Personalized nutrition regimen optimization system based on gut microbiota metabolism
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HUAYAN SHENGTAI (SHANXI) HEALTH TECHNOLOGY CO LTD
- Filing Date
- 2026-05-20
- Publication Date
- 2026-08-07
AI Technical Summary
[0004]针对上述情况，为克服现有技术的缺陷，本发明提出的基于肠菌代谢的个性化营养方案优化系统，有效的解决了目前个性化营养方案缺乏肠菌代谢全链路关联、静态设计无法适配菌群动态变化、效果评估主观、调整依赖人工，且长期效果不稳定、精准度不足的问题
[0013]采用上述结构本发明取得的有益效果如下：本方案提出的基于肠菌代谢的个性化营养方案优化系统，通过构建“菌群—代谢—表型—营养”全链路量化模型，实现全流程智能化，从数据采集到方案迭代全程自动化，无需人工干预；通过个性化营养方案生成模块，实现高度个性化，实现“一人一策、动态调整”，适配菌群高度异质性；使用动态监测与效果评估模块，通过多维指数客观评估，避免主观判断偏差，实现可量化；通过AI自学习迭代优化模块，实现闭环迭代持续适配菌群变化，防止效果反弹，规避过敏或不耐受风险，实时监测不良反应，解决了现有个性化营养干预技术的诸多痛点，填补了现有技术空白，具有广泛的应用前景。
Smart Images
Figure CN122531636A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of biotechnology, specifically referring to a personalized nutrition optimization system based on gut microbiota metabolism. Background Technology
[0002] As the core of the human gut micro-ecosystem, the gut microbiota's metabolic activities are closely related to human nutrient absorption, metabolic balance, immune function, and the occurrence and development of various diseases. The metabolic products of gut microbiota directly regulate human physiological phenotypes and affect the risk of metabolic-related problems such as obesity, abnormal blood sugar, and intestinal inflammation.
[0003] Currently, personalized nutrition solutions are designed solely based on human physiological indicators or the abundance of a single gut microbiota, resulting in significant technical limitations: 1. They fail to establish a complete link between gut microbiota, metabolites, physiological phenotypes, and nutritional interventions, hindering precise targeted regulation; 2. Existing nutritional solutions are mostly statically designed, unable to adapt to the dynamic changes in gut microbiota, leading to poor intervention effects and a high risk of rebound; 3. They lack a scientific system for quantifying and evaluating effectiveness, relying on subjective judgment and manual experience for adjustments, resulting in low efficiency and poor individual adaptability. Furthermore, the field of precision nutrition faces challenges such as unverified long-term effects, insufficient algorithm interpretability, and inability to adapt to individual metabolic changes. Existing technologies cannot solve the technical difficulties of precisely matching gut microbiota metabolism with nutritional interventions, dynamically optimizing them, and quantifying their effects, making it difficult to meet individual needs for personalized, precise, and long-term effective nutritional interventions. Therefore, there is an urgent need to develop a personalized nutrition solution optimization system based on gut microbiota metabolism. Summary of the Invention
[0004] In response to the above situation and to overcome the shortcomings of the existing technology, the present invention proposes a personalized nutrition program optimization system based on gut microbiota metabolism. This system effectively solves the problems of current personalized nutrition programs lacking full-chain correlation of gut microbiota metabolism, static design being unable to adapt to dynamic changes in the gut microbiota, subjective effect evaluation, reliance on manual adjustment, and unstable long-term effects and insufficient accuracy.
[0005] The technical solution adopted in this invention is as follows: The personalized nutrition plan optimization system based on gut microbiota metabolism proposed in this invention includes a multi-source data acquisition and standardization module, a gut microbiota metabolism correlation modeling module, a personalized nutrition plan generation module, a dynamic monitoring and effect evaluation module, and an AI self-learning iterative optimization module. The multi-source data acquisition and standardization module, the gut microbiota metabolism correlation modeling module, the personalized nutrition plan generation module, the dynamic monitoring and effect evaluation module, and the AI self-learning iterative optimization module work together to complete the entire closed loop from multi-source data acquisition, gut microbiota metabolism correlation modeling, personalized plan generation, to dynamic monitoring and evaluation and AI self-learning iteration, thereby realizing the precise, dynamic, and intelligent optimization of nutrition plans.
