Disclosure of Invention
The technical problem to be solved by the invention is to provide a light and convenient wearable sweat detection system and a detection method thereof, wherein the sweat detection system can be flexibly allocated according to the use scene, can continuously collect and detect sweat, and can realize real-time monitoring and early warning of human health state by utilizing sweat in a mode of integrating data acquisition, analysis and prejudgment.
The technical scheme adopted for solving the technical problems is as follows:
a wearable sweat detection system comprising a sweat sensor patch, a wearable sweat detection device, a software module, and a cloud platform;
the sweat sensor patch is used for collecting sweat secreted by human epidermis;
the wearable sweat detection equipment comprises 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 acquiring and preprocessing sweat flow rate signals and sweat concentration signals, the sweat acquisition and preprocessing module is used for acquiring the impedance of an electrode on the sweat sensor patch by inputting a high-frequency oscillation excitation signal into the sweat sensor patch, the electrode comprises a fixed electrode and an interdigital electrode, the high-frequency oscillation signal drives the interdigital electrode to form a sweat flow rate alternating current signal, the high-frequency oscillation signal drives the fixed electrode to form a sweat concentration alternating current signal, and the sweat flow rate alternating current signal and the sweat concentration alternating current signal are converted into corresponding direct current signal values through the preprocessing of the sweat signal acquisition and preprocessing module;
The sweat signal analysis module is used for extracting sweat flow rate, calculating sweat concentration, calculating sweat loss, calculating electrolyte loss, calculating water shortage degree and calculating water supplementing;
The communication module is used for carrying out wired communication or wireless communication with the software module, transmitting the sweat data processed by the sweat signal analysis module to the software module, and carrying out visual display and display on the sweat data through the software module;
And the software module transmits the sweat data to a cloud platform through a network, performs big data analysis and builds a big data model through the cloud platform, and feeds back a result after trend prediction processing to the software module through the network for display.
Further, the software module comprises software integrated in the wearable sweat detection device, or software installed on a computer side, or software installed on a mobile side.
Further, the wearable sweat detection device further comprises a housing, a PCB motherboard, chips and electronic devices required by each module, electrode contacts electrically connected with the sweat sensor patches, a power management module and a battery.
A method of detecting a wearable sweat detection system, comprising the steps of:
S1, sweat signal acquisition and pretreatment;
s2, sweat signal analysis and treatment;
S3, data storage and data transmission;
S4, data display;
S5, cloud platform trend prediction.
Further, 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 sweat alternating current signals to sweat direct current signals.
Furthermore, the sweat flow signal acquisition and the sweat concentration signal acquisition are switched and controlled by an electrode sampling switching technology, wherein the electrode sampling switching technology is that the interdigital electrode and the fixed electrode share a pair of electrodes, and the sweat flow signal acquisition and the sweat concentration signal acquisition are alternately acquired by a switching control module in the wearable sweat detection equipment according to a time division multiplexing principle.
Furthermore, the sweat flow rate signal acquisition adopts an interdigital electrode self-adaption technology, the interdigital electrode self-adaption technology sets operational amplifier circuits with different amplification factors in hardware of the wearable sweat detection equipment, and switches on the operational amplifier circuits with different amplification factors through a control software analog switch chip in the wearable detection equipment so as to realize self-adaption switching of interdigital electrode gears, thereby adapting to sweat with different flow rates, and the gear selection and switching method comprises the following steps:
when V (n) > VH (n) and V (n+1) > VL (n+1), shift to n+1 gear;
When V (n) < VL (n) and V (n-1) > VL (n-1), shift to n-1;
Wherein n (n=1, 2, 3,..m) is the interdigital electrode gear, VL (n) is a low voltage threshold corresponding to the n gear, VH (n) is a high voltage threshold corresponding to the n gear, and V (n) is a voltage value measured in the n gear.
Further, the conversion formula for converting the sweat alternating current signal into the sweat direct current signal is as follows:
Wherein X is a signal sampling point, n is a sampling point number, and Xrms is a converted direct current characteristic value;
or the formula:
wherein, ampout is the amplitude of the high-frequency injection sinusoidal signal;
The amplitude value of the input signal acquired for the nth time is set;
The amplitude ratio of the n-th input and the n-th output is set;
is the amplitude ratio after filtering and smoothing.
Further, in step S2, the sweat signal analysis processing further includes the steps of:
S21, sweat flow rate extraction treatment;
s22, sweat concentration calculation processing;
S23, sweat loss calculation processing;
S24, electrolyte loss amount calculation processing;
S25, calculating and processing the water shortage degree;
S26, water supplementing calculation processing.
