A method for non-invasive monitoring of blood glucose indicators in a home diabetic

Abstract

The application discloses a blood glucose index non-invasive monitoring method for home diabetic patients, and aims at the problem that the existing home urine sugar detection is affected by the dynamic dilution of the toilet bottom water and the change of individual physiological state, resulting in large calculation error. The method first acquires a toilet residual water constant C, and matches a specific blood sugar urine sugar ratio constant K2 based on the current physiological state of the user. Then, a sudden change signal such as temperature or conductivity is collected by an immersion urine sensor, the urination duration is calculated to dynamically estimate the single urine volume M, and the mixed urine sugar concentration CDG is acquired. Finally, the dilution ratio constant K1 is obtained based on C and M, and multi-dimensional dynamic compensation is carried out in combination with K2 and CDG, so that the current blood glucose index BG can be accurately calculated. Through dynamic estimation of the underlying signal and matrix matching of the physiological state, the application effectively eliminates the interference of the environment and metabolic fluctuations, and realizes painless and high-precision invisible monitoring.

Description

Technical Field

[0001] This invention relates to the field of medical auxiliary detection technology, and in particular to a non-invasive method for monitoring blood glucose levels in diabetic patients at home. Background Technology

[0002] Current blood glucose testing technologies are mainly divided into two categories: invasive and non-invasive. Invasive testing typically involves collecting blood samples using a needle for analysis. While highly accurate, it is extremely inconvenient for people who are afraid of pain or prone to fainting at the sight of blood. Furthermore, wounds heal poorly in diabetic patients, and frequent needle pricks can easily lead to infection and skin damage. Non-invasive testing currently mostly uses urine glucose test strips. Although this avoids the pain of sampling, it can only roughly determine the blood glucose range, lacking specific numerical values, and cannot be systematically compared and analyzed with historical data.

[0003] When using urine for blood glucose monitoring in a home environment, existing technologies face two core challenges: At home, urination usually occurs in the toilet, where urine is significantly diluted by the residual water at the bottom. Because the amount of residual water varies between toilets, and the amount of urine urinated each time is random, the dilution ratio is constantly changing. Simply using a fixed constant for estimation will result in a large deviation in the final measured blood glucose level.

[0004] The proportion of glucose in the blood that is filtered into urine by the kidneys (the blood glucose-to-urine glucose ratio) is not a constant value, but fluctuates dramatically with changes in physiological states such as before and after meals and before sleep. Existing technologies often overlook the interference of these physiological states on computational models, resulting in low reliability of test results and making them unsuitable for long-term monitoring. Therefore, how to achieve high-precision, dynamically compensated, non-invasive monitoring of blood glucose levels in the highly uncertain environment of a home toilet is a pressing technical problem that needs to be solved in this field. Summary of the Invention

[0005] The purpose of this invention is to provide a non-invasive method for monitoring blood glucose levels in diabetic patients at home, aiming to solve the technical problems of inaccurate blood glucose estimation caused by uncertain dilution rates of toilet water, fluctuations in single urination volume, and changes in the patient's physiological state in existing home urine glucose testing.

[0006] To address the aforementioned technical problems, this invention provides a non-invasive method for home-based monitoring of blood glucose levels in diabetic patients, comprising the following steps: Obtain the residual water volume constant C of the household toilet; The system obtains the user's current physiological state command and, based on a preset physiological state mapping relationship, determines the blood glucose-to-urine glucose ratio constant K2 corresponding to the physiological state command. The urine signal in the toilet is collected by an immersion urine sensor. Based on the urine signal, the estimated urine volume M of this urination is determined, and the concentration of glucose CDG in the mixed toilet urine is obtained. Based on the residual water volume constant C and the estimated urine volume M, the dilution ratio constant K1 is calculated, where K1 = (M + C) / M; Based on the dilution ratio constant K1, the blood glucose-to-urine glucose ratio constant K2, and the urine glucose concentration in the toilet (CDG), the user's current blood glucose level (BG) is calculated, where BG = K1 × K2 × CDG.

