In-vivo water-soluble vitamin dynamic monitoring method based on multi-modal biosensing

Abstract

The invention provides an in-vivo water-soluble vitamin dynamic monitoring method based on multi-modal biosensing, and relates to the multi-modal technical field, and the in-vivo water-soluble vitamin dynamic monitoring method comprises the following steps: collecting in-vivo signals through a wearable sensing terminal, processing the in-vivo signals, transmitting the processed in-vivo signals to a user terminal, establishing an individualized mapping relation, combining dietary intake information and urine load test data, and determining the in-vivo water-soluble vitamin dynamic monitoring result. According to the method, calibration and analysis of vitamin concentration time sequence data are achieved, a comprehensive nutrition state evaluation result is generated, the problems that traditional vitamin detection is discontinuous and low in precision are solved, and real-time, accurate and dynamic monitoring of in-vivo water-soluble vitamins is achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of multi-modal technology, and in particular to an in-vivo water-soluble vitamin dynamic monitoring method based on multi-modal biosensing. BACKGROUND

[0002] Water-soluble vitamins are essential nutrients for the human body, playing an important role in metabolism and various physiological functions. Unlike fat-soluble vitamins, water-soluble vitamins are not easily stored in the body and need to be continuously ingested through daily meals. Traditional vitamin level evaluation mainly relies on blood sample analysis, dietary questionnaires, and clinical symptom observation methods. With the development of wearable devices and biosensing technology, real-time monitoring of in-vivo water-soluble vitamin levels becomes possible, providing new technical means for individualized nutrition management.

[0003] Currently, there are the following deficiencies in the monitoring of in-vivo water-soluble vitamins:

[0004] Existing vitamin detection methods mainly rely on laboratory blood tests, requiring blood sampling and professional analysis equipment, and cannot achieve continuous monitoring, only providing static information at a certain time point, making it difficult to reflect the dynamic change characteristics of vitamins in the body and unable to meet the needs of individualized nutrition management.

[0005] Traditional nutrition status evaluation relies too much on subjective reported dietary intake, with memory bias and estimation error, and cannot accurately quantify the actual absorption and utilization of vitamins. Although the urine load test can assess vitamin metabolism, the existing method is complex to operate and needs to be completed in a medical institution, which is not convenient for daily use, limiting its widespread application. SUMMARY

[0006] The embodiments of the present application provide an in-vivo water-soluble vitamin dynamic monitoring method based on multi-modal biosensing, which can solve the problems in the prior art.

[0007] In a first aspect, the embodiments of the present application provide an in-vivo water-soluble vitamin dynamic monitoring method based on multi-modal biosensing, comprising:

[0008] The biological sensor in the wearable sensing terminal continuously collects in-vivo interstitial fluid or transdermal signals as raw signals, filters and digitizes the raw signals, and generates digitized concentration signals;

[0009] The digitized concentration signals are transmitted to the user terminal through wireless communication;

[0010] The user terminal receives external calibration data, establishes an individualized mapping relationship between the digitized concentration signal and the blood vitamin concentration, and performs calibration conversion on the digitized concentration signal according to the individualized mapping relationship to generate calibrated vitamin concentration time series data.

[0011] Dietary intake information is obtained, and a vitamin intake estimate is calculated according to the dietary intake information by accessing a food composition database of a cloud service platform.

[0012] After receiving a urine load test start instruction, the time when the vitamin preparation is taken is recorded as the test start time point, and the calibrated vitamin concentration time series data is continuously collected within a preset test time period.

[0013] After the end of the preset test time period, the total volume of urine input by the user is received, the calibrated vitamin concentration time series data within the preset test time period is subjected to time integration operation, and the total amount of vitamin excretion within the time period is calculated in combination with the total volume of urine to generate a urine load test report.

[0014] The calibrated vitamin concentration time series data and the vitamin intake estimate are transmitted to a cloud service platform, and fusion analysis is performed on the calibrated vitamin concentration time series data and the vitamin intake estimate based on a nutritional status evaluation model to generate a comprehensive nutritional status evaluation result.

[0015] The biosensor includes at least two sensing units for different water-soluble vitamins, each sensing unit adopts electrochemical sensing technology or optical sensing technology to generate a raw signal corresponding to a different water-soluble vitamin.

[0016] The user terminal receives external calibration data, establishes an individualized mapping relationship between the digitized concentration signal and the blood vitamin concentration, and performs calibration conversion on the digitized concentration signal according to the individualized mapping relationship to generate calibrated vitamin concentration time series data, which includes:

[0017] The user terminal receives external calibration data input by the user, the external calibration data including a blood vitamin concentration measured value obtained through in-vitro detection and a sampling time corresponding to the blood vitamin concentration measured value;

[0018] According to the sampling time, the sensor output signal value at the corresponding time is extracted from the digitized concentration signal, and a calibration data pair is established between the sensor output signal value and the blood vitamin concentration measured value.

[0019] A regression function is constructed based on the calibration data pair, the regression function representing an individualized mapping relationship between the digitized concentration signal and the blood vitamin concentration.

[0020] inputting the digitized concentration signal into the regression function for conversion operation, and outputting a sequence of blood vitamin concentration estimation values corresponding to the digitized concentration signal in time sequence as the calibrated vitamin concentration time series data.