Further, in step S21, the sweat flow extraction process employs a step signal edge recognition technique including the steps of:
s211, presetting a queue with the length of 2n, and setting a value of identification degree as k;
s212, emptying a queue;
s213, storing the collected and preprocessed sweat data into the queue;
s214, when the queue is full, obtaining an average value m in the queue data;
recording L, wherein L is the number of values of which the queue 0~n part is smaller than (m-k);
recording R, wherein R is the number of values of which the n-2 n parts of the queue are larger than (m+k);
S215, acquiring a step edge signal and time, wherein if L is more than 2n f and R is more than 2n f (f is a matching factor), the step edge signal is a step edge signal, the current time is recorded as Tn, and meanwhile, the step S212 is skipped, and otherwise, the step S213 is skipped;
S216, calculating step signal time T, wherein T=Tn-Tn-1, and T is the nth step edge signal time minus the (n-1) th step edge signal time;
s217, calculating sweat flow rate S, wherein S=V/T, and V is the volume between two interdigital electrodes;
S218, jumping to step S212;
or the sweat flow rate is obtained by the following steps:
s211a of acquiring a step edge signal,
The calculation of the Teen _ err,
Teen_err=Teen_Vn-Teen_Vn-1;
Wherein Teen_Vn is the corresponding converted nth DC characteristic value of the high-frequency oscillation signal injection of the interdigital electrode, and n is an integer.
It is determined that the number of the cells to be processed,
Teen_err>Teen_K;
Wherein Teen_K is a set value (preset value).
If Teen_err > Teen_K, it indicates that a step edge signal is present at Teen_tn and the current station is recorded
A time point Teen_tn corresponding to the edge signal;
s212a to calculate the sweat flow rate S,
S=VOL÷(Teen_tn-Teen_tn-1);
Wherein VOL is the volume between two interdigital electrodes, tentn is the time point corresponding to the step edge signal of tenvn, and tentn-1 is the time point corresponding to the last step edge signal.
Further, the cloud platform trend prediction further comprises the following steps:
S51, calculating and obtaining the current whole body electrolyte estimation amount An,
An=f(Mc Whole body,M sweat liquid);
Wherein An is the estimated total body electrolyte, n is a positive integer, mc Whole body is the total body electrolyte loss, M sweat liquid is the total body sweat loss;
S52, calculating and obtaining the electrolyte loss rate Bs corresponding to the S motion type,
Bs=ma(Vc Whole body);
Wherein ma is a smoothing filter, bs is an electrolyte loss rate corresponding to the s motion type, and Vc Whole body is a whole body electrolyte loss rate;
s53 is to calculate the acquisition Cn,
Cn=pettitt(Bn...B1);
Wherein Pettitt is a mutation detection algorithm, and Bn to B1 are previous data;
s54 the acquired reference contrast deviation ratio Kn is calculated,
Kn=h(Cn);
Wherein the h function is a standard reference contrast function, and Kn is a reference contrast deviation ratio;
S55, constructing a data model, carrying out trend prediction,
When |kn| <10%, the electrolyte content trend is normal, and the body is healthy;
when 10% </kn| <30%, the electrolyte content trend is slightly abnormal, suggesting reasonable exercise and diet;
when 30% </kn| <50%, the electrolyte content trend is moderate anomaly degree, and the body has potential health problem advice examination;
S56, calculating and acquiring recommended movement duration Tn,
Tn=g(Bs,Kn);
Wherein g is the reasonable movement duration Tn under the calculation k movement.
The beneficial effects of the invention are as follows:
1. the wearable sweat detection system and the detection method are light and convenient, can be flexibly allocated according to the use scene, can continuously collect and detect sweat, and realize real-time monitoring and early warning of human health state by utilizing sweat in a mode of integrating data acquisition, analysis and prejudgment.
2. The sweat flow rate signal acquisition and sweat concentration signal acquisition in the invention are switched and controlled by an electrode sampling switching control technology. According to the change characteristics of the signals, in order to reduce the connection points of the sweat detection equipment and the sweat sensor patch, the sweat concentration detection electrode (namely, the fixed electrode in the sweat sensor patch) and the sweat flow detection electrode (namely, the interdigital electrode in the sweat sensor patch) share a pair of electrodes, and the signals are alternately detected by the switching control module according to the time-sharing multiplexing principle, so that the accuracy of the signals is ensured, and the simplicity and convenience of the equipment structure are ensured. The sweat concentration and sweat flow time-sharing detection mechanism greatly simplifies the structure of the sweat equipment, reduces the number of external interfaces while reducing the weight, and improves the waterproofness of the equipment.
