A wearable sweat detection system and its detection method

Abstract

This invention relates to a wearable sweat detection system and method, comprising a sweat sensor patch, a wearable sweat detection device, a software module, and a cloud platform. The wearable sweat detection system and method of this invention are lightweight and convenient, flexibly adaptable to different usage scenarios, and capable of continuous sweat collection and detection. Through a three-dimensional approach of data acquisition, analysis, and prediction, it achieves real-time monitoring and early warning of human health status using sweat. This invention incorporates a cloud platform to introduce a big data trend suggestion data model. The software module synchronizes data to the cloud platform via the network. The cloud platform builds a big data model based on trends to process the collected data and feeds back the processing results (trend prediction) to the software module for display. The software module provides health reminders and suggestions, as well as exercise recommendations, making predictions and exercise suggestions more reasonable and accurate.

Description

Technical Field

[0001] This invention relates to the field of sensor detection technology, specifically to a wearable sweat detection system and its detection method. Background Technology

[0002] Sweat is a bodily fluid secreted by sweat glands, containing a wealth of information related to health indicators. Analyzing the biochemical markers in sweat can provide a better understanding of one's health, such as exercise intensity, body water content, and muscle fatigue. Furthermore, sweat can indirectly reflect or predict potential diseases; for example, the salt concentration in sweat may be associated with cystic fibrosis symptoms, and Na+ is related to blood pressure and blood sugar regulation.

[0003] Online collection and testing of sweat can yield a series of dynamic biomarker concentration data that change over time, such as the timing of sweating, the amount of sweat, and the biochemical components in the sweat. Therefore, human sweat testing has become one of the hot topics in "personalized medicine," particularly in the field of body fluid diagnostics, in recent years.

[0004] However, current wearable sweat detection systems typically use multiple sensors stacked together to collect data, resulting in large size, high sensor costs, and insufficient flexibility in use. They cannot be easily adjusted according to the detection location and usage scenario, and do not form a closed loop of data collection, data analysis, and trend prediction. Summary of the Invention

[0005] The technical problem solved by this invention is to provide a lightweight and convenient wearable sweat detection system and its detection method, which can be flexibly configured according to the usage scenario, can continuously collect and detect sweat, and then realize real-time monitoring and early warning of human health status through a three-dimensional approach of data acquisition, analysis and prediction.

[0006] The technical solution adopted by the present invention to solve its technical problem is:

[0007] A wearable sweat detection system includes a sweat sensor patch, a wearable sweat detection device, a software module, and a cloud platform;

[0008] The sweat sensor patch is used to collect sweat secreted by the human skin.

[0009] The wearable sweat detection device includes a sweat signal acquisition and preprocessing module, a sweat signal analysis module, a data storage module, and a communication module;

[0010] The sweat signal acquisition and preprocessing module is used for acquiring and preprocessing sweat flow rate and sweat concentration signals. The module inputs a high-frequency oscillation signal into the sweat sensor patch to obtain the impedance of the electrodes on the patch, which include fixed electrodes and interdigitated electrodes. The high-frequency oscillation signal drives the interdigitated electrodes to generate an AC signal for sweat flow rate, and drives the fixed electrodes to generate an AC signal for sweat concentration. The sweat flow rate and concentration AC signals are preprocessed by the sweat signal acquisition and preprocessing module and converted into corresponding DC signal values.

[0011] The sweat signal analysis module is used to extract sweat flow rate, calculate sweat concentration, calculate sweat loss, calculate electrolyte loss, calculate dehydration level, and calculate hydration replenishment.

[0012] The communication module is used to communicate with the software module via wired or wireless communication, transmit the sweat data processed by the sweat signal analysis module to the software module, and visualize and display the sweat data through the software module.

[0013] The software module transmits the sweat data to a cloud platform via a network, performs big data analysis and builds a big data model on the cloud platform, and then feeds back the processed trend prediction results to the software module for display via the network.

[0014] Furthermore, the software module includes software integrated into the wearable sweat detection device, software installed on a computer, or software installed on a mobile device.

[0015] Furthermore, the wearable sweat detection device also includes a shell, a PCB motherboard, chips and electronic components required for each module, electrode contacts electrically connected to the sweat sensor patch, a power management module, and a battery.

[0016] A detection method for a wearable sweat detection system includes the following steps:

[0017] S1: Sweat signal acquisition and preprocessing;

[0018] S2: Sweat signal analysis and processing;

[0019] S3: Data storage and data transmission;

[0020] S4: Data Presentation;

[0021] S5: Cloud Platform Trend Forecast.

[0022] Furthermore, in step S1, the sweat signal acquisition and preprocessing also includes the following steps:

[0023] S11: Sweat flow rate signal acquisition;

[0024] S12: Sweat concentration signal acquisition;

[0025] S13: Preprocessing of converting sweat alternating current signal to sweat direct current signal.

[0026] Further, the sweat flow rate signal acquisition and the sweat concentration signal acquisition are switched and controlled through electrode sampling switching technology. The electrode sampling switching technology uses a pair of electrodes shared by the interdigitated electrodes and the fixed electrodes, and is alternately acquired by the switching control module in the wearable sweat detection device according to the principle of time-division multiplexing.