[0021] performing time integration operation on the calibrated vitamin concentration time series data in the preset test time period, and combining the total urine volume to calculate the total vitamin excretion in the time period, including:

[0022] extracting all data points in the preset test time period from the calibrated vitamin concentration time series data, the all data points including a plurality of sampling time points and corresponding vitamin concentration values;

[0023] taking the time interval between adjacent sampling time points as the integration step, and taking the corresponding vitamin concentration value as the concentration representative value in the time interval, calculating the concentration and time product in each time interval to obtain a plurality of time interval concentration product values;

[0024] performing accumulation summation operation on the plurality of time interval concentration product values to obtain the vitamin concentration time integral value in the preset test time period;

[0025] performing multiplication operation on the vitamin concentration time integral value and the total urine volume to obtain the total vitamin excretion in the time period.

[0026] performing fusion analysis on the calibrated vitamin concentration time series data and the vitamin intake estimation value based on a nutritional status evaluation model to generate comprehensive nutritional status evaluation results, including:

[0027] extracting feature parameters representing the dynamic changes of vitamins in the body from the calibrated vitamin concentration time series data, the feature parameters including concentration mean value, concentration fluctuation amplitude and concentration trough duration;

[0028] inputting the feature parameters and the vitamin intake estimation value as input variables into the nutritional status evaluation model;

[0029] the nutritional status evaluation model calculates a vitamin metabolism state index in the body based on the feature parameters, the vitamin metabolism state index representing the utilization efficiency and storage capacity of the body to vitamins;

[0030] the nutritional status evaluation model calculates a dietary supply adequacy index based on the vitamin intake estimation value, the dietary supply adequacy index representing the satisfaction degree of external intake to the body's needs;

[0031] The nutrition state evaluation model performs weighted comprehensive operation on the in-vivo vitamin metabolism state index and the dietary supply sufficiency index, and generates a comprehensive nutrition state evaluation result, which represents the overall level of vitamin nutrition of the body.

[0032] The method further comprises:

[0033] acquiring a preset vitamin physiological requirement reference value, which represents the vitamin concentration level required by the body to maintain normal physiological functions;

[0034] performing ratio operation on the concentration mean value in the characteristic parameter and the vitamin physiological requirement reference value to obtain a concentration sufficiency component;

[0035] performing normalization operation on the concentration fluctuation amplitude in the characteristic parameter and the vitamin physiological requirement reference value to obtain a metabolic stability component;

[0036] performing comparison operation on the concentration trough duration in the characteristic parameter and a preset safety duration threshold to obtain a deficiency risk component;

[0037] performing weighted summation operation on the concentration sufficiency component, the metabolic stability component and the deficiency risk component to obtain the in-vivo vitamin metabolism state index;

[0038] acquiring a preset vitamin daily recommended intake standard value;

[0039] performing ratio operation on the vitamin intake estimation value and the vitamin daily recommended intake standard value to obtain the dietary supply sufficiency index.

[0040] In a second aspect of the embodiment, an in-vivo water-soluble vitamin dynamic monitoring system based on multi-modal biosensing is provided, comprising:

[0041] A first unit is configured to continuously collect interstitial fluid or transdermal signals in the body as original signals through a biosensor in a wearable sensing terminal, perform filtering and analog-digital conversion processing on the original signals, and generate digitized concentration signals;

[0042] A second unit is configured to transmit the digitized concentration signals to a user terminal through wireless communication;

[0043] A third unit is configured to receive external calibration data in the user terminal, establish an individualized mapping relationship between the digitized concentration signals and blood vitamin concentration, perform calibration conversion on the digitized concentration signals according to the individualized mapping relationship, and generate calibrated vitamin concentration time series data;

[0044] The fourth unit is configured to acquire dietary intake information, and calculate a vitamin intake estimation value according to the dietary intake information by accessing a food composition database of a cloud service platform.

[0045] The fifth unit is configured to record a vitamin preparation taking time as a test starting time point after receiving a urine load test starting instruction, and continuously collect the calibrated vitamin concentration time series data within a preset test time period.

[0046] The sixth unit is configured to receive a total urine volume input by a user after the end of the preset test time period, perform a time integration operation on the calibrated vitamin concentration time series data within the preset test time period, calculate a total vitamin excretion amount within the time period in combination with the total urine volume, and generate a urine load test report.

[0047] The seventh unit is configured to transmit the calibrated vitamin concentration time series data and the vitamin intake estimation value to a cloud service platform, perform fusion analysis on the calibrated vitamin concentration time series data and the vitamin intake estimation value based on a nutritional status evaluation model, and generate a comprehensive nutritional status evaluation result.

[0048] The third aspect of the embodiment of the application,

[0049] An electronic device is provided, comprising:

[0050] a processor;

[0051] a memory for storing processor-executable instructions;

[0052] The processor is configured to invoke the instructions stored in the memory to perform the method described above.

[0053] The fourth aspect of the embodiment of the application,

[0054] A computer-readable storage medium is provided, which stores computer program instructions, and the computer program instructions are executed by a processor to implement the method described above.