[0006] Preferably, the multi-source data acquisition and standardization module includes a data acquisition unit and a data standardization unit. The data acquisition unit integrates a gut microbiota detection unit, a metabolomics detection unit, a physiological phenotype unit, and a behavioral data unit. The gut microbiota detection unit uses three detection methods: 16S rRNA, metagenomic sequencing, and qPCR quantification, to achieve accurate quantitative analysis of gut microbiota, effectively improving the accuracy of microbiota structure detection and reducing the impact of experimental operations on the results. The metabolomics detection unit detects short-chain fatty acids, bile acids, tryptophan metabolites, and lipopolysaccharides in feces / blood. The physiological phenotype unit collects physiological indicators such as weight, body fat, blood glucose, blood lipids, inflammatory factors, and intestinal symptom scores. The behavioral data unit collects dietary logs, exercise, sleep, and stress-perceived behavioral data. The data standardization unit is used to construct a unified data format, remove batch effects, fill missing values, and transform gut microbiota, metabolomics, physiological phenotype, and behavioral data into a standardized feature matrix, providing a unified data foundation for subsequent modeling and analysis.
[0007] To achieve a precise correlation between gut microbiota metabolism and physiological phenotype, the gut microbiota metabolism correlation modeling module includes a core microbiota-metabolite correlation network and a metabolism-phenotype mapping model. The core microbiota-metabolite correlation network uses a multivariate Bayesian network + LASSO regression algorithm to mine the quantitative causal relationship between core driving microbiota and key metabolic pathways from microbiota OTU, species abundance, and metabolite concentration data, and outputs an "Individual Gut Microbiota Metabolic Risk Map". The metabolism-phenotype mapping model uses a random forest + XGBoost fusion model, inputting standardized gut microbiota metabolic feature data, and outputting an individual metabolic risk score, including four core risk scores: obesity risk, abnormal blood glucose risk, intestinal inflammation risk, and intestinal barrier damage risk, providing precise targeting basis for the generation of personalized nutrition plans.
[0008] Furthermore, the personalized nutrition plan generation module includes a nutrient target library and a multi-objective optimization algorithm. The nutrient target library is a pre-built database containing 12 types of nutrients and 36 functional components, clearly recording the regulatory effects of each component on core gut microbiota and metabolic pathways. For example, fructooligosaccharides can proliferate Bifidobacteria, thereby increasing butyrate content, and resistant starch can regulate the ratio of Bacteroides to Firmicutes, thereby improving insulin resistance. The multi-objective optimization algorithm uses "maximizing beneficial metabolites, minimizing harmful metabolites, matching individual dietary preferences, and avoiding food allergies" as its objective function. It automatically generates personalized nutrition formulas using a non-dominated sorting genetic algorithm. The nutrition formulas specifically include the proportion of macronutrients (carbohydrates, protein, and fat), the type and dosage of dietary fiber, the combination of functional components (prebiotics, probiotics, and plant polyphenols), as well as dietary restrictions and meal recommendations, achieving precise matching between the nutrition plan and individual gut microbiota metabolic characteristics and dietary preferences.
[0009] To achieve real-time monitoring and accurate evaluation of the effectiveness of nutritional intervention programs, the dynamic monitoring and effectiveness evaluation module includes a real-time monitoring unit and an effectiveness quantification evaluation system. The real-time monitoring unit is equipped with portable rapid metabolite test strips and a smart terminal, regularly collecting fecal SCFA and urinary oxidative stress indicators. Combined with daily dietary and symptom records, a dynamic monitoring data stream is generated, enabling real-time tracking of key indicators during the intervention process. The effectiveness quantification evaluation system constructs a multi-dimensional effectiveness index, calculated using the following formula:
[0010]
[0011] Among them, weight - It adaptively adjusts to individual metabolic risk type and achieves objective and quantitative assessment of intervention effects through multidimensional effect indices, avoiding subjective judgment bias.
