Personalized health intervention system based on brain-intestine axis theory

Abstract

The invention provides a personalized health intervention system based on a braingut axis theory, and the specific operation comprises the following steps: S1, household standardized sampling and low-temperature transportation: providing a standardized household sampling suite, enabling a user to collect a feces sample, carrying out compliance storage and transportation, and transporting the sample to a detection point position; s2, performing high-throughput sequencing analysis on the intestinal flora, namely receiving the excrement sample acquired by the household sampling module, performing sequencing analysis on the intestinal microbiome in the sample through a high-throughput sequencing technology, and outputting composition, abundance and function annotation data of the intestinal flora; and S3, user emotion and sleep phenotype data acquisition: acquiring emotion and sleep related phenotype data autonomously reported by the user. According to the invention, convenient collection and stable transportation of intestinal flora samples are realized through the standardized household sampling suite, a user does not need to go to a professional institution, the detection threshold is reduced, and the user participation degree is improved.

Description

Technical Field

[0001] This invention relates to the field of medical technology, and in particular to a personalized health intervention system based on the brain-gut axis theory. Background Technology

[0002] People typically learn about their health status through physical examinations and carry out targeted interventions based on the results. However, traditional physical examinations have significant limitations: First, the examination cycle is generally once every six months or once a year, which is a long time span and makes it difficult to track health status in real time. Second, the interpretation of physical examination reports requires professional medical resources, which is difficult for ordinary users to understand on their own, resulting in a high threshold for health management. Third, for sub-health issues with hidden symptoms or early potential health risks, the improvement effect after intervention cannot be intuitively perceived in the short term. It often requires waiting for the next physical examination to verify the data, resulting in insufficient timeliness of feedback. Fourth, traditional physical examinations focus on the screening and treatment of organic diseases and lack specific assessments of mental health status such as emotions and sleep, making it even more difficult to provide targeted intervention guidance and failing to meet people's needs for managing holistic physical and mental health. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to overcome the defects of the existing technology. The present invention proposes a personalized health intervention system based on the brain-gut axis theory.

[0004] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A personalized health intervention system based on the gut-brain axis theory includes the following steps: Step S1, standardized home sampling and low-temperature transportation, providing a standardized home sampling kit for users to collect their own fecal samples, ensuring compliant preservation and transportation, and transporting the samples to the testing site; Step S2, high-throughput sequencing analysis of gut microbiota, receiving the fecal samples acquired by the home sampling module, performing sequencing analysis on the gut microbiome in the samples using high-throughput sequencing technology, and outputting gut microbiota composition, abundance, and functional annotation data; Step S3, user emotion and sleep phenotype data collection, collecting user-reported emotion and sleep-related phenotype data; Step S4, cloud platform data integration, building a cloud platform A to receive the microbiota data output by the high-throughput sequencing analysis module and read... Step S5 involves integrating user-reported mood and sleep-related phenotypic data after cleaning and standardization. Step S6 involves building a personalized gut-brain response model based on the integrated gut microbiota and phenotypic data using machine learning algorithms to establish a user-specific gut-brain response model and clarify the correlation between gut microbiota and mental health indicators. Step S7 involves generating and pushing personalized digital nutrition formulas based on the gut-brain response model, dynamically generating digital nutrition formulas containing customized dietary recommendations and probiotic combinations, and pushing them to the user. Step S8 involves dynamically updating the model and nutrition formulas, iteratively optimizing the gut-brain response model based on periodically submitted phenotypic data and gut microbiota data obtained from resequencing, and simultaneously updating the digital nutrition formulas.

[0005] Preferably, the sampling kit in step S1 includes a home sampling kit comprising a sterile sampling tube, a sample preservation solution, a sealed bag, and a sampling operation instruction manual. The sample preservation solution is a microbial preservation system free from nuclease contamination, which can maintain the stability of the intestinal flora structure for at least 72 hours.

[0006] Preferably, the equipment used for sample preservation in step S1 is a liquid nitrogen tank, with the sample stored in the center of the liquid nitrogen tank to extend the sample preservation time.

[0007] Preferably, the sequencing target region in step S2 covers the full-length 16S rRNA gene or the complete metagenomic sequence, and the analysis process includes nucleic acid extraction, library construction, sequencing, data filtering, and microbial community functional annotation.

[0008] Preferably, the emotion-related data collected in step S3 includes pleasure level, anxiety level, and stress level, and the sleep-related data includes sleep duration, sleep onset time, and sleep quality score.

[0009] Preferably, the cloud platform data integration in step S4 is achieved by collecting data through the user interface on cloud platform A, which supports regular updates and real-time uploads.