[0027] Further, the sweat flow rate signal acquisition adopts interdigitated electrode adaptive technology. The interdigitated electrode adaptive technology sets operational amplifier circuits with different magnification factors in the hardware of the wearable sweat detection device, and switches through the control software's analog switch chip in the wearable sweat detection device on the operational amplifier circuits with different magnification factors to achieve adaptive switching of the interdigitated electrode gears, so as to adapt to sweat with different flow rates. The method for selecting and switching gears includes:

[0028] When V(n)>VH(n) and V(n + 1)>VL(n + 1), switch to the n + 1 gear;

[0029] When V(n)<VL(n) and V(n - 1)>VL(n - 1), switch to the n - 1 gear;

[0030] Among them, n (n = 1, 2, 3,..., m) is the interdigitated electrode gear, VL(n) is the low voltage threshold corresponding to the n gear, VH(n) is the high voltage threshold corresponding to the n gear, and V(n) is the voltage value measured at the n gear.

[0031] Further, the conversion formula for converting the sweat alternating current signal to the sweat direct current signal is:

[0032]

[0033] Among them, x is the signal sampling point, n is the number of sampling points, X rms is the converted DC eigenvalue;

[0034] Or use the formula:

[0035]

[0036] Among them, Amp out is the amplitude of the high-frequency injected sine signal;

[0037] The amplitude of the input signal acquired in the Nth acquisition;

[0038] The amplitude ratio of the nth input to the output;

[0039] This represents the amplitude ratio after filtering and smoothing.

[0040] Furthermore, in step S2, the sweat signal analysis and processing also includes the following steps:

[0041] S21: Sweat flow rate extraction process;

[0042] S22: Calculation and processing of sweat concentration;

[0043] S23: Calculation and processing of sweat loss;

[0044] S24: Calculation and processing of electrolyte loss;

[0045] S25: Calculation and processing of water shortage degree;

[0046] S26: Water replenishment calculation and processing.

[0047] Furthermore, in step S21, the sweat flow rate extraction process employs step signal edge recognition technology, which includes the following steps:

[0048] S211: Preset a queue of length 2n, and set a recognition value of k;

[0049] S212: Clear the queue;

[0050] S213: Store the collected and preprocessed sweat data into the queue;

[0051] S214: When the queue is full, the average value of the queue data is calculated as m;

[0052] Record L, where L is the number of values ​​less than (mk) in the 0 to n part of the queue;

[0053] Record R, where R is the number of values ​​greater than (m+k) in the n to 2n part of the queue;

[0054] S215: Obtain the step edge signal and time: If L>2n*f and R>2n*f (f is the matching factor), then it is a step edge signal. Record the current time as Tn and jump to step S212. Otherwise, jump to step S213.

[0055] S216: Calculate the step signal time T, T = T n -T n-1T is the time of the edge signal of the nth step minus the time of the edge signal of the (n-1)th step.

[0056] S217: Calculate the sweat flow rate S, S = V / T, where V is the volume between the two interposed electrodes;

[0057] S218: Proceed to step S212;

[0058] Alternatively, the sweat flow rate can be obtained using the following steps:

[0059] S211a: Obtain the step edge signal.

[0060] Calculate Teen_err,

[0061] Teen_err=Teen_V n -Teen_V n-1 ;

[0062] Among them, Teen_V n The nth DC characteristic value corresponding to the conversion is injected into the high-frequency oscillation signal of the interdigital electrode; where n is an integer.

[0063] determination,

[0064] Teen_err>Teen_K;

[0065] Where Teen_K is a set value (preset value);

[0066] If Teen_err > Teen_K, then it means that in Teen_t n A step edge signal appeared at the current station, and the current station was recorded.

[0067] Teen_t, the time point corresponding to the edge signal. n ;

[0068] S212a: Calculate the sweat flow rate S.

[0069] S=VOL÷(Teen_t n -Teen_t n-1 );

[0070] Where VOL is the volume between the two interdigitated electrodes, and Teen_t n For Teen_V n The time point corresponding to the step edge signal, Teen_t n-1 This is the time point corresponding to the edge signal of the previous step.

[0071] Furthermore, the cloud platform trend prediction also includes the following steps:

[0072] S51: Calculate and obtain the estimated total body electrolyte level An.

[0073] An = f(Mc) 全身 M 汗液 );

[0074] Where An is the estimated total body electrolyte level, and n is a positive integer; Mc 全身 M represents the total electrolyte loss. 汗液 This represents the amount of sweat lost throughout the body.

[0075] S52: Calculate and obtain the electrolyte loss rate Bs corresponding to the s-type exercise.

[0076] Bs = ma(Vc) 全身 );

[0077] Where ma is the smoothing filter, Bs is the electrolyte loss rate corresponding to motion type s, and Vc 全身 This represents the rate of electrolyte loss throughout the body.

[0078] S53: Calculate and obtain Cn.

[0079] Cn = pettitt(Bn...B1);

[0080] Among them, Pettitt is the mutation detection algorithm, and Bn to B1 are historical data;

[0081] S54: Calculate and obtain the reference comparison deviation ratio Kn.