[0007] Further, determining the estimated urine volume M for this urination based on the urine signal includes: Monitor the abrupt changes in the physical property signals collected by the immersion urine sensor; the physical property signals include temperature signals or conductivity signals; Calculate the duration of urination based on the start and end times of the mutation characteristics; By combining a preset user vital sign model with the urination duration, the estimated urine volume M is dynamically estimated.

[0008] Furthermore, the step of determining the estimated urine volume M for this urination based on the urine signal also includes a backup mapping step: The mobile application provides an interactive interface with categorization options. Receive active selection commands triggered by the user on the interactive interface; The active selection instruction is mapped to the corresponding estimated urine volume value M.

[0009] Furthermore, the preset physiological state mapping relationship is represented as a two-dimensional matrix of physiological states and ratio constants; The physiological state instructions include any one of the following: before meals, after meals, before sleep, or after sleep; If no physiological state instruction is received from the user, the system will default to processing in the post-meal stage and call the corresponding blood glucose-to-urine glucose ratio constant K2.

[0010] Further, the immersion urine sensor includes an electrode-type immersion urine sensor or a photoelectric-type immersion urine sensor; the acquisition of the mixed toilet urine glucose concentration (CDG) includes: When using an electrode-type immersion urine sensor, an electric field is established between the positive and negative electrodes of the sensor to cause an electrochemical reaction of glucose in the urine. The electrical signal is detected and converted into a voltage or current signal, which is then further converted into a digital signal to obtain the concentration of glucose (CDG) in the toilet urine. When a photoelectric type immersion urine sensor is used, the light wave transmission module is activated to emit light waves of a specific wavelength that penetrate the urine. After being attenuated by the urine, the light waves are received by the light wave receiving module, which converts them into voltage or current signals, and further converts them into digital signals to obtain the concentration of glucose (CDG) in the toilet urine.

[0011] Furthermore, obtaining the residual water volume constant C of the home toilet includes: During the system initialization phase, the application provides a toilet specification matching interface or a volume parameter input interface. The system receives toilet parameters input by the user and maps the residual water constant C of the current home to the local database accordingly.

[0012] Furthermore, after calculating the user's current blood glucose level (BG), the calculation also includes: Determine whether the blood glucose index BG is within a preset valid value range; if invalid, output a warning message. When the blood glucose index BG is valid, the blood glucose index BG is transmitted to the mobile terminal application via wireless communication. The application displays the blood glucose index BG graphically and retrieves past records for comparison and analysis; if the blood glucose index BG exceeds the upper limit, it is determined to be hyperglycemia, and if it is below the lower limit, it is determined to be hypoglycemia.

[0013] A non-invasive blood glucose monitoring system, comprising: An immersion urine sensor is used to collect urine signals from inside the toilet. A non-invasive blood glucose meter, connected to the immersion urine sensor, is used to perform the calculation and processing steps as described in the monitoring method to obtain the current blood glucose level BG; A mobile terminal is wirelessly connected to the non-invasive blood glucose meter to receive and display the blood glucose index BG.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: By capturing abrupt changes in temperature or conductivity in real time using an immersion urine sensor to accurately define the duration of urination, and combining this with a user's vital signs model to dynamically estimate the volume M of a single urine stream, the precise dilution ratio K1 for a single urination is derived together with the pre-calibrated residual water volume C in the toilet. This completely eliminates the distortion of underlying signals caused by "differences in bottom water volume" and "fluctuations in single urine volume" in the home toilet environment, significantly improving the baseline accuracy of non-invasive blood glucose estimation.