[0010] Preferably, the machine learning algorithm used in the personalized "gut-brain" response model construction in step S5 includes random forest, gradient boosting tree or neural network algorithm. The model training process uses gut microbiota data as input features and emotion and sleep-related phenotypic data as output labels, and optimizes model parameters through cross-validation.

[0011] Preferably, in step S6, the push method is to push the digital nutrition formula to the user's cloud platform account in the form of a graphic report, which supports online viewing, downloading and printing; at the same time, a reminder is pushed through SMS or mobile application to ensure that the user receives it in a timely manner.

[0012] Preferably, the periodic feedback time in step S7 is set to once a month to determine the changes in the gut microbiota after the user uses the nutritional formula, thereby facilitating the determination of subsequent nutritional formula updates.

[0013] In step S7, new applicable fungal strains and ingredient information are also updated.

[0014] Compared with the prior art, the beneficial effects of the present invention include: 1. This invention enables convenient collection and stable transportation of gut microbiota samples through a standardized home sampling kit, eliminating the need for users to visit professional institutions, thus lowering the testing threshold and increasing user participation.

[0015] 2. By integrating gut microbiota data obtained from high-throughput sequencing with user-reported phenotypic data, a personalized gut-brain response model is constructed using machine learning algorithms. This breaks through the limitations of the traditional "one-size-fits-all" intervention approach and enables precise analysis of the relationship between gut microbiota and mental health.

[0016] 3. Based on a dedicated model, it generates customized dietary recommendations and probiotic combinations, which can regulate the structure and function of the gut microbiota in a targeted manner, improve the user's mental health through the brain-gut axis pathway, and has a strong intervention and significant effect.

[0017] 4. It has a dynamic model update function, which can continuously optimize the intervention plan according to changes in user status, realize long-term and dynamic health management, has a wide range of applications, and can be promoted and applied to the daily mental health maintenance of the general population. Attached Figure Description

[0018] The disclosure of this invention is illustrated with reference to the accompanying drawings. It should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of protection of this invention. In the drawings, the same reference numerals are used to refer to the same parts. Wherein: Figure 1 This is a flowchart of a personalized health intervention system based on the brain-gut axis theory proposed in this invention. Detailed Implementation

[0019] It is readily understood that, based on the technical solution of this invention, those skilled in the art can propose various interchangeable structural methods and implementations without altering the essential spirit of the invention. Therefore, the following detailed embodiments and accompanying drawings are merely illustrative examples of the technical solution of this invention and should not be considered as the entirety of the invention or as limitations or restrictions on the technical solution of this invention.

[0020] According to one embodiment of the present invention, Figure 1 As shown.

[0021] A personalized health intervention system based on the brain-gut axis theory includes the following steps: Step S1: Standardized Home Sampling and Cryogenic Transportation We provide pre-sterilized, standardized home sampling kits, enabling users to independently collect, preserve, and transport fecal samples in compliance with regulations, ensuring that the gut microbiota structure remains stable when the samples arrive at the testing site. The sampling kit consists of: one sterile screw-cap sampling tube (10-15 mL capacity, with built-in graduations), 3 mL of nuclease-free microbial preservation solution (containing a compound preservative, pH 7.0-7.2, which can maintain the stability of the intestinal flora structure for ≥72 hours), a double-layer leak-proof sealed bag (with an outer layer of absorbent cotton to prevent sample leakage and contamination), and an illustrated sampling operation manual (including step diagrams, precautions, and transportation requirements). Core operating procedures: ① Following the instructions, the user collects a stool sample the size of a soybean using the small spoon inside the sampling tube, adds it to the preservation solution, tightens the cap, and inverts the tube 6 times to ensure thorough mixing of the sample and preservation solution; ② Place the sampling tube into the inner sealing bag, squeeze out the air, seal it, and then place it into the outer sealing bag for sealing; ③ If the transportation time exceeds 72 hours, the sealed sampling tube must be placed in the matching liquid nitrogen storage container (central sleeve structure, the inner cylinder holds the sampling tube, the outer cavity is filled with liquid nitrogen, maintaining a low temperature of -196℃, extending the sample preservation time to 7 days); ④ According to the cold chain transportation standards specified in the instructions (transportation temperature 0-4℃, avoid violent shaking), send the sample to the designated testing point, along with a unique sample identification code (linked to the user account); Step S2: High-throughput sequencing analysis of gut microbiota After receiving the samples, the testing site uses a standardized high-throughput sequencing process to complete the sequencing and bioinformatics analysis of the gut microbiome, outputting accurate microbial community data.