[0082] Kn = h(Cn);

[0083] Where h is the standard reference contrast function, and Kn is the reference contrast deviation ratio;

[0084] S55: Build data models and perform trend prediction.

[0085] When |Kn| < 10%, the electrolyte level trend is normal, indicating good health.

[0086] When 10% < |Kn| < 30%, the electrolyte level trend is slightly abnormal, and reasonable exercise and diet are recommended;

[0087] When 30% < |Kn| < 50%, the electrolyte level trend is moderately abnormal, and it is recommended to check if there is an underlying health problem; when 50% < |Kn|, the electrolyte level trend is severely abnormal, and it is recommended to check if there is an abnormal health condition.

[0088] S56: Exercise duration suggestion, calculate and obtain the suggested exercise duration Tn.

[0089] Tn = g(Bs, Kn);

[0090] Where g is the reasonable motion duration Tn for calculating motion k.

[0091] The beneficial effects of the present invention are:

[0092] 1. The wearable sweat detection system and method of the present invention are lightweight and convenient, and can be flexibly configured according to the usage scenario. They can continuously collect and detect sweat, and then realize real-time monitoring and early warning of human health status by means of data acquisition, analysis and prediction in a three-dimensional integrated manner.

[0093] 2. In this invention, the acquisition of sweat flow rate and sweat concentration signals is switched and controlled using electrode sampling switching control technology. Based on the signal variation characteristics, to reduce the number of connection points between the sweat detection device and the sweat sensor patch, the sweat concentration detection electrode (i.e., the fixed electrode in the sweat sensor patch) and the sweat flow rate detection electrode (i.e., the interdigitated electrode in the sweat sensor patch) share a common pair of electrodes. The switching control module detects these electrodes alternately according to a time-division multiplexing principle. This ensures both signal accuracy and device structural simplicity. The time-division detection mechanism for sweat concentration and sweat flow rate greatly simplifies the structure of the sweat device, reducing weight and the number of external interfaces, and improving the device's waterproofness.

[0094] 3. This invention employs interdigitated electrode adaptive technology, which uses different detection levels in the hardware to handle sweat flow rates of varying speeds. This expands the range of sweat flow rate acquisition across different concentrations, improving the applicability and accuracy of sweat flow rate acquisition. Furthermore, the time-division detection mechanism and automatic shifting mechanism allow the sweat detection device to be compatible with various sensor patch sizes, expanding its application scenarios and improving cost-effectiveness.

[0095] 4. To acquire sweat concentration and velocity signals, this invention innovatively applies a high-frequency AC excitation signal to both ends of the electrodes of the sweat sensor patch to obtain the electrode impedance (i.e., resistance). Therefore, the wearable sweat detection device acquires the signal returned by the sensor patch as an AC signal. Real-time processing of AC signals requires a powerful MCU. To reduce the requirements on the MCU, this invention converts the AC signal of sweat into a DC signal in step S13, thereby facilitating MCU acquisition and processing.

[0096] 5. This invention integrates a cloud platform with a big data trend prediction model. The software module synchronizes data to the cloud platform via the network. The cloud platform uses this data to build a big data model based on trends, processes the collected data, and feeds back the results (trend predictions) to the software module for display. The software module provides health reminders and suggestions, as well as exercise recommendations, making predictions and exercise suggestions more reasonable and accurate. Attached Figure Description

[0097] Figure 1 This is a system framework diagram of the present invention;

[0098] Figure 2 This is a flowchart of the detection method of the present invention;

[0099] Figure 3 Schematic diagram of the detection principle of the sweat sensor patch;

[0100] Figure 4 This is a flowchart of the method for sweat signal acquisition and preprocessing in this invention;

[0101] Figure 5 A step signal diagram formed by the interdigital electrode under different sweat concentrations;

[0102] Figure 6 This is a flowchart of the sweat signal analysis and processing method in this invention. Detailed Implementation

[0103] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of the present invention. However, the present invention can be practiced in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0104] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0105] like Figure 1 As shown, the present invention provides a wearable sweat detection system, including a sweat sensor patch, a wearable sweat detection device, a software module, and a cloud platform.

[0106] The sweat sensor patch uses a flexible wearable dual-channel sweat sensor as disclosed in patent publication number CN118319299A, which is used to collect sweat secreted by the human skin.

[0107] The wearable sweat detection device includes a sweat signal acquisition and preprocessing module, a sweat signal analysis module, a data storage module, and a communication module.

[0108] Furthermore, the wearable sweat detection device also includes a housing, a PCB motherboard, the chips and peripheral components required for the aforementioned modules, electrode contacts that make electrical contact with the sweat sensor patch, a data interface, and a power management module and battery. The power supply preferably uses a rechargeable lithium battery, which is integrated into the housing of the detection device along with the PCB motherboard.

[0109] Furthermore, the data interface primarily uses a USB port, allowing the battery to be charged via USB. Of course, the wearable sweat detection device also includes a wireless charging module, enabling the battery to be charged wirelessly via a wireless charger and wireless charging module.

[0110] The software modules include, but are not limited to, software integrated into wearable sweat detection devices, software installed on computers, or software installed on mobile devices. Of course, the software can be a standalone program, a mobile app, or a mini-program within other software.