[0015] By introducing a two-dimensional physiological state matrix, the constant K2 of the blood glucose-to-urine glucose ratio is accurately mapped to different specific metabolic stages such as "before meals, after meals, and before bedtime". This effectively compensates for the algorithm offset caused by the drastic fluctuations in the glucose metabolism rate of the human body under different work and rest and dietary conditions. It upgrades the conventional static linear estimation to a nonlinear restoration model that conforms to the dynamic laws of individual physiology, giving the test results extremely high medical reference value.

[0016] By setting a manual urine volume mapping mechanism for the mobile terminal interactive interface in parallel in the control method, and a minimum threshold cleaning judgment step for the underlying urine glucose concentration CDG, the monitoring continuity of the sensor is ensured when the power consumption is low or some detection modules fail. At the same time, the cross-interference of chemical substances such as detergent residue in the toilet is successfully filtered out, giving the system extremely high operating condition robustness.

[0017] By using in-situ detection of the toilet's water area and linking it with the data visualization processing of a smart terminal, the physiological trauma and hemophobia associated with traditional needle blood collection are completely eliminated. At the same time, relying on the automatic saving and periodic comparison mechanism, an intuitive long-term health map is constructed for users, achieving truly painless, high-frequency, and long-term home-based invisible monitoring. Attached Figure Description

[0018] 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.

[0019] Figure 1 This is an architectural block diagram of a non-invasive blood glucose monitoring system provided in an embodiment of the present invention; Figure 2 This is a main flowchart of a non-invasive home-based blood glucose monitoring method for diabetic patients provided by an embodiment of the present invention; Figure 3 This is a flowchart of the dynamic estimation and interactive mapping of urine volume provided in an embodiment of the present invention; Figure 4 This is a block diagram illustrating the working principle of an analog signal-based non-invasive blood glucose meter provided in an embodiment of the present invention. Figure 5 This is a block diagram illustrating the working principle of a digital signal-based non-invasive blood glucose meter provided in an embodiment of the present invention. Figure 6 This is a block diagram illustrating the working principle of an electrode-type immersion urine sensor provided in an embodiment of the present invention. Figure 7 This is a block diagram illustrating the working principle of an immersion urine sensor of photoelectric type provided in an embodiment of the present invention.

[0020] In the diagram: S1 is the physical signal acquisition probe, S2 is the biochemical signal acquisition probe, U1 is the algorithm processing module, U2 is the preprocessing module, U3 is the wireless transmission module, TP1 is the command input module, and LCD1 is the data display module. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this invention clearer, the following detailed description of the non-invasive home blood glucose monitoring method for diabetic patients, in conjunction with the accompanying drawings and specific embodiments, provides further insight. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of the invention.

[0022] like Figure 1-7 As shown, the non-invasive home blood glucose monitoring method for diabetic patients provided by this invention is based on a dynamic measurement model that integrates the physiological mechanism of urination with the water environment of the toilet. This method is typically executed by a non-invasive blood glucose monitoring system, which mainly includes an immersion urine sensor placed inside the toilet, a non-invasive blood glucose meter (containing a pre-processing module and an algorithm processing module) located outside the toilet, and a smart mobile terminal (such as a smartphone) that wirelessly connects to the meter.

[0023] When performing a complete non-invasive blood glucose monitoring procedure, the system first needs to obtain the residual water volume constant C of the user's home toilet to eliminate initial volume errors caused by different household bathroom fixtures. During the system initialization phase, when the user first opens the dedicated application (APP) on their smartphone, a toilet specification matching interface will pop up. The user can search and bind their toilet model (e.g., selecting "Brand X1"), or manually enter the bottom water volume specified in the toilet's instruction manual directly in the volume parameter input interface. Simultaneously, the user needs to enter their baseline physiological parameters such as gender, age, and weight in the application for the system to build and store in its local vital sign model. For example, if the user enters a residual water volume of 1500ml, the system will use this value "1500" as the residual water volume constant C, store it in the local database, and establish a long-term mapping with the user's account.