[0055] The beneficial effects of the present application are as follows:

[0056] The present application realizes continuous monitoring of water-soluble vitamins in the body through a wearable sensing terminal, establishes individual mapping relationships for data calibration, overcomes the shortcomings of traditional detection methods, such as strong discontinuity and inability to reflect dynamic changes of vitamins, and provides a more comprehensive and accurate evaluation method for individual vitamin status.

[0057] The present application combines dietary intake information with urine load test, realizes two-way evaluation of intake and excretion, generates a comprehensive nutritional status evaluation report through data fusion analysis of the cloud platform, improves the scientificity and individuality of vitamin level monitoring, and provides data support for precise nutrition intervention.

[0058] The non-invasive and continuous monitoring method of the present application significantly improves user experience, reduces monitoring cost, solves the pain and inconvenience brought by traditional blood and urine detection, and provides real-time nutritional status feedback for users through intelligent analysis, which helps users to self-manage and maintain health in daily vitamin intake. BRIEF DESCRIPTION OF DRAWINGS

[0059] Figure 1 A flowchart of the in-vivo water-soluble vitamin dynamic monitoring method based on multi-modal biosensing of the embodiment of the present application is shown. DETAILED DESCRIPTION

[0060] To make the objectives, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described below in detail with reference to the drawings of the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0061] The technical solutions of the present application will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0062] Figure 1 A flowchart of the in-vivo water-soluble vitamin dynamic monitoring method based on multi-modal biosensing of the embodiment of the present application is shown. Figure 1 As shown in the figure, the method comprises:

[0063] The biological sensor in the wearable sensing terminal continuously collects interstitial fluid or transdermal signals in the body as raw signals, and performs filtering and analog-to-digital conversion processing on the raw signals to generate digitalized concentration signals;

[0064] The digitalized concentration signals are transmitted to the user terminal through wireless communication;

[0065] External calibration data is received at the user terminal, an individualized mapping relationship between the digitalized concentration signals and blood vitamin concentration is established, the digitalized concentration signals are calibrated and converted according to the individualized mapping relationship, and calibrated vitamin concentration time series data is generated;

[0066] Obtaining dietary intake information, accessing a food composition database of a cloud service platform, and calculating a vitamin intake estimate based on the dietary intake information;

[0067] After receiving the urine load test start instruction, recording the vitamin preparation taking time as the test starting time point, and continuously collecting the calibrated vitamin concentration time series data in the preset test time period;

[0068] After the end of the preset test time period, receiving the total urine volume input by the user, performing time integration operation on the calibrated vitamin concentration time series data in the preset test time period, combining the total urine volume to calculate the total vitamin excretion in the time period, and generating a urine load test report;

[0069] Transmitting the calibrated vitamin concentration time series data and the vitamin intake estimate to the cloud service platform, performing fusion analysis on the calibrated vitamin concentration time series data and the vitamin intake estimate based on a nutritional status evaluation model, and generating a comprehensive nutritional status evaluation result.

[0070] In an optional embodiment, the biosensor includes at least two sensing units for different water-soluble vitamins, each sensing unit adopts an electrochemical sensing technology or an optical sensing technology to generate a raw signal corresponding to a different water-soluble vitamin.

[0071] In an optional embodiment, the user terminal receives external calibration data to establish an individualized mapping relationship between the digitized concentration signal and the blood vitamin concentration, and performs calibration conversion on the digitized concentration signal according to the individualized mapping relationship to generate calibrated vitamin concentration time series data, which includes:

[0072] The user terminal receives external calibration data input by the user, the external calibration data including a blood vitamin concentration measured value obtained through in vitro detection and a sampling time corresponding to the blood vitamin concentration measured value;

[0073] According to the sampling time, the sensor output signal value at the corresponding time is extracted from the digitized concentration signal, and a calibration data pair is established between the sensor output signal value and the blood vitamin concentration measured value;

[0074] Based on the calibration data pair, a regression function is constructed, which represents the individualized mapping relationship between the digitized concentration signal and the blood vitamin concentration;

[0075] The digitized concentration signal is input into the regression function for conversion operation, and a blood vitamin concentration estimate value sequence corresponding to the digitized concentration signal in time sequence is output as the calibrated vitamin concentration time series data.

[0076] The user terminal can receive external calibration data input by the user through its user interface. External calibration data includes measured blood vitamin concentration values ​​obtained through in vitro testing and the corresponding sampling time. For example, a user can have a vitamin D test performed in a laboratory or hospital, obtaining a blood vitamin D concentration of 35 ng / mL at a sampling time of 10:30 AM on June 15, 2023. The user then inputs this vitamin D concentration value of 35 ng / mL and the corresponding sampling time of 10:30 AM on June 15, 2023 into the calibration data input interface of the user terminal. The user terminal stores this data in its local database as external calibration data.

[0077] The user terminal extracts the sensor output signal value corresponding to the sampling time from the digitized concentration signal. The user terminal first queries the locally stored time-series data of the digitized concentration signal to locate the record matching the sampling time. If no record perfectly matches the sampling time in the time series, the user terminal can estimate the sensor output value at the sampling time using interpolation. For example, if the digitized concentration signal value is 0.85 volts at 10:25 AM on June 15, 2023, and 0.89 volts at 10:35 AM, then linear interpolation can be used to calculate the sensor output signal value at 10:30 AM as 0.87 volts.