[0022] Sample pretreatment: ① After receiving the sample, first check the identification code to confirm that there is no leakage of the sample and that the preservation solution is not turbid; ② Place the sampling tube in a 4℃ refrigerator to thaw (if it is a liquid nitrogen frozen sample, it needs to be thawed in a -20℃ refrigerator for 2 hours before being transferred to 4℃ to thaw); ③ Take 1mL of sample suspension for subsequent nucleic acid extraction; Sequencing and Analysis Workflow: ① Nucleic Acid Extraction: Total microbial DNA is extracted from the sample using a microbial total DNA extraction kit. DNA purity is assessed using a nucleic acid purity testing device (A260 / A280 ratio 1.8-2.0), and DNA integrity is verified by agarose gel electrophoresis. ② Library Construction: PCR amplification, end repair, adapter ligation, and library quantification are performed on the target sequencing regions (full-length 16S rRNA using a long-read sequencing platform, and full-length metagenomic sequences using a short-read high-throughput sequencing platform). ③ Sequencing: Qualified libraries (concentration ≥10 ng / μL, fragment size meeting platform requirements) are loaded into the sequencer, and the sequencing process is run according to standard procedures to obtain raw sequencing data. ④ Data Filtering: Remove low-quality reads (Q30≤80%), adapter sequences, and host genome contamination sequences to obtain clean data; ⑤ Microbial community functional annotation: Perform OTU (operational taxonomic unit) clustering based on clean data, compare with microbial species classification databases for species classification and identification, and obtain gut microbiota composition (phylum, class, order, family, genus, species level) and abundance data; perform functional annotation through microbial functional annotation databases, and analyze functional information such as microbial metabolic pathways and enzyme activities; ⑥ Output results: Generate a "Gut Microbiota Sequencing Analysis Report", which includes microbial species abundance table, dominant microbial community ranking, functional pathway enrichment analysis, microbial community diversity indices (Shannon, Simpson indices), etc. Step S3: Collection of User Emotion and Sleep Phenotype Data Through the user interface on the cloud platform, standardized data on mood and sleep-related phenotypes reported by users are collected to ensure data authenticity and consistency.

[0023] Data collection channels: Users log in to the cloud platform (supports mobile applications, mini-programs, and web pages), complete the data entry in the designated modules, and the platform provides data entry guidelines (including indicator definition explanations). Data Collection Indicators and Standards: ① Emotion-related data: Pleasure level (1-5 points, 1 point = extremely unpleasant, 5 points = extremely pleasant), anxiety level (1-5 points, 1 point = no anxiety, 5 points = extreme anxiety), stress level (1-5 points, 1 point = no stress, 5 points = extremely stressful), which users need to fill in based on their average state over the past 7 days; ② Sleep-related data: Sleep duration (unit: hours, accurate to 0.1 hours), time to fall asleep (unit: minutes, accurate to 5 minutes), sleep quality score (1-5 points, 1 point = very poor, 5 points = excellent), which users need to fill in based on their sleep situation over the past 7 days; ③ Data verification: The platform sets logical verification rules (e.g., sleep duration must be within the range of 3-12 hours). If the entered data is abnormal, the user will be prompted to reconfirm.

[0024] Step S4: Cloud Platform Data Integration Build a cloud platform with data storage, cleaning, standardization and integration functions to achieve efficient fusion of microbial community sequencing data and phenotypic data; The platform's core functions include: ① Data reception: Automatically synchronizing sequencing data output from step S2 (interfacing with the sequencing system at the testing site via an interface), reading user-submitted emotion and sleep phenotype data in real time, and associating it with the user's unique identifier; ② Data cleaning: Removing outliers from sequencing data (such as rare species with abundance below 0.001%) and logically incorrect data from phenotype data (such as duplicate entries or data exceeding a reasonable range), and supplementing missing data (if phenotype data is partially missing, the platform will remind the user to fill it in; if not filled in within the deadline, it will be marked as "missing"); ③ Data standardization: Normalizing microbial abundance data (converting it to relative abundance), and standardizing phenotype data (converting 1-5 scale data to 0-1 range values), eliminating differences in data units; ④ Data integration: Building a user-specific database, associating standardized microbial and phenotype data by time dimension (marking time nodes such as first sampling and subsequent retesting), supporting regular data updates (real-time updates after users fill in phenotype data) and real-time uploads (automatically uploaded to the platform within 24 hours after sequencing data is generated).