3. According to the invention, the interdigital electrode gear self-adaption technology is adopted, and sweat at different flow rates is handled by setting different detection gears in hardware, so that sweat flow rate acquisition in different concentration ranges can be widened, and the applicable scene and accuracy of sweat flow rate acquisition are improved. Meanwhile, the time-sharing detection mechanism and the automatic gear shifting mechanism can enable sweat detection equipment to be suitable for sensor patches of various specifications, and therefore the use scene is expanded and the economical efficiency is improved.
4. In order to acquire sweat concentration signals and sweat velocity signals, the invention innovatively applies high-frequency alternating-current excitation signals to both ends of an electrode of a sweat sensor patch to obtain impedance (i.e. resistance) of the electrode. Therefore, the signal returned by the sensor patch collected by the wearable sweat detection equipment is an alternating current signal, and a powerful MCU is required for real-time processing of the alternating current signal, so that the requirement on the MCU is reduced.
5. According to the method, a big data trend suggestion data model is introduced by combining a cloud platform, a software module synchronizes data into the cloud platform through a network, the cloud platform builds the big data model based on trends to process collected data, and a processed result (trend prediction) is fed back to the software module to be displayed. Health reminding and suggestion, sports suggestion and the like are carried out through the software module, so that prediction and sports suggestion can be more reasonable and accurate.
Detailed Description
In order that the above objects, features and advantages of the invention will be readily understood, a more particular description of the invention will be rendered by reference to the appended drawings. In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. The present invention may be embodied in many other forms than described herein and similarly modified by those skilled in the art without departing from the spirit of the invention, whereby the invention is not limited to the specific embodiments disclosed below.
Unless defined otherwise, 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 belongs. 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.
As shown in fig. 1, the present invention provides a wearable sweat detection system comprising a sweat sensor patch, a wearable sweat detection device, a software module, and a cloud platform.
Wherein, sweat sensor paster adopts a flexible wearable binary channels sweat sensing in patent publication No. CN118319299A for collect the sweat that human epidermis secreted.
The wearable sweat detection device comprises a sweat signal acquisition and preprocessing module, a sweat signal analysis module, a data storage module and a communication module.
Further, the wearable sweat detection device further comprises a housing, a PCB motherboard, chips and peripheral components required by the respective modules, electrode contacts in electrical contact with the sweat sensor patches, a data interface, a power management module and a battery. The power supply is preferably a rechargeable lithium battery, and is integrally arranged in the shell of the detection device together with the PCB main board.
Further, the data interface preferably adopts a USB interface, so that the battery can be charged through the USB interface. Of course, the wearable sweat detection device also includes a wireless charging module to charge the battery through the wireless charger and the wireless charging module.
The software module includes, but is not limited to, software integrated in the wearable sweat detection device, or software installed on a computer side, or software installed on a mobile side. Of course, the software may be independent program software, or mobile terminal APP software, or applet software based on other software.
Furthermore, the software module preferably adopts mobile terminal APP software or applet software taking mobile terminal equipment as a carrier.
The cloud platform is used for analysis of big data and model construction of the big data.
The sweat signal acquisition principle of the wearable sweat detection system is that the sweat impedance is obtained by injecting a high-frequency oscillation signal into a sweat sensor patch, and the sweat concentration value and the sweat flow velocity value are obtained by a sweat signal analysis module.
The sweat collecting method comprises the steps of sticking one surface of a sweat sensor patch with sweat injection holes on human skin, sticking the other surface of the sensor patch on wearable sweat detection equipment, and enabling sweat discharged by sweat glands of the human body to enter micro-channels in the sweat sensor patch so as to collect sweat signals.
As shown in fig. 2, the invention further provides a detection method of the wearable sweat detection system based on the wearable sweat detection system, which comprises the following steps:
S1, sweat signal acquisition and pretreatment;
s2, sweat signal analysis and treatment:
S3, data storage and data transmission;
S4, data display;
S5, cloud platform trend prediction.