[0111] Furthermore, the software modules should prioritize mobile APP software or mini-program software that are based on mobile devices.

[0112] The cloud platform is used for big data analysis and big data model building.

[0113] The wearable sweat detection system of the present invention acquires sweat signals by injecting a high-frequency oscillation signal into a sweat sensor patch to obtain the impedance of sweat, and by using a sweat signal analysis module to obtain the concentration and flow rate of sweat.

[0114] The method of sweat collection is as follows: the side of the sweat sensor patch with sweat injection holes is attached to the human skin, and the other side of the sensor patch is attached to the wearable sweat detection device. The sweat secreted by the human sweat glands enters the microchannels in the sweat sensor patch, thereby collecting the sweat signal.

[0115] like Figure 2 As shown, based on the wearable sweat detection system of the present invention, the present invention also provides a detection method for the wearable sweat detection system, comprising the following steps:

[0116] S1: Sweat signal acquisition and preprocessing;

[0117] S2: Sweat signal analysis and processing:

[0118] S3: Data storage and data transmission;

[0119] S4: Data Presentation;

[0120] S5: Cloud Platform Trend Forecast.

[0121] like Figure 3 The diagram illustrates the detection principle of the sweat sensor patch. Sweat enters the signal acquisition area of ​​the sweat sensor patch from its chamber. As the amount of sweat increases, the total resistance output by the sensor patch changes abruptly, forming a step signal. The height of the step, i.e., the conductivity value, is related to the sweat concentration; the width of the step, i.e., the time duration, is related to the sweat flow rate.

[0122] Therefore, as Figure 4 As shown, in step S1, the sweat signal acquisition and preprocessing step further includes the following steps:

[0123] S11: Sweat flow rate signal acquisition;

[0124] S12: Sweat concentration signal acquisition;

[0125] S13: Preprocessing of sweat AC signal to sweat DC signal.

[0126] The acquisition of sweat flow rate and sweat concentration signals is achieved through electrode sampling switching control technology. Based on the signal variation characteristics, to reduce the number of connection points between the sweat detection device and the sweat sensor patch, the sweat concentration detection electrode (i.e., the fixed electrode in the sweat sensor patch) and the sweat flow rate detection electrode (i.e., the interdigitated electrode in the sweat sensor patch) share a common pair of electrodes. The switching control module takes turns detecting these electrodes according to a time-division multiplexing principle, thus ensuring both signal accuracy and device simplicity.

[0127] Furthermore, in steps S11 and S12, the sweat flow rate signal acquisition is used to collect the sweat flow rate signal, and the sweat concentration signal acquisition is used to collect the sweat concentration signal. To collect the sweat concentration and flow rate signals, this invention innovatively uses a high-frequency oscillation signal (high-frequency AC excitation signal) applied to both ends of the electrodes of the sweat sensor patch to obtain the electrode impedance (i.e., resistance). Therefore, the wearable sweat detection device collects the signal returned by the sensor patch as an AC signal. Real-time processing of AC signals requires a powerful MCU. To reduce the requirements on the MCU, this invention converts the AC signal of sweat into a DC signal in step S13, thereby facilitating MCU acquisition and processing.

[0128] Furthermore, when the sweat sensor patch sampling switches to the interdigital electrodes according to the time-division multiplexing principle (i.e., electrode sampling switching control technology) to sample sweat flow rate, the high-frequency output signal at this time carries sweat flow rate information. For example... Figure 5 As shown, the slope of the step signal formed by the interdigital electrode varies with different sweat concentrations; the higher the concentration, the higher the slope, and the slope varies significantly across different concentrations. Figure 5 As shown, from top to bottom, they are 200 mmol / L. -1, 150 mmol / L -1 , 100 mmol / L -1 , 50 mmol / L -1 , 20 mmol / L -1 and 10 mmol / L -1 step signals at different concentrations

[0129] Too high slope signals will exceed the range set by the hardware. Therefore, in order to detect sweat within a wide concentration range, the present invention adopts an adaptive technology for interdigitated electrode positions. By setting different detection positions in the hardware of the wearable sweat detection device to cope with sweat at different flow rates, and then through the control software of the device, the analog switch chip is switched to operational amplifier circuits with different magnification factors to achieve adaptive switching of the interdigitated electrode positions. The method for selecting and switching positions is as follows:

[0130] A: When V(n)>VH(n) and V(n + 1)>VL(n + 1), switch to position n + 1;

[0131] B: When V(n)<VL(n) and V(n - 1)>VL(n - 1), switch to position n - 1;

[0132] Here, n (n = 1, 2, 3,..., m) is the position of the interdigitated electrode, VL(n) is the low voltage threshold corresponding to position n, VH(n) is the high voltage threshold corresponding to position n, and V(n) is the voltage value measured at position n.

[0133] By introducing an automatic electrode position shifting mechanism, the present invention can detect sweat parameters within a wide range (1 - 10000). At the same time, the time-sharing detection mechanism and the automatic position shifting mechanism enable the sweat detection device to adapt to various specifications of sensor patches, expand the usage scenarios, and improve the economy.