[0024] To ensure reliable physical execution of the method of this invention, this embodiment provides a detailed design for the signal flow of the underlying hardware. The specific hardware working principle is as follows: Figures 4 to 7 As shown: As a first system hardware architecture embodiment of the present invention (in combination with) Figure 4(See the analog signal block diagram shown). When sensor S1 is inserted into the non-invasive blood glucose meter, the meter powers S1 and starts its operation. Sensor S1 outputs an analog signal of voltage or current to the pre-processing module U2 inside the meter. After receiving the analog signal, the pre-processing module U2 parses and performs analog-to-digital conversion (A / D). The parsed result is transmitted as a digital signal to the algorithm processing module U1 via data interfaces such as UART, I2C, SPI, or USB. The algorithm processing module U1 determines the digital signal and converts it into a specific concentration value. The user can input option commands through the command input module TP1. If module U1 determines that the value is valid and within a specific range, the value is displayed on the data display module LCD1 in the form of numbers or icons; if the value is invalid (e.g., due to severe interference), a warning message is displayed on LCD1 to inform the user. At the same time, if the user selects wireless transmission through TP1, the wireless transmission module U3 is activated to send the data to the mobile APP.

[0025] As a second system hardware architecture embodiment of the present invention (in combination with) Figure 5 The block diagram of digital signal classes shown below, and its relationship with Figure 4 The difference lies in the fact that the sensor S1 has a built-in conversion circuit, and after acquiring data, it directly outputs digital signals to the algorithm processing module U1 of the measuring instrument through data interfaces such as UART, I2C, SPI or USB. After receiving the signal, U1 directly performs analysis, judgment and numerical conversion. The subsequent local display (LCD1), warning and wireless transmission (U3) logic is consistent with the analog signal architecture.

[0026] Regarding the aforementioned sensor S1, this embodiment preferably employs an immersion probe based on two principles: Firstly, such as Figure 6 The electrode type shown is an immersion urine sensor. Its working principle is as follows: After receiving the command from the measuring instrument, the sensor processing module U1 applies a rated voltage to the positive electrode P1 and the negative electrode P2, establishing a stable micro-electric field between P1 and P2. This induces a specific electrochemical reaction in the urine's glucose within the electric field range. Once the electric field stabilizes, U1 activates the thermocouple U2 to detect the glucose concentration, and the result is determined according to the underlying microscopic formula V=K. V ×BG (where BG is the current urine glucose concentration, V is the voltage, and K is the voltage) V (where I is a constant) or I=K I ×BG (I is the current, K) I (The concentration is constant), converting it into an analog signal such as voltage or current. This analog signal can be converted into a digital signal by the U1's built-in pre-processing module, or directly transmitted to the measuring instrument for processing.

[0027] Secondly, such as Figure 7The photoelectric immersion urine sensor shown is as follows: After receiving a command, the sensor processing module U1 activates the light transmission module LED1, emitting light waves of a specific wavelength that penetrate a certain distance into the urine. After being attenuated / reflected by the glucose in the urine, the light waves are received by the light reflection (or receiving) module LED2. After receiving the light waves, LED2 also calculates the value according to the formula V=K. V ×BG or I=K I ×BG converts the attenuation characteristics into voltage or current signals, and then outputs analog or digital signals to the measuring instrument.

[0028] After calibrating the environmental parameters, the process enters the physiological state acquisition phase when the user prepares for urination testing. Clinical studies have shown that the level of glucose metabolism and the renal glucose threshold differ significantly in different eating and sleeping stages. Therefore, the system needs to obtain the user's current physiological state command. The user selects their current stage, such as "2 hours after a meal," through the main interface of the mobile app. The system pre-loads or stores a "two-dimensional matrix of physiological state-metabolic ratio," which is based on baseline data derived from a large number of clinical samples. The horizontal axis represents the physiological stage (e.g., before meal, after meal, before sleep, after sleep), and the vertical axis represents the user's historical disease rating (e.g., mild, moderate). When the system receives the "after meal" command, the algorithm processing module retrieves and calls the blood glucose-urine glucose ratio constant K2 specifically for the postprandial state from this matrix. For example, the system extracts a postprandial ratio constant K2 of 1.2. If a user forgets to make any selections in the app, the system will activate an error prevention mechanism, which will default to obtaining the current phone system time and forcibly perform a backup process based on the "after-meal period" according to the user's normal schedule, to ensure that the calculation is not interrupted.

