined; Alternatively, based on the first candidate human body release curve and the second candidate human body release curve, determine the candidate human body release curve corresponding to each candidate type.

7. The preparation method of coenzyme Q10 with the function of protecting cardiovascular and cerebrovascular systems as claimed in claim 6, wherein The steps of determining the candidate human body release curve corresponding to each candidate type based on the first candidate human body release curve and the second candidate human body release curve specifically include: Calculate the first correlation degree between the first candidate human body release curve and the second candidate human body release curve of the same candidate type; If the first correlation degree is greater than or equal to the first correlation degree threshold, fit the first candidate human body release curve and the second candidate human body release curve to obtain the candidate human body release curve corresponding to the candidate type; If the first correlation degree is less than the first correlation degree threshold, keep the second candidate human body release curve unchanged as a benchmark, and perform fitting iteration on the fitting curve between the first candidate human body release curve and the second candidate human body release curve and the second candidate human body release curve. When the fitting iteration reaches the number of iterations, determine the fitting curve obtained from the last fitting iteration as the candidate human body release curve corresponding to the candidate type, and the number of iterations is determined according to the difference between the first correlation degree and the first correlation degree threshold.

8. The preparation method of coenzyme Q10 with the function of protecting cardiovascular and cerebrovascular systems as claimed in claim 4, characterized in that, The steps of determining the target preparation scheme among all preparation schemes based on the candidate human body release curve and determining the human body release curve corresponding to the target preparation scheme among all candidate human body release curves specifically include: Extract the N-dimensional features of the candidate human body release curve, where N is an integer greater than 1; Cluster all the candidate human body release curves according to the N-dimensional features to obtain N clustering clusters, and each clustering cluster includes at least one candidate human body release curve; For each clustering cluster, determine the preparation schemes corresponding to the M candidate human body release curves closest to the cluster center as candidate preparation schemes, where M is an integer greater than or equal to 1; Remove duplicates from all the candidate preparation schemes, determine the de-duplicated candidate preparation schemes as the target preparation scheme, and determine the candidate human body release curve corresponding to the target preparation scheme as the human body release curve corresponding to the target preparation scheme.

9. The preparation method of coenzyme Q10 with the function of protecting cardiovascular and cerebrovascular system according to any one of claims 3 to 8, characterized in that, The steps of determining the microcapsule ratio corresponding to the target targeting data according to the historical targeting data and the human body release curves of each type of microcapsule specifically include: Based on the historical targeting data and the human body release curves of each type of microcapsule, construct a first mapping model between the targeting data and the microcapsule ratio. The first mapping model includes the mapping path between the targeting data and the microcapsule ratio, and each mapping path corresponds to a mapping weight; According to the collected real-time process parameters, determine the real-time dynamic path adjustment parameters, and adjust the mapping weights corresponding to each mapping path in the first mapping model according to the real-time dynamic path adjustment parameters to obtain a second mapping model; According to the second mapping model, determine the microcapsule ratio corresponding to the target targeting data.

10. The preparation method of coenzyme Q10 with the function of protecting cardiovascular and cerebrovascular systems as described in claim 9, characterized in that, The steps of constructing a first mapping model between the targeting data and the microcapsule ratio based on the historical targeting data and the human release curves of each type of the microcapsules specifically include: Based on the human release curves of each type of the microcapsules, determining a comprehensive human release curve corresponding to different microcapsule ratios; Determining the coenzyme Q10 absorption curve required for each of the historical targeting data; Calculating a second correlation degree between the comprehensive human release curve and the coenzyme Q10 absorption curve; Associating the comprehensive human release curve and the coenzyme Q10 absorption curve with the second correlation degree greater than or equal to the second correlation degree threshold to obtain an association relationship between the comprehensive human release curve and the coenzyme Q10 absorption curve; Based on the association relationship between the comprehensive human release curve and the coenzyme Q10 absorption curve, determining a mapping path between the microcapsule ratio and the historical targeting data, and determining a mapping weight corresponding to the mapping path according to the second correlation degree to obtain a first mapping model between the microcapsule ratio and the historical targeting data.

Images