[0134] Furthermore, when the sweat sensor patch samples and switches to a fixed electrode according to the time-sharing multiplexing principle (i.e., the electrode sampling switching control technology) for sweat concentration sampling, the high-frequency output signal at this time carries sweat concentration information. Then, this signal is passed through an AC-DC conversion module to obtain a DC characteristic value. As long as the correspondence between the concentration and the DC characteristic value is established, the corresponding sweat concentration value can be obtained from the DC characteristic value of the fixed electrode signal.

[0135] As Figure 6 shown, it is the steps of sweat signal analysis and processing. The sweat signal analysis and processing of the present invention further includes the following steps:

[0136] S21: Extracting sweat flow rate;

[0137] S22: Calculating sweat concentration;

[0138] S23: Calculating sweat loss amount;

[0139] S24: Calculation of electrolyte loss;

[0140] S25: Calculation of water shortage level;

[0141] S26: Water replenishment calculation.

[0142] In step S21, when sweat passes through the interdigitated electrode array in the sweat sensor patch, a step signal is formed. The width of the step signal is related to the flow velocity. In this invention, sweat flow velocity extraction employs step signal edge recognition technology or abrupt change point recognition technology. The recognition algorithm includes the following steps:

[0143] S211: Preset a queue of length 2n, and set a recognition value of k;

[0144] S212: Clear the queue;

[0145] S213: Store the collected data (DC characteristic values ​​after AC signal is converted to DC signal) into the queue;

[0146] S214: When the queue is full, calculate the average value of the data in the queue as m;

[0147] Record L, where L is the number of values ​​less than (mk) in the 0 to n part of the queue;

[0148] Record R, where R is the number of values ​​greater than (m+k) in the n to 2n part of the queue;

[0149] S215: Obtain the step edge signal and time: If L>2n*f and R>2n*f (f is the matching factor), then it is a step edge signal. Record the current time as Tn and jump to step S212; otherwise, jump to step S213.

[0150] S216: Calculate the step signal time T, T = T n -T n-1 T is the time of the edge signal of the nth step minus the time of the edge signal of the (n-1)th step.

[0151] S217: Calculate the sweat flow rate S, S = V / T, where V is the volume between the two interdigitated electrodes. Different models of sweat sensor patches have different interdigitated electrode volumes between their internal interdigitated electrodes. This volume is calculated during the design of the sweat sensor patch.

[0152] S218: Jump to step S212.

[0153] Furthermore, the present invention also provides another method for extracting sweat flow rate, comprising the following steps:

[0154] S211: Obtain the step edge signal.

[0155] Calculate Teen_err,

[0156] Teen_err=Teen_V n -Teen_V n-1 ;

[0157] Among them, Teen_V n The nth DC characteristic value corresponding to the conversion is injected into the high-frequency oscillation signal of the interdigital electrode, where n is an integer.

[0158] determination,

[0159] Teen_err>Teen_K;

[0160] Here, Teen_K is a set value (preset value).

[0161] If Teen_err > Teen_K, then it means that in Teen_t n A step edge signal appears at this point, and the corresponding time point Teen_t is recorded. n .

[0162] S212: Calculate the sweat flow rate S.

[0163] S=VOL÷(Teen_t n -Teen_t n-1 );

[0164] Where VOL is the volume between the two interdigitated electrodes, and Teen_t n For Teen_V n The time point corresponding to the step edge signal, Teen_t n-1 This is the time point corresponding to the edge signal of the previous step.

[0165] like Figure 6 As shown, the method for calculating sweat concentration is as follows:

[0166] S221: Establish a linear fitting equation for sweat concentration, g.

[0167]

[0168] Where f is the univariate linear regression model function, and g is the obtained linear fitting equation;

[0169] Take n points for the sweat concentration range, where the concentration value of the i-th point is;

[0170] For DC characteristic value of the high-frequency signal of the fixed electrode corresponding to the concentration;

[0171] S222: Sweat concentration conversion, calculation of sweat concentration d.

[0172] d = g(v);

[0173] The sweat concentration can be obtained by substituting the DC value v of the high-frequency signal converted from the fixed electrode into the g function.

[0174] like Figure 6 As shown, the calculation of sweat loss also includes the following steps:

[0175] S231: Calculate and obtain the human body surface area (BSA).

[0176] BSA=0.0061*H+0.0124*W-0.0099;

[0177] H represents height (in cm), and W represents weight (in kg).

[0178] BSA stands for body surface area (unit: m²). 2 ).

[0179] S232: Calculate and obtain the rate of sweating throughout the body, V. 全身 ,

[0180] V 全身 =a*V 局部 +b;

[0181] V 全身 The unit is L / (min*cm) 2 );

[0182] Where parameters a and b depend on the position where the sensor is attached, V 局部 The sweat flow rate is calculated by multiplying the extracted sweat flow rate by the cross-sectional area coefficient of the flow channel inside the sweat sensor patch. The cross-sectional area coefficient of the flow channel is calculated during the sensor design.

[0183] S233: Calculate and obtain the total amount of sweat loss M. 汗液 ,

[0184] U 全身 =BSA*10000*V 全身 *t, where t is time, in minutes;

[0185] M 汗液 =U 全身 *p 汗液 ;

[0186] Where, p 汗液 =1g / cm3; p 汗液 This refers to the density of sweat.