[0012] Furthermore, the AI self-learning iterative optimization module includes a reinforcement learning optimizer and a knowledge base update unit. The reinforcement learning optimizer adopts a deep deterministic policy gradient algorithm, using a multidimensional effect index as a reward signal. It inputs the current nutritional plan, dynamic monitoring data, and individual response data to automatically adjust the proportion of nutritional components. The specific adjustment logic is as follows: when the individual responds well to the current plan, the plan is maintained and fine-tuned; when the response is poor, the bottleneck flora / metabolic pathway is identified and the functional component is replaced; when component intolerance occurs, the dosage of the stimulating component is reduced and the mild component is replaced. The knowledge base update unit, after completing a 4-week intervention cycle, integrates the individual's detection data, intervention effect data, and plan adjustment data into the global gut microbiota-nutrition knowledge base, continuously optimizing the association model and plan generation rules to achieve dual iteration of collective intelligence and individual precision, thereby improving the overall accuracy and adaptability of the system.
[0013] The beneficial effects achieved by this invention using the above structure are as follows: The personalized nutrition program optimization system based on gut microbiota metabolism proposed in this solution achieves full-process intelligence by constructing a quantitative model of the entire "microbiota-metabolism-phenotype-nutrition" chain. From data collection to program iteration, the entire process is automated without manual intervention. Through the personalized nutrition program generation module, a high degree of personalization is achieved, realizing "one person, one policy, dynamic adjustment" to adapt to the high heterogeneity of the microbiota. Using the dynamic monitoring and effect evaluation module, objective evaluation through multi-dimensional indices avoids subjective judgment bias and achieves quantifiability. Through the AI self-learning iterative optimization module, closed-loop iteration is achieved to continuously adapt to changes in the microbiota, prevent effect rebound, avoid the risk of allergies or intolerances, and monitor adverse reactions in real time. This solves many pain points of existing personalized nutrition intervention technologies, fills the gap in existing technologies, and has broad application prospects. Attached Figure Description
[0014] Figure 1 This is a schematic diagram of the overall process structure of the personalized nutrition solution optimization system based on gut microbiota metabolism proposed in this invention.
[0015] Figure 2 This is a schematic diagram of the application process of the personalized nutrition program optimization system based on gut microbiota metabolism proposed in this invention.
[0016] The system comprises the following modules: 1. Multi-source data acquisition and standardization module; 2. Gut microbiota metabolism association modeling module; 3. Personalized nutrition plan generation module; 4. Dynamic monitoring and effect evaluation module; 5. AI self-learning iterative optimization module; 11. Data acquisition unit; 12. Data standardization unit; 13. Gut microbiota detection unit; 14. Metabolomics detection unit; 15. Physiological phenotype unit; 16. Behavioral data unit; 21. Core microbiota-metabolite association network; 22. Metabolism-phenotype mapping model; 31. Nutritional target library; 32. Multi-objective optimization algorithm; 41. Real-time monitoring unit; 42. Effect quantification evaluation system; 51. Reinforcement learning optimizer; and 52. Knowledge base update unit.
[0017] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof. Detailed Implementation
[0018] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.
[0019] like Figure 1 and Figure 2 As shown, the personalized nutrition program optimization system based on gut microbiota metabolism proposed in this invention includes a multi-source data acquisition and standardization module 1, a gut microbiota metabolism association modeling module 2, a personalized nutrition program generation module 3, a dynamic monitoring and effect evaluation module 4, and an AI self-learning iterative optimization module 5. The multi-source data acquisition and standardization module 1, the gut microbiota metabolism association modeling module 2, the personalized nutrition program generation module 3, the dynamic monitoring and effect evaluation module 4, and the AI self-learning iterative optimization module 5 work together to complete the entire closed loop from multi-source data acquisition, gut microbiota metabolism association modeling, personalized program generation, to dynamic monitoring and evaluation and AI self-learning iteration, thereby achieving precise, dynamic, and intelligent optimization of the nutrition program.