[0078] The user terminal establishes calibration data pairs between sensor output signal values ​​and measured blood vitamin concentration values. The user terminal combines the sensor output signal value extracted in the previous step with the user-input measured blood vitamin concentration value to form a calibration data pair. For example, pairing the sensor output signal value of 0.87 volts with a vitamin D concentration of 35 ng / mL creates a calibration data pair (0.87 volts, 35 ng / mL). The user terminal can continuously collect multiple calibration data pairs to form a calibration dataset. For example, a calibration dataset containing 5 calibration data pairs could be: (0.78 volts, 28 ng / mL), (0.87 volts, 35 ng / mL), (0.95 volts, 42 ng / mL), (1.03 volts, 49 ng / mL), and (1.12 volts, 56 ng / mL).

[0079] The user terminal constructs a regression function based on calibration data. This regression function characterizes the individualized mapping relationship between the digitized concentration signal and blood vitamin D concentration. The user terminal can employ various regression algorithms, such as linear regression, multinomial regression, or piecewise linear regression, selecting the most suitable regression model based on the characteristics of the calibration dataset. Taking linear regression as an example, the user terminal analyzes the data pairs in the calibration dataset and calculates the slope and intercept of the best-fit line. Assuming the calculated regression function is: Blood Vitamin D Concentration (ng / mL) = 82.35 × Sensor Output Signal Value (volts) - 36.64, this function indicates that for every 0.01 volt increase in the sensor output signal value, the estimated blood vitamin D concentration increases by 0.8235 ng / mL.

[0080] The user terminal inputs the digitized concentration signal into a regression function for conversion and calculation. The user terminal inputs the entire time series of stored digitized concentration signals one by one into the constructed regression function to calculate the corresponding estimated blood vitamin concentration values. For example, for a digitized concentration signal sequence containing 24 hours of data, with one sampling point every 10 minutes, totaling 144 sampling points, the user terminal substitutes the signal values ​​of these 144 sampling points into the regression function to obtain 144 estimated blood vitamin concentration values. If the sensor output signal value at a certain time point is 0.92 volts, the estimated blood vitamin D concentration obtained after substituting it into the regression function is: 82.35 × 0.92 - 36.64 = 39.12 ng / mL.

[0081] The user terminal outputs a time-series sequence of estimated blood vitamin concentrations, corresponding one-to-one with the digitized concentration signals. This estimated sequence serves as the calibrated vitamin concentration time-series data. For example, the calibrated vitamin concentration time-series data might look like this: June 15, 2023, 8:00 AM, 32.5 ng / mL; 8:10 AM, 33.1 ng / mL; 8:20 AM, 34.7 ng / mL; and so on, until the entire 24-hour vitamin concentration change is recorded. The user terminal can display this data on the interface in tabular or graphical form, or export it as a file for further analysis.

[0082] Through the aforementioned technical means, user terminals can establish an individualized mapping relationship between digital concentration signals and actual blood vitamin concentrations using a small amount of external calibration data. This allows continuously collected sensor signals to be accurately converted into physiologically meaningful vitamin concentration values, improving the accuracy and personalization of vitamin monitoring.

[0083] In one optional implementation, the calibrated vitamin concentration time-series data within the preset test time period is integrated over time, and the total vitamin excretion during that time period is calculated in conjunction with the total urine volume, including:

[0084] Extract all data points within the preset experimental time period from the calibrated vitamin concentration time-series data. The total data points include multiple sampling times and their corresponding vitamin concentration values.

[0085] Using the time interval between adjacent sampling times as the integration step size and the corresponding vitamin concentration value as the representative concentration value within that time interval, the product of concentration and time is calculated for each time interval to obtain multiple time interval concentration product values.

[0086] The vitamin concentration time-time integral value is obtained by summing the product values ​​of the multiple time intervals.

[0087] The total amount of vitamins excreted during that time period is obtained by multiplying the time integral value of the vitamin concentration with the total volume of urine.

[0088] The calculation method comprises three main steps: data extraction, time integration, and final discharge calculation. In the data extraction stage, all data points within a preset experimental time period are extracted from the calibrated vitamin concentration time-series data. These data points include multiple sampling times and corresponding vitamin concentration values. During time integration, the system sets the time interval between adjacent sampling times as the integration step size and uses the corresponding vitamin concentration value as the representative concentration value within that time interval. For each time interval, the product of concentration and time is calculated to obtain multiple time interval concentration product values. Specifically, trapezoidal integration or rectangular integration can be used.

[0089] The concentration products at all time intervals are summed to obtain the vitamin concentration over time for the preset experimental period. Finally, the vitamin concentration over time is multiplied by the total urine volume to calculate the total amount of vitamins excreted during that period.

[0090] In practical applications, the total urine volume can be obtained by collecting all urine samples from the subject during the experiment and measuring their volume. In cases where complete urine collection is not possible, the total volume can be calculated by measuring the urine volume at each sampling point and then summing the results. Furthermore, to improve calculation accuracy, a more frequent sampling frequency and a shorter sampling time interval can be used, resulting in more accurate integral calculations.

[0091] Using the above calculation method, researchers can accurately measure the vitamin excretion characteristics of different subjects under various conditions, providing important reference data for personalized nutritional supplementation and the diagnosis and treatment of related diseases. The advantage of this method is that it can comprehensively capture the dynamic process of vitamin excretion, avoiding the random errors caused by single sampling and providing more reliable vitamin metabolism data.