[0029] Subsequently, the process moves to the core stage of dynamic volume acquisition and concentration detection. The user places the immersion urine sensor in the residual water in the toilet bowl and begins urinating. At this time, the sensor acquisition end begins to capture the changes in the physical property signals of the urine mixture in real time. Specifically, this embodiment preferably uses the abrupt change characteristics of the temperature signal to estimate the urine volume. The temperature of the water at the bottom of a normal household toilet is usually close to the room temperature (e.g., 20°C), while the temperature of urine excreted by the human body is stable at around 37°C. When the urine comes into contact with the sensor, the temperature probe will capture a rapidly rising temperature step signal. The algorithm processing module sets a sudden change threshold (e.g., the temperature change rate exceeds 5°C / second and the absolute temperature reaches above 32°C). Once this threshold is met, the system immediately triggers the internal high-precision timer and records it as the start time of urination. As urination ends, the temperature of the mixed liquid in the toilet tends to level off or even begins to drop slightly due to heat exchange. The system captures this decay characteristic and stops timing, recording it as the end time of urination.

[0030] The system then calculates the time difference between the start and end times, i.e., the urination duration T (e.g., recorded as 15 seconds). Subsequently, the algorithm processing module combines a preset user vital sign model (which integrates baseline parameters such as the user's gender, age, and weight to deduce an individual's average urine flow rate baseline value, e.g., a baseline urine flow rate of 20 ml / s for males) for product or integral calculations. Assuming the system calculates the user's dynamic urine volume estimate M based on the vital sign model as: 15 seconds × 20 ml / s = 300 ml. As an important alternative embodiment of the present invention, if the dynamic temperature measurement function is disabled in certain low-power application scenarios, the system will display an interactive option for "Estimated Urine Volume" on the mobile APP interface, allowing the user to actively select "High (preset 400 ml), Medium (preset 250 ml), Low (preset 100 ml)" based on their own feeling. The system uses this manually mapped value as the urine volume estimate M.

[0031] After obtaining the precise dynamic urine volume M (such as 300ml as mentioned above) and the preset bottom water volume C (1500ml), the system immediately uses the dilution model algorithm to calculate the dilution ratio constant K1 formed by this urination in the toilet. The calculation is performed using the formula K1=(M+C) / M, i.e., K1=(300+1500) / 300=6. This indicates that the user's original urine was diluted 6 times in the toilet.

[0032] Simultaneously with the volumetric calculations, the underlying sensor performs biochemical concentration extraction. Upon receiving instructions from the measuring instrument, the immersion urine sensor (using electrode type as an example) applies a rated operating voltage between its positive and negative electrodes, establishing a stable micro-electric field. The diluted and mixed urine in the toilet undergoes a specific electrochemical reaction within this electric field. The internal thermocouple of the sensor detects the weak electrical signal generated in real time and converts it into an analog voltage signal according to a built-in linear conversion formula. This analog-to-digital conversion (A / D) by the pre-processing module converts it into a digital signal. This digital signal represents the concentration of mixed urine glucose (CDG) in the toilet (e.g., a detected concentration of 1.5 mmol / L). Furthermore, to eliminate interference from foreign matter such as detergents, the system performs a cleaning judgment at this point. If the CDG is below the extremely low threshold of 0.1 mmol / L, the system determines the signal is invalid and sends a "water quality abnormality interference" warning to the app, terminating subsequent calculations.