[0020] like Figure 1 and Figure 2As shown, the multi-source data acquisition and standardization module 1 includes a data acquisition unit 11 and a data standardization unit 12. The data acquisition unit 11 integrates a gut microbiota detection unit 13, a metabolomics detection unit 14, a physiological phenotype unit 15, and a behavioral data unit 16. The gut microbiota detection unit 13 uses three detection methods: 16S rRNA, metagenomic sequencing, and qPCR quantification to achieve accurate quantitative analysis of gut microbiota, effectively improving the accuracy of microbiota structure detection and reducing the impact of experimental operations on the results. The metabolomics detection unit 14 detects short-chain fatty acids, bile acids, tryptophan metabolites, and lipopolysaccharides in feces / blood. The physiological phenotype unit 15 collects physiological indicators such as weight, body fat, blood glucose, blood lipids, inflammatory factors, and intestinal symptom scores. The behavioral data unit 16 collects dietary diaries, exercise, sleep, and stress-perceived behavioral data. The data standardization unit 12 is used to construct a unified data format, remove batch effects, fill missing values, and transform gut microbiota, metabolomics, physiological phenotype, and behavioral data into a standardized feature matrix, providing a unified data foundation for subsequent modeling and analysis.
[0021] like Figure 1 and Figure 2 As shown, the gut microbiota metabolism association modeling module 2 includes a core microbiota-metabolite association network 21 and a metabolism-phenotype mapping model 22. The core microbiota-metabolite association network 21 uses a multivariate Bayesian network + LASSO regression algorithm to mine the quantitative causal relationship between the core driving microbiota and key metabolic pathways from microbiota OTU, species abundance and metabolite concentration data, and outputs the "Individual Gut Microbiota Metabolic Risk Map". The metabolism-phenotype mapping model 22 uses a random forest + XGBoost fusion model. It takes standardized gut microbiota metabolic feature data as input and outputs an individual metabolic risk score, including four core risk scores: obesity risk, abnormal blood glucose risk, intestinal inflammation risk and intestinal barrier damage risk, providing a precise targeting basis for the generation of personalized nutrition plans.
[0022] Furthermore, the personalized nutrition plan generation module 3 includes a nutrient target library 31 and a multi-objective optimization algorithm 32. The nutrient target library 31 is a pre-built database containing 12 types of nutrients and 36 functional components, clearly recording the regulatory effects of each component on the core flora and metabolic pathways. For example, fructooligosaccharides can proliferate Bifidobacteria and thus increase butyrate content, while resistant starch can regulate the ratio of Bacteroides and Firmicutes and thus improve insulin resistance. The multi-objective optimization algorithm 32 uses "maximizing beneficial metabolites, minimizing harmful metabolites, matching individual dietary preferences, and avoiding food allergies" as the objective function. It uses a non-dominated sorting genetic algorithm to automatically generate personalized nutrition formulas. The nutrition formulas specifically include the proportion of macronutrients (carbohydrates, proteins, and fats), the type and dosage of dietary fiber, the combination of functional components (prebiotics, probiotics, and plant polyphenols), as well as dietary restrictions and meal recommendations, to achieve precise matching between the nutrition plan and the individual's gut microbiota metabolic characteristics and dietary preferences.
[0023] like Figure 1 and Figure 2 As shown, the dynamic monitoring and effect evaluation module 4 includes a real-time monitoring unit 41 and an effect quantification evaluation system 42. The real-time monitoring unit 41 is equipped with portable rapid metabolite test strips and a smart terminal, and regularly collects fecal SCFA and urinary oxidative stress indicators. Combined with daily diet and symptom records, it forms a dynamic monitoring data stream to achieve real-time tracking of key indicators during the intervention process. The effect quantification evaluation system 42 constructs a multi-dimensional effect index, the calculation formula of which is:
[0024]
[0025] Among them, weight - It adaptively adjusts to individual metabolic risk type and achieves objective and quantitative assessment of intervention effects through multidimensional effect indices, avoiding subjective judgment bias.