[0092] In one optional implementation, the calibrated vitamin concentration time-series data and the estimated vitamin intake are fused and analyzed based on a nutritional status assessment model to generate a comprehensive nutritional status assessment result, including:

[0093] Characteristic parameters representing the dynamic changes of vitamins in vivo were extracted from the calibrated vitamin concentration time-series data. These characteristic parameters included the average concentration, the amplitude of concentration fluctuations, and the duration of concentration troughs.

[0094] The characteristic parameters and the estimated vitamin intake are used as input variables and input into the nutritional status assessment model.

[0095] The nutritional status assessment model calculates in vivo vitamin metabolism status indicators based on the aforementioned characteristic parameters. These in vivo vitamin metabolism status indicators characterize the body's utilization efficiency and storage capacity of vitamins.

[0096] The nutritional status assessment model calculates a dietary adequacy index based on the estimated vitamin intake, and this index characterizes the degree to which external intake meets the body's needs.

[0097] The nutritional status assessment model performs a weighted comprehensive calculation on the in vivo vitamin metabolism status indicators and the dietary supply adequacy indicators to generate a comprehensive nutritional status assessment result, which characterizes the overall level of vitamin nutrition in the body.

[0098] In this embodiment, the system performs a fusion analysis on the calibrated vitamin concentration time-series data and the estimated vitamin intake based on the nutritional status assessment model to generate a comprehensive nutritional status assessment result.

[0099] The system extracts characteristic parameters representing the dynamic changes of vitamins in vivo from calibrated vitamin concentration time-series data. These characteristic parameters include the average concentration, concentration fluctuation range, and duration of concentration troughs. Specifically, for each user's calibrated vitamin A concentration data, the system calculates the average concentration over 30 days, e.g., a user's average vitamin A concentration is 1.2 μmol / L; it calculates the difference between the highest and lowest concentrations as the fluctuation range, e.g., this user's vitamin A fluctuation range is 0.5 μmol / L; and it counts the number of consecutive days the concentration is below the recommended threshold (e.g., vitamin A is 0.7 μmol / L) as the duration of the trough, e.g., this user's vitamin A trough duration is 3 days. The system uses the same method to extract characteristic parameters for other vitamin types such as vitamin D and vitamin E.

[0100] The system uses the aforementioned characteristic parameters and estimated vitamin intake as input variables into the nutritional status assessment model. The input variables form a feature vector containing the average concentration, fluctuation range, duration of troughs, and corresponding estimated daily vitamin intake for each vitamin. For example, for vitamin A, the input features include: average concentration 1.2 μmol / L, fluctuation range 0.5 μmol / L, trough duration 3 days, and estimated daily intake of 700 μg retinol equivalent.

[0101] The nutritional status assessment model calculates vitamin metabolic status indicators based on characteristic parameters, which characterize the body's utilization efficiency and storage capacity of vitamins. Specifically, the model analyzes the position of the mean concentration relative to the population reference range to assess basal nutritional levels; analyzes the ratio of concentration fluctuation amplitude to the population standard fluctuation range to assess homeostatic regulation capacity; and analyzes the relationship between the duration of troughs and critical values ​​(e.g., 5 days for vitamin A) to assess reserve mobilization capacity. These assessment results are converted into metabolic status indicators through pre-defined calculation rules. For example, for vitamin A, a user's calculated metabolic status indicator is 0.82 (out of 1), indicating that the user's vitamin A metabolism and storage status is good.

[0102] The nutritional status assessment model also calculates a dietary adequacy index based on estimated vitamin intake, which characterizes the degree to which external intake meets the body's needs. Specifically, the system compares the user's estimated daily vitamin intake with the recommended intake for their corresponding age and gender, calculating the satisfaction rate. For example, the recommended intake of vitamin A for a 35-year-old adult male is 800 μg retinol equivalents / day, and the user's daily intake is 700 μg, resulting in a satisfaction rate of 87.5%. Considering the differences in the physiological importance of different vitamins, the system applies weight adjustments to the satisfaction rate to generate the dietary adequacy index. This user's vitamin A dietary adequacy index is 0.85 (out of 1), indicating that the dietary supply is basically sufficient but slightly insufficient.

[0103] The nutritional status assessment model performs a weighted comprehensive calculation on in vivo vitamin metabolic status indicators and dietary supply adequacy indicators to generate a comprehensive nutritional status assessment result. Specifically, the system sets weight coefficients for different vitamin types, such as vitamin A with a weight of 0.15 and vitamin D with a weight of 0.2, reflecting the degree of influence of each vitamin on overall health. Simultaneously, in vivo metabolic status indicators and dietary supply adequacy indicators are weighted in a 6:4 ratio, reflecting the relative importance of internal environmental status and external supply. Through this weighted calculation, the system generates a comprehensive score between 0 and 1, mapping the score to nutritional status levels: 0.9-1.0 is excellent, 0.8-0.9 is good, 0.7-0.8 is average, 0.6-0.7 is insufficient, and below 0.6 is deficient. This user's comprehensive vitamin A score is 0.83, corresponding to a "good" level.