[0033] Finally, after all the parameters of the three key dimensions—K1 (6), the blood glucose-to-urine glucose ratio constant K2 (1.2), and CDG (1.5 mmol / L)—are collected in the algorithm processing module, the system executes the final multiplication operation: current blood glucose index BG = K1 × K2 × CDG. By substituting the above specific values, the system obtains the estimated value of the user's current true blood glucose index as: BG = 6 × 1.2 × 1.5 = 10.8 mmol / L.

[0034] After obtaining the final blood glucose level (BG), the system sends the data packet to the mobile terminal via wireless communication modules such as Bluetooth or WiFi. Upon receiving the data, the mobile app uses a graphics rendering engine to visually present the value of "10.8 mmol / L" as a graph with scale lines on the screen. The system compares this value to a historical database. If the preset safe upper limit for postprandial blood glucose is 11.1 mmol / L, although the data is within the safe range, it is close to the upper limit, so the app will display a yellow icon and pop up a health warning message: "Value is high; please pay attention to your diet and exercise moderately." Through this coherent, rigorous, and multi-dimensional dynamic compensation calculation process, this invention completely eliminates the painful traditional method of needle-based blood sampling. Relying on highly customized environmental and physiological compensation algorithms, it achieves extremely high-precision, non-invasive, real-time blood glucose monitoring at home.

[0035] To further illustrate the working process of the multidimensional dynamic compensation mechanism of this invention in a real home environment, two typical application scenarios are listed below: Example Scenario 1: Fully automated high-precision monitoring in the morning on an empty stomach (master control mode) Scenario preset: User A (male, mild diabetes patient), whose APP account has been initialized and bound to a home toilet residual water volume constant C of 1500ml.

[0036] Operation process: After waking up in the morning, user A clicks on the "After Sleep / Fasting in the Morning" status on the mobile APP. At this time, the system retrieves the blood glucose-urine glucose ratio constant K2 (assuming the system setting value is 1.0) corresponding to the fasting state from the physiological state-metabolic ratio two-dimensional matrix.

[0037] Dynamic calculation: When user A urinates, the immersion urine sensor quickly captures a constant temperature step signal of approximately 37°C, lasting for 20 seconds. The system, combined with the user's vital signs model (preset baseline urine flow rate of 20 ml / s), automatically estimates the urine volume M to be 400 ml.

[0038] System processing: The system immediately calculates the dilution ratio K1 = (400 + 1500) / 400 = 4.75. Simultaneously, the sensor detects a mixed urinary glucose concentration (CDG) of 1.2 mmol / L in the toilet bowl water. The system finally calculates BG = 4.75 × 1.0 × 1.2 = 5.7 mmol / L.

[0039] Results feedback: 5.7 mmol / L is within the normal fasting blood glucose range. The mobile app displays this value as a green curve and indicates "Current fasting blood glucose is well controlled".

[0040] Example Scenario 2: Degraded backup monitoring (standby mode) in situations where sensors are partially limited and the user is in a post-meal state. Scenario preset: User B (female, moderate diabetes patient), due to low sensor battery or temperature detection module not being turned on, the system automatically switches to backup interaction mode; the residual water volume of the toilet C bound to her is also 1500ml.

[0041] Operating procedure: Two hours after dinner, user B prepares to urinate, clicks "Post-meal stage" on the APP, and manually selects the "low urine volume" option in the pop-up estimated urine volume assistance interface based on their own feeling.

[0042] Dynamic calculation: The system calls the exclusive ratio constant K2 (assuming the set value is 1.2) based on the "post-meal stage"; at the same time, it maps the "low urine output" command to the preset volume estimate value M (assuming it is 150ml).

[0043] System processing: The system calculates the dilution ratio K1 at this point as K1 = (150 + 1500) / 150 = 11 (the dilution factor has increased significantly). Subsequently, the sensor detects a urine glucose concentration (CDG) of 0.8 mmol / L in the toilet bowl. The system ultimately calculates BG = 11 × 1.2 × 0.8 = 10.56 mmol / L.