[0026] Furthermore, the AI self-learning iterative optimization module 5 includes a reinforcement learning optimizer 51 and a knowledge base update unit 52. The reinforcement learning optimizer 51 adopts a deep deterministic policy gradient algorithm, using a multidimensional effect index as a reward signal. It inputs the current nutrition plan, dynamic monitoring data, and individual response data to automatically adjust the proportion of nutritional components. The specific adjustment logic is as follows: when the individual responds well to the current plan, the plan is maintained and fine-tuned; when the response is poor, the bottleneck flora / metabolic pathway is identified and the functional components are replaced; when component intolerance occurs, the dosage of the stimulating component is reduced and the mild component is replaced. Every time a 4-week intervention cycle is completed, the knowledge base update unit 52 integrates the individual's detection data, intervention effect data, and plan adjustment data into the global gut microbiota-nutrition knowledge base, continuously optimizing the association model and plan generation rules, realizing dual iteration of collective intelligence and individual precision, and improving the overall accuracy and adaptability of the system.
[0027] In practical use, multiple users complete the collection of gut microbiota, metabolomics, physiological phenotype, and behavioral data through the gut microbiota detection unit 13, metabolomics detection unit 14, physiological phenotype unit 15, and behavioral data unit 16 of the data acquisition unit 11. After the data is standardized by the data standardization unit 12, it is transmitted to the gut microbiota metabolism association modeling module 2. The core microbiota-metabolite association network 21 generates the "Individual Gut Microbiota Metabolic Risk Map" and performs metabolic risk scoring through the metabolism-phenotype mapping model 22. The personalized nutrition plan generation module 3 generates a personalized nutrition formula based on the above data through the multi-objective optimization algorithm 32 and pushes it to the user. The user performs the intervention according to the nutrition formula, and at the same time, the real-time monitoring unit 41 regularly uploads monitoring data. The effect quantification evaluation system 42 calculates the MEI to evaluate the intervention effect. The AI self-learning iterative optimization module 5 automatically adjusts the nutrition plan according to the MEI and monitoring data. After each intervention cycle is completed, the knowledge base update unit 52 updates the global knowledge base to achieve continuous optimization of the plan until the user's gut microbiota metabolic status and physiological indicators reach the ideal level.
[0028] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.
[0029] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.
[0030] The present invention and its embodiments have been described above. This description is not restrictive, and the accompanying drawings are only one embodiment of the present invention; the actual structure is not limited thereto. In conclusion, if those skilled in the art are inspired by this description and design similar structures and embodiments without departing from the spirit of the invention, such designs should fall within the protection scope of the present invention.
Claims
1. A personalized nutrition program optimization system based on gut microbiota metabolism includes a multi-source data acquisition and standardization module (1), a gut microbiota metabolism association modeling module (2), a personalized nutrition program generation module (3), a dynamic monitoring and effect evaluation module (4), and an AI self-learning iterative optimization module (5). The multi-source data acquisition and standardization module (1), the gut microbiota metabolism association modeling module (2), the personalized nutrition program generation module (3), the dynamic monitoring and effect evaluation module (4), and the AI self-learning iterative optimization module (5) work together to complete the entire closed loop from multi-source data acquisition, gut microbiota metabolism association modeling, personalized program generation, to dynamic monitoring and evaluation and AI self-learning iteration, thereby achieving precise, dynamic, and intelligent optimization of the nutrition program.