[0025] Step S5: Construction of a Personalized Gut-Brain Response Model Based on multi-dimensional data integrated from the cloud platform, a user-specific "gut-brain" response model is constructed using machine learning algorithms to quantify the correlation between gut microbiota and mental health indicators.

[0026] Data preprocessing: Target data is extracted from the integrated database. Gut microbiota data (species abundance, functional pathway abundance) are used as input features, and emotion and sleep-related phenotypic data (standardized pleasure level, anxiety level, stress level, sleep duration, sleep onset time, and sleep quality score) are used as output labels to construct a training dataset (accounting for 80% of the total data volume) and a validation dataset (accounting for 20% of the total data volume).

[0027] Model Construction and Optimization: ① Algorithm Selection: Select a suitable machine learning algorithm based on data characteristics. Random forest (suitable for non-linear data, resistant to overfitting), gradient boosting tree (suitable for improving the accuracy of feature importance identification), or neural network algorithm (suitable for high-dimensional data, capturing complex correlations) can be selected. ② Model Training: Train the model using the training dataset, and optimize the model hyperparameters (such as the number of decision trees in random forest and the number of hidden layer nodes in neural network) using mean squared error (MSE) and coefficient of determination (R²) as evaluation indicators. ③ Cross-validation: Use 5-fold cross-validation to verify the model's generalization ability and ensure that the prediction error of the model is stable (R²≥0.7) on different data subsets. ④ Model Output: Generate a user-specific "gut-brain" response model, identify key related bacteria (such as beneficial bacteria genera negatively correlated with anxiety level and functional bacteria positively correlated with sleep quality) and functional pathways, and output correlation heatmaps and prediction reports.

[0028] Step S6: Personalized Digital Nutrition Formula Generation and Delivery Based on the analysis results of the "gut-brain" response model and combined with individual user characteristics, a customized digital nutrition formula is dynamically generated and pushed to users through the platform.

[0029] Formula generation logic: ① Microbiome optimization goals: Targeting key related microbiome groups identified by the model, determine the direction of nutritional intervention (e.g., if the abundance of beneficial bacteria is low, increase probiotic supplementation; if the ratio of dominant bacteria is unbalanced, adjust the dietary structure); ② Customized dietary recommendations: Based on user dietary preferences (collected in advance by the platform), recommend suitable ingredients (e.g., whole grains rich in dietary fiber, prebiotic ingredients that promote the proliferation of probiotics), daily intake and cooking methods, and identify ingredients that need to be reduced (e.g., high-sugar, high-fat ingredients); ③ Probiotic combination design: Based on the microbiome gap, recommend suitable probiotic strains, daily dosage (including CFU value) and time of administration (e.g., taken with warm water half an hour after meals); ④ Other intervention suggestions: Supplement with auxiliary suggestions such as sleep and rest adjustments and stress relief methods to synergistically improve the function of the gut-brain axis.

[0030] Push method: Digital nutrition formulas are pushed to users' cloud platform accounts in the form of graphic reports, which can be viewed, downloaded and printed online; at the same time, reminders are pushed via SMS or mobile application to ensure that users receive them in a timely manner.

[0031] Step S7: Dynamic updates of the model and nutritional formula Establish a regular feedback and retesting mechanism to iteratively optimize the gut-brain response model and digital nutrition formula based on the latest user data, so as to achieve continuous adaptation of personalized intervention; Regular Feedback and Retesting Arrangements: ① Phenotypic Data Feedback: A monthly feedback cycle is set. Users are required to fill in their emotional and sleep phenotypic data for the past 30 days through the cloud platform within 7 days after the end of the cycle (indicators are the same as in step S3); ② Microbiome Data Retesting: It is recommended that users retest their fecal samples every 3 months (sampling and transportation process is the same as in step S1). After the testing site completes high-throughput sequencing, the latest microbiome data will be synchronized to the cloud platform; ③ Special Circumstances Handling: If users report significant changes in their phenotypic data (such as an increase in anxiety level from 2 to 4), they can apply for early microbiome retesting; Model and Formula Updates: ① Data Integration: The platform incorporates the latest phenotypic data and retested gut microbiota data into the user database, and performs cleaning and standardization according to the standards in step S4; ② Model Iteration: Based on the updated dataset, the gut-brain response model is retrained and optimized, and the weights of key associated microbiota and functional pathways are adjusted; ③ Formula Updates: Based on the iterated model results, the digital nutrition formula is dynamically adjusted (e.g., adjusting the combination of probiotic strains, optimizing the types and intake of recommended foods), generating an updated formula report and pushing it to users, while also highlighting the key update points (e.g., "Due to the increased abundance of target beneficial bacteria, the probiotic dosage is reduced; food recommendations rich in calming active ingredients are added"). Network Updates: Information on newly applicable probiotic strains and functional food ingredients emerging in the medical field is systematically collected, categorized, stored, and managed in a standardized manner, providing accurate and scalable data support for the system to continuously optimize and upgrade as medical technologies iterate.