[0044] Results feedback: Although the detected original solution concentration of 0.8 is not high, the system underwent significant dilution compensation and post-meal status compensation, restoring the true blood glucose prediction to 10.56 mmol / L. Because this value is close to the upper limit threshold after a meal, the APP interface displays an orange warning and prompts "Post-meal blood glucose is high, it is recommended to walk for 30 minutes."

[0045] Based on the aforementioned multidimensional dynamic compensation calculations, to further eliminate the uneven diffusion of urine in a home toilet and the underlying errors caused by the oxidation of glucose in urine by air and biological processes, the non-invasive blood glucose monitoring system of this invention also incorporates an independent "blood glucose verification method." This method can exist independently of conventional large-sample statistical models. Its prerequisite is that the diabetic patient has undergone blood glucose testing at a reputable medical institution within a short period and obtained an authoritative and reliable baseline blood glucose concentration value. .

[0046] The specific verification process is as follows: Before using the non-invasive blood glucose meter, users should select "Blood Glucose Verification Mode" through the instrument's interactive menu or mobile app, and manually enter their personal blood glucose concentration baseline value obtained from a recent hospital test. Due to varying user waiting times, the system offers two branching modes: "low-precision verification" and "high-precision verification." Firstly, the low-precision rapid calibration mode: After urination, the user waits for the urine to initially diffuse in the toilet. The sensor collects the real-time urine glucose concentration (CDG) in the toilet at the 5th and / or 10th minute. (5min) and CDG (10min) At this point, the system calculates the single-time verification constant K according to the verification formula. (5min) = / CDG (5min) and K (10min) = / CDG (10min) The system will take the arithmetic mean of the obtained single-test constants (i.e., K = (K...). (5min) +K (10min) ) / 2) is saved as the user's personalized verification constant.

[0047] Secondly, the high-precision antioxidant weighted verification mode: If the user selects high-precision verification, the system mandates a 30-minute verification process to fully capture diffusion characteristics. The non-invasive blood glucose meter triggers sampling verification at six specific times: 5, 10, 15, 20, 25, and 30 minutes. If no signal is collected within 3 minutes at any time, the system automatically determines that the verification has failed to prevent data distortion caused by excessive oxidation.

[0048] In this mode, considering that urine diffusion becomes more complete over time (and its weight should increase), but the interference from air oxidation and biological oxidation also increases, the system introduces a time gradient weighted algorithm: K = 5% × K (5min) +10%×K (10min) +15%×K (15min) +20%×K (20min) +25%×K (25min) +30%×K (30min)Since the sum of the weights mentioned above reaches 105% (5%+10%+15%+20%+25%+30%=105%), the algorithm engine introduces a proportionally scaled correction coefficient (100 / 105) in the final step to reduce the summation result, thereby accurately offsetting the error caused by oxidation loss. The final calculated high-precision calibration constant K will be updated to local storage for future daily monitoring to perform conversion corrections of underlying urine glucose concentrations that are highly tailored to individual physiological environments.

[0049] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A non-invasive method for home-based monitoring of blood glucose levels in diabetic patients, characterized in that, Includes the following steps: Obtain the residual water volume constant C of the household toilet; The system obtains the user's current physiological state command and, based on a preset physiological state mapping relationship, determines the blood glucose-to-urine glucose ratio constant K2 corresponding to the physiological state command. The urine signal in the toilet is collected by an immersion urine sensor. Based on the urine signal, the estimated urine volume M of this urination is determined, and the concentration of glucose CDG in the mixed toilet urine is obtained. Based on the residual water volume constant C and the estimated urine volume M, the dilution ratio constant K1 is calculated, where K1 = (M + C) / M; Based on the dilution ratio constant K1, the blood glucose-to-urine glucose ratio constant K2, and the urine glucose concentration in the toilet (CDG), the user's current blood glucose level (BG) is calculated, where BG = K1 × K2 × CDG.