2. The personalized nutrition program optimization system based on gut microbiota metabolism according to claim 1, characterized in that: The multi-source data acquisition and standardization module (1) includes a data acquisition unit (11) and a data standardization unit (12). The data acquisition unit (11) is integrated by a gut microbiota detection unit (13), a metabolomics detection unit (14), a physiological phenotype unit (15), and a behavioral data unit (16). The gut microbiota detection unit (13) uses three detection methods: 16S rRNA, metagenomic sequencing, and qPCR quantification to achieve accurate quantitative analysis of gut microbiota. The metabolomics detection unit (14) detects short-chain fatty acids, bile acids, tryptophan metabolites, and lipopolysaccharides in feces / blood. The physiological phenotype unit (15) collects physiological indicators such as weight, body fat, blood glucose, blood lipids, inflammatory factors, and intestinal symptom scores. The behavioral data unit (16) collects dietary logs, exercise, sleep, and stress-perceived behavioral data. The data standardization unit (12) is used to construct a unified data format, remove batch effects, fill missing values, and transform gut microbiota, metabolomics, physiological phenotype, and behavioral data into a standardized feature matrix, providing a unified data foundation for subsequent modeling and analysis.
3. The personalized nutrition program optimization system based on gut microbiota metabolism according to claim 2, characterized in that: 。 4. The personalized nutrition program optimization system based on gut microbiota metabolism according to claim 3, characterized in that: The gut microbiota metabolism association modeling module (2) includes a core microbiota-metabolite association network (21) and a metabolism-phenotype mapping model (22). The core microbiota-metabolite association network (21) uses a multivariate Bayesian network + LASSO regression algorithm to mine the quantitative causal relationship between the core driving microbiota and key metabolic pathways from microbiota OTU, species abundance and metabolite concentration data, and outputs the "Individual Gut Microbiota Metabolic Risk Map". The metabolism-phenotype mapping model (22) uses a random forest + XGBoost fusion model, inputs standardized gut microbiota metabolic feature data, and outputs an individual metabolic risk score, including four core risk scores: obesity risk, abnormal blood glucose risk, intestinal inflammation risk and intestinal barrier damage risk, providing a precise targeting basis for the generation of personalized nutrition plans.
5. The personalized nutrition program optimization system based on gut microbiota metabolism according to claim 4, characterized in that: The personalized nutrition plan generation module (3) includes a nutrient target library (31) and a multi-objective optimization algorithm (32). The nutrient target library (31) is a pre-built database containing 12 types of nutrients and 36 functional components, clearly recording the regulatory effects of each component on the core microbiota and metabolic pathways. The multi-objective optimization algorithm (32) uses "maximizing beneficial metabolites, minimizing harmful metabolites, matching individual dietary preferences, and avoiding food allergies" as its objective function and automatically generates personalized nutrition formulas using a non-dominated sorting genetic algorithm.
6. The personalized nutrition program optimization system based on gut microbiota metabolism according to claim 5, characterized in that: The dynamic monitoring and effect evaluation module (4) includes a real-time monitoring unit (41) and an effect quantification evaluation system (42). The real-time monitoring unit (41) is equipped with a portable metabolite rapid test strip and a smart terminal, and regularly collects fecal SCFA and urinary oxidative stress indicators. Combined with daily diet and symptom records, it forms a dynamic monitoring data stream. The effect quantification evaluation system (42) constructs a multidimensional effect index, the calculation formula of which is: Among them, weight - It is adaptively adjusted based on the individual's metabolic risk type.
7. The personalized nutrition program optimization system based on gut microbiota metabolism according to claim 6, characterized in that: The AI self-learning iterative optimization module (5) includes a reinforcement learning optimizer (51) and a knowledge base update unit (52). The reinforcement learning optimizer (51) adopts a deep deterministic policy gradient algorithm, uses a multidimensional effect index as a reward signal, inputs the current nutrition plan, dynamic monitoring data, and individual response data, and automatically adjusts the proportion of nutritional components. The knowledge base update unit (52) integrates the individual's detection data, intervention effect data, and plan adjustment data into the global gut microbiota-nutrition knowledge base after each four-week intervention cycle, and continuously optimizes the association model and plan generation rules.