[0104] Based on a comprehensive score of all vitamin types, the system generates a detailed nutritional status assessment report. This report includes: a comprehensive score and level display for each vitamin; an analysis of nutritional status strengths, such as "Sufficient vitamin E reserves, strong antioxidant capacity"; an analysis of nutritional status deficiencies, such as "Slightly low vitamin D levels, recommendation to increase outdoor activity time"; and targeted improvement suggestions, such as "Increase intake of dark green vegetables to improve vitamin A levels." Users can view the assessment report through a mobile application interface. The system updates the comprehensive assessment results every 30 days to help users track changes in their nutritional status.

[0105] Furthermore, the assessment model also considers individual differences, including age, sex, and physiological stage. For example, for pregnant women, the weighting coefficient for vitamin folic acid is adjusted to 0.25 to reflect its special importance during pregnancy; for the elderly, the assessment standard for vitamin absorption and utilization efficiency is lowered to account for age-related physiological changes. This personalized assessment method makes the assessment results more accurate and tailored to the actual needs of different population groups.

[0106] In one optional implementation, the method further includes:

[0107] Obtain a preset baseline value for vitamin physiological requirements, which represents the concentration level of vitamins required by the body to maintain normal physiological functions;

[0108] The concentration mean value in the characteristic parameters is compared with the baseline value of vitamin physiological requirements to obtain the concentration satisfaction component.

[0109] The concentration fluctuation range in the characteristic parameters is normalized to the baseline value of vitamin physiological requirements to obtain the metabolic stability component;

[0110] The duration of the concentration trough in the characteristic parameters is compared with a preset safe duration threshold to obtain the lack of risk component.

[0111] The in vivo vitamin metabolic status index is obtained by performing a weighted summation operation on the concentration satisfaction component, the metabolic stability component, and the deficiency risk component.

[0112] Obtain the preset recommended daily intake of vitamins;

[0113] The dietary adequacy index is obtained by comparing the estimated vitamin intake with the recommended daily intake of vitamins.

[0114] In this embodiment, preset baseline values ​​for vitamin physiological requirements are obtained. These baseline values ​​characterize the vitamin concentration levels required by the body to maintain normal physiological functions. For adults, the baseline value for vitamin D can be set at 30 ng / mL, the baseline value for vitamin C at 1.0 mg / dL, and the baseline value for vitamin B12 at 500 pg / mL. These baseline values ​​were determined through extensive clinical research data and reflect the vitamin levels required for the normal functioning of human tissues and organs.

[0115] The concentration satisfaction component is obtained by comparing the average concentration of the characteristic parameters with the baseline value of vitamin D physiological requirements. Specifically, for a user's vitamin D metabolism test, assuming the average concentration is 24 ng / mL and the baseline value of vitamin D physiological requirements is 30 ng / mL, the concentration satisfaction component is calculated as 24 ÷ 30 = 0.8. This component reflects the degree to which the average vitamin D concentration in the user's body meets the physiological requirements, and its value ranges from 0 to 2. The larger the value, the higher the degree of satisfaction, with 1.0 indicating that the requirement is just met.

[0116] The metabolic stability component is obtained by normalizing the concentration fluctuation range in the characteristic parameters against the baseline value of vitamin physiological requirements. For example, if a user's vitamin D concentration fluctuation range is 9 ng / mL and the baseline value of physiological requirements is 30 ng / mL, then the metabolic stability component is calculated as 1 - (9 ÷ 30) = 0.7. This component reflects the stability of the user's vitamin concentration fluctuation relative to the baseline value, and its value ranges from 0 to 1. The larger the value, the more stable the metabolism.

[0117] The duration of the vitamin D concentration trough in the characteristic parameters is compared with a preset safe duration threshold to obtain the deficiency risk component. The safe duration threshold for vitamin D can be set to 14 days, meaning that a concentration below 20% of the physiological requirement baseline for 14 consecutive days is considered a high-risk state. If the user's vitamin D concentration trough lasts for 5 days, the deficiency risk component is calculated as 1 - (5 ÷ 14) = 0.64. This component ranges from 0 to 1, with a higher value indicating a lower deficiency risk.

[0118] A weighted summation of the concentration satisfaction component, metabolic stability component, and deficiency risk component yields an indicator of vitamin metabolic status in the body. Pre-defined recommended daily intake (RDI) values ​​for vitamins are then obtained. Based on factors such as gender and age, the RDI for vitamin D can be set at 600 IU, for vitamin C at 75-90 mg, and for vitamin B12 at 2.4 μg. These RDI values ​​are typically established by a committee of nutrition experts, taking into account the general needs of the population.

[0119] Finally, the estimated vitamin intake is compared with the recommended daily intake (RDA) to obtain the dietary adequacy index. For example, based on food diaries and food composition database analysis, the estimated daily vitamin D intake is 400 IU, while the RDA is 600 IU. Therefore, the dietary adequacy index is calculated as 400 ÷ 600 = 0.67. This index ranges from 0 to 2; a higher value indicates a more adequate dietary intake, with 1.0 indicating that the recommended level has been met.

[0120] In practical applications, the weights of each parameter can be adjusted according to different user characteristics. For example, for pregnant women, the weight of the vitamin folic acid concentration satisfaction component can be increased to 0.7 to reflect its importance; for the elderly, the weight of the vitamin D deficiency risk component can be increased to 0.35 to reflect the high risk of vitamin D deficiency in the elderly.