[0187] S234: Calculate and obtain the total body dehydration percentage W 全身 ,

[0188] M 丢失 =M 汗液 ;

[0189] W 全身 =(M 丢失 / M 体重 )*100%;

[0190] Among them, M 体重 The user's body weight is then configured via a software module and a mobile app. The configuration data is then transmitted to the wearable sweat detection device via a wireless communication module.

[0191] like Figure 6 As shown, the calculation of electrolyte loss also includes the following steps:

[0192] S241: Calculate and obtain the total body electrolyte concentration C 全身 ,

[0193] C 全身 =a*C 局部 +b;

[0194] C 全身 The unit is: mmol / L;

[0195] Among them, parameters a and b depend on the position where the sensor is attached.

[0196] S242: Calculate and obtain the rate of total electrolyte loss Vc 全身 ,

[0197] Vc 全身 =C 全身 *V 全身 *BSA*10000.

[0198] S243: Calculate total electrolyte loss Mc 全身 ,

[0199] Mc 全身 =Vc 全身 *M NaCl *t;

[0200] t represents the motion time, in minutes;

[0201] M NaCl Molar mass, measured in g / mol = mg / mmol, is the molar mass M in one embodiment. NaCl Set to

[0202] 58.5 g / mol.

[0203] like Figure 6 As shown, the determination of water shortage level is based on the following logic:

[0204] When 2%≤W 全身 When the level is less than 3%, the software module will display and remind you: Thirsty;

[0205] When 3%≤W 全身 When the level is less than 4%, the software module will display and remind you of the symptoms: restlessness and loss of appetite.

[0206] When 4%≤W 全身 When the temperature is less than 5%, the software module will display and alert you to symptoms such as flushed skin, elevated body temperature, and fatigue.

[0207] When 5% ≤ W 全身 When the level is less than 8%, the software module will display and remind you of the following symptoms: fever, dizziness, weakness, and decreased urination.

[0208] When 8%≤W 全身 When the percentage is less than 10%, the software module will display and alert: convulsions, shock;

[0209] When W 全身 When the percentage is ≥10%, the software module will display and alert: no urine, death.

[0210] like Figure 6 As shown, the formula for calculating water replenishment is as follows:

[0211] M w 丢失 =M 汗液 -M 补水 .

[0212] like Figure 1 As shown, the software module can be integrated into the wearable sweat detection device of this invention, or installed in a computer, or in a mobile device such as a mobile phone, tablet, or smartwatch. In this invention, a mobile device is preferred as the carrier of the software APP.

[0213] Furthermore, the communication module in the wearable sweat detection device of the present invention preferentially adopts a low-power wireless Bluetooth communication chip to form a wireless communication module. The wearable sweat detection device sends the detected and calculated data to a mobile APP through the wireless communication module, and also backs up a copy of the data through the device's internal storage module. This ensures that data is not lost even if the device experiences wireless communication failure. Users only need to carry the wearable sweat detection device of the present invention to perform sweat detection and analysis, which greatly reduces the user's carrying burden and ensures long-term data integrity.

[0214] Further, such as Figure 2 As shown, the data transmitted by the mobile app will be further stored locally in the mobile device's storage chip.

[0215] Meanwhile, the mobile device can also send corresponding configuration information to the wearable sweat detection device via the communication module, thereby flexibly adjusting the device's operating parameters. For example, if different specifications of sweat sensor patches are selected according to the type of exercise, the wearable sweat detection device can be used simply by reconfiguring it through the mobile app. Compared to other sweat detection devices that can only use one type of sensor, this greatly improves the convenience and cost-effectiveness of use.

[0216] Furthermore, the wearable sweat detection device sends the sweat data processed by the sweat signal analysis module to the software module via the communication module. The software module performs visual analysis and display of the sweat data, showing in real time the changes in sweat concentration, sweat flow rate, and curves of body water and electrolyte loss over time. Combined with dehydration level judgment logic, it provides reminders about the degree of dehydration during exercise and prompts for rehydration. Simultaneously, the software module synchronizes the data to a cloud platform via the network. The cloud platform uses a big data model based on trends to process the collected data and feeds back the processing results (trend predictions) to the software module for display. The software module then provides health reminders and suggestions, as well as exercise recommendations.

[0217] Furthermore, the software module is installed in smartphones, smartwatches, and tablets, using these devices as platforms to visualize and display data.

[0218] like Figure 1 As shown, the cloud platform utilizes cloud computing to analyze sweat big data and build big data models. Based on trends, the cloud platform establishes data models to predict physical condition and provide exercise suggestions.

[0219] Furthermore, trend prediction in cloud platforms includes the following steps:

[0220] S51: Calculate and obtain the estimated total body electrolyte level An.

[0221] An = f(Mc) 全身 M 汗液 );

[0222] Where An is the estimated total body electrolyte level, n is a positive integer, and f is the total body electrolyte estimation function, which estimates the total body electrolyte concentration based on the total body electrolyte loss and the total body sweat loss.

[0223] S52: Calculate and obtain the electrolyte loss rate Bs corresponding to the s-type exercise.