2. The monitoring method according to claim 1, characterized in that, The step of determining the estimated urine volume M for this urination based on the urine signal includes: Monitor the abrupt changes in the physical property signals collected by the immersion urine sensor; the physical property signals include temperature signals or conductivity signals; Calculate the duration of urination based on the start and end times of the mutation characteristics; By combining a preset user vital sign model with the urination duration, the estimated urine volume M is dynamically estimated.

3. The monitoring method according to claim 1, characterized in that, The step of determining the estimated urine volume M based on the urine signal also includes a backup mapping step: The mobile application provides an interactive interface with categorization options. Receive active selection commands triggered by the user on the interactive interface; The active selection instruction is mapped to the corresponding estimated urine volume value M.

4. The monitoring method according to claim 1, characterized in that, The preset physiological state mapping relationship is represented as a two-dimensional matrix of physiological state and ratio constant; The physiological state instructions include any one of the following: before meals, after meals, before sleep, or after sleep; If no physiological state instruction is received from the user, the system will default to processing in the post-meal stage and will invoke the corresponding blood glucose-to-urine glucose ratio constant K2.

5. The monitoring method according to claim 1, characterized in that, The process of obtaining the mixed toilet urine glucose concentration (CDG) includes: An immersion urine sensor of electrode type detects the electrical signal generated by the electrochemical reaction in urine, and converts it to obtain the concentration of glucose (CDG) in the toilet urine; or, The toilet urine glucose concentration (CDG) is obtained by detecting the attenuation signal of light waves after penetrating urine using a photoelectric immersion urine sensor.

6. The monitoring method according to claim 1, characterized in that, The method for obtaining the residual water volume constant C of a home toilet includes: During the initialization process, a toilet specification matching interface or a volume parameter input interface is provided through a mobile terminal application. The system receives toilet parameters input by the user and maps the residual water constant C of the current home into a preset storage unit.

7. The monitoring method according to claim 1, characterized in that, The method also includes a blood glucose verification step based on authoritative reference values: Obtain the baseline blood glucose concentration value of the medical institution input by the user; After urination, the urine is left to stand for several preset diffusion time points, and the corresponding toilet urine glucose concentration is obtained using the immersion urine sensor; based on the ratio of the blood glucose concentration benchmark value to the toilet urine glucose concentration, the single-time verification constant corresponding to each diffusion time point is calculated. Based on multiple single-time verification constants, the final calibration constant K is calculated by averaging or weighted summation. When using weighted summation, the corresponding single-time verification constants are obtained at the 5th, 10th, 15th, 20th, 25th, and 30th minutes, and then summed with weights of 5%, 10%, 15%, 20%, 25%, and 30% respectively. The summation is then multiplied by a preset correction coefficient of 100 / 105 to obtain the final calibration constant K. If a timeout is detected at any corresponding moment, the verification is deemed to have failed.

8. The monitoring method according to claim 1, characterized in that, After calculating the user's current blood glucose level (BG), the following is also included: Determine whether the blood glucose index BG is within a preset valid value range; if invalid, output a warning message. When the blood glucose index BG is valid, the blood glucose index BG is transmitted to the mobile terminal application via wireless communication. The application displays the blood glucose index BG graphically and retrieves past records for comparison and analysis; if the blood glucose index BG exceeds the upper limit, it is determined to be hyperglycemia, and if it is below the lower limit, it is determined to be hypoglycemia.

9. A non-invasive blood glucose monitoring system, characterized in that, include: An immersion urine sensor is used to collect urine signals from inside the toilet. A non-invasive blood glucose meter, connected to the immersion urine sensor, is used to perform the calculation and processing steps in the monitoring method as described in any one of claims 1 to 8 to obtain the current blood glucose index BG; A mobile terminal is wirelessly connected to the non-invasive blood glucose meter to receive and display the blood glucose index BG.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the monitoring method as described in any one of claims 1 to 8.

Images