[0121] The system can provide corresponding suggestions based on the calculation results. For example, if the vitamin D metabolic status index is 0.738 and the dietary supply adequacy index is 0.67, both of which are lower than the ideal value of 1.0, the system can suggest that the user increase outdoor activity time to promote vitamin D synthesis, supplement with vitamin D preparations as appropriate, and increase the intake of foods rich in vitamin D, such as fish oil and egg yolks.

[0122] Additionally, a metabolic stability component of 0.7 indicates significant fluctuations in the user's vitamin D metabolism, potentially related to irregular lifestyle habits. The system can advise the user to maintain a regular sleep schedule, a balanced diet, and, if necessary, use slow-release vitamin supplements to stabilize blood concentrations. The deficiency risk component is 0.64, indicating a moderate risk level. The system can advise the user to avoid consuming foods that are deficient in vitamin D for several consecutive days, especially during winter or prolonged indoor work periods.

[0123] The above methods can be used to scientifically assess a user's vitamin metabolism and dietary vitamin supply, providing a basis for personalized nutritional intervention, thereby improving the user's health and reducing the risk of vitamin deficiency or excess.

[0124] This invention provides an in vivo dynamic monitoring system for water-soluble vitamins based on multimodal biosensing, comprising:

[0125] The first unit is used to continuously collect in vivo interstitial fluid or transdermal signals as raw signals through biosensors in a wearable sensing terminal, and to filter and perform analog-to-digital conversion processing on the raw signals to generate digital concentration signals;

[0126] The second unit is used to transmit the digitized concentration signal to the user terminal via wireless communication.

[0127] The third unit is used to receive external calibration data at the user terminal, establish an individualized mapping relationship between the digitized concentration signal and the blood vitamin concentration, calibrate and convert the digitized concentration signal according to the individualized mapping relationship, and generate calibrated vitamin concentration time-series data.

[0128] The fourth unit is used to obtain dietary intake information. By accessing the food composition database of the cloud service platform, it calculates an estimated value of vitamin intake based on the dietary intake information.

[0129] The fifth unit is used to record the time of vitamin preparation administration as the start time of the test after receiving the urine load test start command, and to continuously collect the calibrated vitamin concentration time series data within the preset test time period;

[0130] The sixth unit is used to receive the total urine volume input by the user after the preset test time period ends, perform time integration on the calibrated vitamin concentration time-series data within the preset test time period, calculate the total vitamin excretion within the time period in combination with the total urine volume, and generate a urine load test report;

[0131] The seventh unit is used to transmit the calibrated vitamin concentration time-series data and the estimated vitamin intake to the cloud service platform, and to perform a fusion analysis on the calibrated vitamin concentration time-series data and the estimated vitamin intake based on the nutritional status assessment model to generate a comprehensive nutritional status assessment result.

[0132] A third aspect of the present invention provides an electronic device, comprising:

[0133] processor;

[0134] Memory used to store processor-executable instructions;

[0135] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0136] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0137] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0138] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for in vivo dynamic monitoring of water-soluble vitamins based on multi-modal biosensing, characterized in that, The method comprises the following steps: The biological sensor in the wearable sensing terminal continuously collects interstitial fluid or transdermal signals in the body as raw signals, filters and digitizes the raw signals to generate digital concentration signals, and transmits the digital concentration signals to the user terminal through wireless communication; The user terminal receives external calibration data, establishes an individualized mapping relationship between the digital concentration signals and blood vitamin concentration, and converts the digital concentration signals according to the individualized mapping relationship to generate calibrated vitamin concentration time series data; Dietary intake information is obtained, the food composition database of the cloud service platform is accessed, the vitamin intake estimate is calculated according to the dietary intake information, after receiving the urine load test start instruction, the vitamin preparation taking time is recorded as the test starting time point, and the calibrated vitamin concentration time series data is continuously collected within the preset test period; After the end of the preset test period, the total volume of urine input by the user is received, the calibrated vitamin concentration time series data within the preset test period is subjected to time integration operation, the total amount of vitamin excretion within the period is calculated in combination with the total volume of urine, and a urine load test report is generated; the calibrated vitamin concentration time series data and the vitamin intake estimate are transmitted to the cloud service platform, and the calibrated vitamin concentration time series data and the vitamin intake estimate are fused and analyzed based on a nutritional status evaluation model to generate a comprehensive nutritional status evaluation result.

2. The method of claim 1, wherein, The biological sensor comprises at least two sensing units for different water-soluble vitamins, each sensing unit adopts electrochemical sensing technology or optical sensing technology to generate raw signals corresponding to different water-soluble vitamins.

3. The method of claim 1, wherein, The user terminal receives external calibration data, establishes an individualized mapping relationship between the digital concentration signals and blood vitamin concentration, and converts the digital concentration signals according to the individualized mapping relationship to generate calibrated vitamin concentration time series data, which comprises: The user terminal receives user input external calibration data, which includes blood vitamin concentration measured values obtained by in-vitro detection and sampling time corresponding to the blood vitamin concentration measured values; According to the sampling time, the sensor output signal value at the corresponding time is extracted from the digital concentration signal, and a calibration data pair between the sensor output signal value and the blood vitamin concentration measured value is established; Based on the calibration data pair, a regression function is constructed, which represents the individualized mapping relationship between the digital concentration signal and the blood vitamin concentration; The digital concentration signal is input into the regression function for conversion operation, and a blood vitamin concentration estimate value sequence corresponding to the digital concentration signal in time sequence is output as the calibrated vitamin concentration time series data.