[0224] Bs = ma(Vc) 全身 );

[0225] Where ma is the smoothing filter and Bs is the electrolyte loss rate corresponding to the s motion type.

[0226] S53: Calculate and obtain the whole-body electrolysis trend value Cn.

[0227] Cn = pettitt(An...A1);

[0228] Here, Pettitt is the mutation detection algorithm, and An to A1 represent historical data. Cn is the whole-body electrolysis trend value calculated by the Pettitt function, a classic trend prediction algorithm.

[0229] S54: Calculate and obtain the reference comparison deviation ratio Kn.

[0230] Kn = h(Cn);

[0231] Where h is the standard reference comparison function, and Kn is the reference comparison deviation ratio.

[0232] S55: Build data models and perform trend prediction.

[0233] When |Kn| < 10%, the electrolyte level trend is normal, indicating good health.

[0234] When 10% < |Kn| < 30%, the electrolyte level trend is slightly abnormal, and reasonable exercise and diet are recommended;

[0235] When 30% < |Kn| < 50%, the electrolyte level trend is moderately abnormal, and it is recommended to check for potential health problems.

[0236] When 50% < |Kn|, the electrolyte level trend indicates a severe abnormality, suggesting an abnormal health condition and recommending further examination.

[0237] S56: Exercise duration suggestion, calculate and obtain the suggested exercise duration Tn.

[0238] Tn = g(Bs, Kn);

[0239] Where g is the reasonable motion duration Tn for calculating motion k.

[0240] like Figure 4 As shown, the formula for converting AC sweat signals to DC sweat signals is as follows:

[0241]

[0242] Where x is the signal sampling point and n is the number of sampling points.

[0243] A high-frequency sinusoidal signal is injected into the sweat sensor patch to obtain an AC signal with impedance information. Then, the AC signal is converted into a DC characteristic value X using the root mean square formula mentioned above. rms The DC characteristic value is input into the microprocessor of the detection device for signal analysis, and the sweat flow rate and sweat concentration values ​​are obtained through the previous algorithm.

[0244] Furthermore, the present invention also provides another conversion formula for converting AC sweat signals to DC sweat signals:

[0245]

[0246] Among them, Amp out The amplitude of the high-frequency injected sinusoidal signal;

[0247] The amplitude of the input signal acquired in the Nth acquisition;

[0248] The amplitude ratio of the nth input to the output;

[0249] This represents the amplitude ratio after filtering and smoothing.

[0250] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A detection method for a wearable sweat detection system, characterized in that: The detection system includes: a sweat sensor patch, a wearable sweat detection device, a software module, and a cloud platform; The sweat sensor patch is used to collect the sweat secreted by the human epidermis; The wearable sweat detection device includes a sweat signal acquisition and preprocessing module, a sweat signal analysis module, a data storage module, and a communication module; The sweat signal acquisition and preprocessing module is used for the acquisition and preprocessing of the sweat flow rate signal and the sweat concentration signal. The sweat acquisition and preprocessing module obtains the impedance of the electrodes on the sweat sensor patch by inputting a high-frequency oscillation signal into the sweat sensor patch. The electrodes include a fixed electrode and interdigital electrodes; the high-frequency oscillation signal drives the interdigital electrodes to form a sweat flow rate alternating current signal, and the high-frequency oscillation signal drives the fixed electrode to form a sweat concentration alternating current signal; the sweat flow rate alternating current signal and the sweat concentration alternating current signal are preprocessed by the sweat signal acquisition and preprocessing module and converted into corresponding direct current signal values; The sweat signal analysis module is used for the extraction of the sweat flow rate, the calculation of the sweat concentration, the calculation of the sweat loss amount, the calculation of the electrolyte loss amount, the calculation of the degree of water shortage, and the calculation of water replenishment; The communication module is used for wired or wireless communication with the software module, transmits the sweat data processed by the sweat signal analysis module to the software module, and visualizes and displays the sweat data through the software module; The software module transmits the sweat data to the cloud platform through the network, performs big data analysis and builds a big data model through the cloud platform, and feeds back the result of the trend prediction processing to the software module through the network for display; The detection method includes the following steps: S1: Sweat signal acquisition and preprocessing; S2: Sweat signal analysis and processing; S3: Data storage and data transmission; S4: Data presentation; S5: Cloud platform trend prediction; In step S1, the sweat signal acquisition and preprocessing further includes the following steps: S11: Sweat flow rate signal acquisition; S12: Sweat concentration signal acquisition; S13: Preprocessing of converting sweat alternating current signal to sweat direct current signal; In step S11: The sweat flow rate signal acquisition adopts the interdigital electrode adaptive technology. The interdigital electrode adaptive technology sets operational amplifier circuits with different magnification factors in the hardware of the wearable sweat detection device, and switches on the operational amplifier circuits with different magnification factors through the control software simulation switch chip in the wearable sweat detection device to achieve the adaptive switching of the interdigital electrode gears, so as to adapt to sweats with different flow rates. The method for selecting and switching gears includes: When V(n)>VH(n) and V(n + 1)>VL(n + 1), switch to the n + 1 gear; When V(n)<VL(n) and V(n - 1)>VL(n - 1), switch to the n - 1 gear; Wherein, n (n = 1, 2, 3, ..., m) is the interdigitated electrode setting, VL(n) is the low voltage threshold corresponding to setting n, VH(n) is the high voltage threshold corresponding to setting n, and V(n) is the voltage value measured at setting n.