[0032] The technical scope of this invention is not limited to the content described above. Those skilled in the art can make various modifications and variations to the above embodiments without departing from the technical concept of this invention, and all such modifications and variations should fall within the protection scope of this invention.

Claims

1. A personalized health intervention system based on the brain-gut axis theory, characterized in that, The specific operation includes the following steps: Step S1, standardized home sampling and low-temperature transportation, is used to provide standardized home sampling kits for users to collect their own fecal samples and to compliantly preserve and transport the samples to the testing site. Step S2, high-throughput sequencing analysis of gut microbiota, is used to receive fecal samples obtained by the home sampling module, and to perform sequencing analysis on the gut microbiome in the sample using high-throughput sequencing technology, outputting gut microbiota composition, abundance and functional annotation data. Step S3: Collect user emotion and sleep phenotype data, collect emotion and sleep-related phenotype data reported by users themselves; Step S4, cloud platform data integration, is used to build cloud platform A, receive the microbial community data output by the high-throughput sequencing analysis module, and read the mood and sleep-related phenotypic data reported by the user. The two types of data are cleaned and standardized before integration is completed. Step S5: Construction of a personalized "gut-brain" response model. This model is used to build a user-specific "gut-brain" response model based on the integrated gut microbiota data and phenotypic data, using machine learning algorithms to clarify the correlation between gut microbiota and mental health indicators. Step S6, Personalized digital nutrition formula generation and push, is used to dynamically generate a digital nutrition formula containing customized dietary advice and probiotic combination based on the "gut-brain" response model, and push it to the user; Step S7 involves dynamic updates to the model and nutritional formula. This is used to iteratively optimize the "gut-brain" response model based on phenotypic data regularly provided by users and gut microbiota data obtained from resequencing, while simultaneously updating the digital nutritional formula.

2. The personalized health intervention system based on the brain-gut axis theory according to claim 1, characterized in that, The sampling kit mentioned in step S1 includes a home sampling kit containing sterile sampling tubes, sample preservation solution, a sealed bag, and sampling operation instructions. The sample preservation solution is a microbial preservation system free from nuclease contamination, which can maintain the stability of the intestinal flora structure for at least 72 hours.

3. The personalized health intervention system based on the brain-gut axis theory according to claim 1, characterized in that, In step S1, the equipment used for sample preservation is a liquid nitrogen tank, which stores the sample in the center to extend the sample preservation time.

4. The personalized health intervention system based on the brain-gut axis theory according to claim 1, characterized in that, The target region for sequencing in step S2 covers the full-length 16S rRNA gene or the complete metagenomic sequence. The analysis process includes nucleic acid extraction, library construction, sequencing, data filtering, and microbial community functional annotation.

5. The personalized health intervention system based on the brain-gut axis theory according to claim 1, characterized in that, The emotion-related data collected in step S3 includes pleasure level, anxiety level, and stress level, while the sleep-related data includes sleep duration, sleep onset time, and sleep quality score.

6. The personalized health intervention system based on the brain-gut axis theory according to claim 1, characterized in that, The cloud platform data integration in step S4 involves data collection through the user interface of the A cloud platform, which supports regular updates and real-time uploads.

7. The personalized health intervention system based on the brain-gut axis theory according to claim 1, characterized in that, The personalized "gut-brain" response model construction in step S5 uses machine learning algorithms such as random forest, gradient boosting tree, or neural network algorithm. The model training process uses gut microbiota data as input features and emotion and sleep-related phenotypic data as output labels, and optimizes model parameters through cross-validation.

8. The personalized health intervention system based on the brain-gut axis theory according to claim 1, characterized in that, In step S6, the digital nutrition formula is pushed to the user's cloud platform account in the form of a graphic report, which can be viewed, downloaded and printed online; at the same time, reminders are pushed via SMS or mobile application to ensure that the user receives them in a timely manner.

9. The personalized health intervention system based on the brain-gut axis theory according to claim 1, characterized in that, The periodic feedback time in step S7 is set to once a month to determine the changes in the gut microbiota after the user uses the nutritional formula, thereby facilitating the determination of subsequent nutritional formula updates.

10. The personalized health intervention system based on the brain-gut axis theory according to claim 1, characterized in that, In step S7, new applicable fungal strains and ingredient information are also updated.

Images