4. The method of claim 1, wherein, The calibrated vitamin concentration time series data within the preset test period is subjected to time integration operation, and the total amount of vitamin excretion within the period is calculated in combination with the total volume of urine, which comprises: extracting all data points in the preset test time period from the calibrated vitamin concentration time series data, the all data points including multiple sampling time points and corresponding vitamin concentration values; taking a time interval between adjacent sampling time points as an integral step, taking a corresponding vitamin concentration value as a concentration representative value in the time interval, calculating a concentration and time product in each time interval, and obtaining multiple time interval concentration product values; performing accumulation sum operation on the multiple time interval concentration product values, and obtaining a vitamin concentration time integral value in the preset test time period; performing product operation on the vitamin concentration time integral value and the total urine volume, and obtaining a total vitamin excretion amount in the time period.

5. The method of claim 1, wherein, based on a nutritional status evaluation model, fusing and analyzing the calibrated vitamin concentration time series data and the vitamin intake estimate value, and generating a comprehensive nutritional status evaluation result, including: extracting feature parameters representing dynamic changes of vitamins in the body from the calibrated vitamin concentration time series data, the feature parameters including a concentration mean value, a concentration fluctuation amplitude, and a concentration trough duration; inputting the feature parameters and the vitamin intake estimate value as input variables into the nutritional status evaluation model; the nutritional status evaluation model calculates a vitamin metabolism state index in the body based on the feature parameters, the vitamin metabolism state index representing utilization efficiency and storage capacity of the body for vitamins; the nutritional status evaluation model calculates a dietary supply adequacy index based on the vitamin intake estimate value, the dietary supply adequacy index representing a satisfaction degree of external intake to body demand; the nutritional status evaluation model performs weighted comprehensive operation on the vitamin metabolism state index in the body and the dietary supply adequacy index, and generates a comprehensive nutritional status evaluation result, the comprehensive nutritional status evaluation result representing an overall level of vitamin nutrition of the body.

6. The method of claim 5, wherein, the method further includes: obtaining a preset vitamin physiological requirement reference value, the vitamin physiological requirement reference value representing a vitamin concentration level required by the body to maintain normal physiological functions; performing ratio operation on the concentration mean value in the feature parameters and the vitamin physiological requirement reference value, and obtaining a concentration satisfaction component; performing normalization operation on the concentration fluctuation amplitude in the feature parameters and the vitamin physiological requirement reference value, and obtaining a metabolic stability component; performing comparison operation on the concentration trough duration in the feature parameters and a preset safety time threshold, and obtaining a deficiency risk component; performing weighted sum operation on the concentration satisfaction component, the metabolic stability component, and the deficiency risk component, and obtaining the vitamin metabolism state index in the body; obtaining a preset vitamin daily recommended intake standard value; performing ratio operation on the vitamin intake estimate value and the vitamin daily recommended intake standard value, and obtaining the dietary supply adequacy index.

7. An in-vivo water-soluble vitamin dynamic monitoring system based on multi-modal biosensing for implementing the method of any one of claims 1-6, characterized in that, including: a first unit configured to continuously collect interstitial fluid or transdermal signals in the body as raw signals through a biological sensor in a wearable sensing terminal, perform filtering and analog-to-digital conversion processing on the raw signals, and generate digitized concentration signals; The second unit is configured to transmit the digitized concentration signal to a user terminal via wireless communication. The third unit is configured to receive external calibration data at the user terminal, establish an individualized mapping relationship between the digitized concentration signal and the blood vitamin concentration, perform calibration conversion on the digitized concentration signal according to the individualized mapping relationship, and generate calibrated vitamin concentration time series data. The fourth unit is configured to obtain dietary intake information, access a food composition database of a cloud service platform, and calculate a vitamin intake estimate value according to the dietary intake information. The fifth unit is configured to record a vitamin preparation taking time as a test starting time point after receiving a urine load test starting instruction, and continuously collect the calibrated vitamin concentration time series data within a preset test time period. The sixth unit is configured to receive a total urine volume input by a user after the end of the preset test time period, perform time integration operation on the calibrated vitamin concentration time series data within the preset test time period, calculate a total vitamin excretion amount within the time period in combination with the total urine volume, and generate a urine load test report. The seventh unit is configured to transmit the calibrated vitamin concentration time series data and the vitamin intake estimate value to the cloud service platform, perform fusion analysis on the calibrated vitamin concentration time series data and the vitamin intake estimate value based on a nutritional status evaluation model, and generate a comprehensive nutritional status evaluation result.

8. An electronic device, comprising: The computer program instructions are executed by the processor to implement the method of any one of claims 1-6. The computer program instructions are executed by the processor to implement the method of any one of claims 1-6. ​ ​ 9. A computer-readable storage medium having stored thereon computer program instructions, wherein, ​

Images