2. The detection method of a wearable sweat detection system according to claim 1, characterized in that: The acquisition of sweat flow rate signal and the acquisition of sweat concentration signal are switched and controlled by electrode sampling switching technology. The electrode sampling switching technology uses a pair of electrodes shared by interdigitated electrodes and fixed electrodes, and the acquisition is carried out in turn by the switching control module in the wearable sweat detection device according to the time-division multiplexing principle.

3. The detection method of a wearable sweat detection system according to claim 1 or 2, characterized in that: The conversion formula for the AC signal of sweat to DC signal of sweat is as follows: Where x is the signal sampling point, n is the number of sampling points, and X rms These are the converted DC characteristic values; Alternatively, the formula can be used: Among them, Amp out The amplitude of the high-frequency injected sinusoidal signal; The amplitude of the input signal acquired in the Nth acquisition; The amplitude ratio of the nth input to the output; This represents the amplitude ratio after filtering and smoothing.

4. The detection method of a wearable sweat detection system according to claim 1, characterized in that: In step S2, the sweat signal analysis and processing further includes the following steps: S21: Sweat flow rate extraction process; S22: Calculation and processing of sweat concentration; S23: Calculation and processing of sweat loss; S24: Calculation and processing of electrolyte loss; S25: Calculation and processing of water shortage degree; S26: Water replenishment calculation and processing.

5. The detection method of a wearable sweat detection system according to claim 4, characterized in that: In step S21, the sweat flow rate extraction process employs step signal edge recognition technology, which includes the following steps: S211: Preset a queue of length 2n, and set a recognition value of k; S212: Clear the queue; S213: Store the collected and preprocessed sweat data into the queue; S214: When the queue is full, the average value of the queue data is calculated as m; Record L, where L is the number of values ​​less than (mk) in the 0 to n part of the queue; Record R, where R is the number of values ​​greater than (m+k) in the n to 2n part of the queue; S215: Obtain the step edge signal and time: If L>2n*f and R>2n*f (f is the matching factor), then it is a step edge signal. Record the current time as Tn and jump to step S212. Otherwise, jump to step S213. S216: Calculate the step signal time T, T = T n -T n-1 T is the time of the edge signal of the nth step minus the time of the edge signal of the (n-1)th step. S217: Calculate the sweat flow rate S, S = V / T, where V is the volume between the two interposed electrodes; S218: Proceed to step S212; Alternatively, the sweat flow rate can be obtained using the following steps: S211a: Obtain the step edge signal. Calculate Teen_err, Teen_err=Teen_V n -Teen_V n-1 ; Among them, Teen_V n The nth DC characteristic value corresponding to the conversion is injected into the high-frequency oscillation signal of the interdigital electrode; where n is an integer. determination, Teen_err>Teen_K; Where Teen_K is a set value (preset value); If Teen_err > Teen_K, then it means that in Teen_t n A step edge signal appears at this point, and the corresponding time point Teen_t is recorded. n ; S212a: Calculate the sweat flow rate S. S=VOL÷(Teen_t n -Teen_t n-1 ); Where VOL is the volume between the two interdigitated electrodes, and Teen_t n For Teen_V n The time point corresponding to the step edge signal, Teen_t n-1 This is the time point corresponding to the edge signal of the previous step.

6. The detection method of a wearable sweat detection system according to claim 1, characterized in that: In step S5, the cloud platform trend prediction further includes the following steps: S51: Calculate and obtain the estimated total body electrolyte level An. An=f(Mc 全身 ,M 汗液 ); Where An is the estimated total body electrolyte level, and n is a positive integer; Mc 全身 M represents the total electrolyte loss. 汗液 This represents the amount of sweat lost throughout the body. S52: Calculate and obtain the electrolyte loss rate Bs corresponding to the s-type exercise. Bs=ma(Vc 全身 ); Where ma is the smoothing filter, Bs is the electrolyte loss rate corresponding to motion type s, and Vc 全身 This represents the rate of electrolyte loss throughout the body. S53: Calculate and obtain Cn. Cn = pettitt(Bn...B1); Among them, Pettitt is the mutation detection algorithm, and Bn to B1 are historical data; S54: Calculate and obtain the reference comparison deviation ratio Kn. Kn = h(Cn); Where h is the standard reference contrast function, and Kn is the reference contrast deviation ratio; S55: Build data models and perform trend prediction. When |Kn| < 10%, the electrolyte level trend is normal, indicating good health. When 10% < |Kn| < 30%, the electrolyte level trend is slightly abnormal, and reasonable exercise and diet are recommended; When 30% < |Kn| < 50%, the electrolyte level trend is moderately abnormal, and it is recommended to check if there is an underlying health problem; when 50% < |Kn|, the electrolyte level trend is severely abnormal, and it is recommended to check if there is an abnormal health condition. S56: Exercise duration suggestion, calculate and obtain the suggested exercise duration Tn. Tn = g(Bs, Kn); Where g is the reasonable motion duration Tn for calculating motion k.

Images