As shown in fig. 3, is the detection principle of the sweat sensor patch. Sweat enters a signal acquisition area of the sweat sensor patch from a cavity of the sweat sensor patch, and the total resistance value output by the sensor patch is suddenly changed along with the continuous increase of the sweat amount, so that a step signal is formed. The height of the step, i.e. the conductance value, is related to the sweat concentration, the step width, the amount of time is related to the sweat rate.
Therefore, as shown in fig. 4, in step S1, the sweat signal acquisition and preprocessing step further includes the following steps:
s11, sweat flow rate signal acquisition;
S12, sweat concentration signal acquisition;
s13, preprocessing sweat alternating current signals to sweat direct current signals.
The sweat flow rate signal acquisition and sweat concentration signal acquisition are switched and controlled by an electrode sampling switching control technology. According to the change characteristics of the signals, in order to reduce the connection points of the sweat detection equipment and the sweat sensor patch, the sweat concentration detection electrode (namely, the fixed electrode in the sweat sensor patch) and the sweat flow detection electrode (namely, the interdigital electrode in the sweat sensor patch) share a pair of electrodes, and the signals are alternately detected by the switching control module according to the time-sharing multiplexing principle, so that the accuracy of the signals is ensured, and the simplicity and convenience of the equipment structure are ensured.
Further, in step S11 and step S12, sweat flow rate signal acquisition is used for acquiring sweat flow rate signals, and sweat concentration signal acquisition is used for acquiring sweat concentration signals. In order to acquire sweat concentration signals and sweat velocity signals, the invention innovatively applies high-frequency alternating-current excitation signals to both ends of an electrode of a sweat sensor patch to obtain impedance (i.e. resistance) of the electrode. Therefore, the signal returned by the sensor patch collected by the wearable sweat detection equipment is an alternating current signal, and a powerful MCU is required for real-time processing of the alternating current signal, so that the requirement on the MCU is reduced.
Furthermore, when sweat sensor patch sampling is switched to an interdigital electrode according to a time division multiplexing principle (namely an electrode sampling switching control technology) to perform sweat flow sampling, the high-frequency output signal carries sweat flow information. As shown in fig. 5, the gradient of the step signal formed by the interdigital electrode is different at different sweat concentrations, the higher the concentration is, the higher the corresponding step gradient is, and the gradient of different concentrations is greatly changed. As shown in fig. 5, the step signals at concentrations of 200mmolL l-1、150mmolL-1、100mmolL-1、50mmolL-1、20mmolL-1 and 10mmolL l-1, respectively, were from top to bottom.
Too high a slope signal would exceed the range set by the hardware, so in order to detect sweat over a wide concentration range, the present invention employs an interdigital electrode shift adaptation technique. Sweat at different flow rates is handled through setting up different detection gear in the hardware of wearable sweat check out test set, then through the control software analog switch chip of equipment switch to have the operation on-line of different magnifications, realize the self-adaptation switching of interdigital electrode gear. The selection and switching method of the gear is as follows:
a, switching to n+1 gear when V (n) > VH (n) and V (n+1) > VL (n+1);
b, switching to n-1 gear when V (n) < VL (n) and V (n-1) > VL (n-1);
Wherein n (n=1, 2,3,..m) is an interdigital electrode gear, VL (n) is a low voltage threshold corresponding to the n gear, VH (n) is a high voltage threshold corresponding to the n gear, and V (n) is a voltage value measured in the n gear.
The invention can detect sweat parameters in a wide range (1-10000) by introducing an electrode automatic gear shifting mechanism. Meanwhile, the time-sharing detection mechanism and the automatic gear shifting mechanism can enable sweat detection equipment to be suitable for sensor patches of various specifications, and therefore the use scene is expanded and the economical efficiency is improved.
Furthermore, when sweat sensor patch sampling is switched to a fixed electrode according to a time division multiplexing principle (namely an electrode sampling switching control technology) to sample sweat concentration, the high-frequency output signal carries sweat concentration information. And then the signal is converted into a direct current characteristic value through an alternating current-direct current module, and the corresponding sweat concentration value can be obtained from the direct current characteristic value of the fixed electrode signal as long as the corresponding relation between the concentration and the direct current characteristic value is established.
As shown in fig. 6, the sweat signal analysis processing steps are shown. The sweat signal analysis processing of the invention further comprises the following steps:
s21, sweat flow rate extraction;
s22, calculating sweat concentration;
S23, sweat loss calculation;
s24, calculating electrolyte loss;
s25, calculating the water shortage degree;
S26, water supplementing calculation.
Wherein, in step S21, sweat passes through the interdigital electrode array in the sweat sensor patch, a step signal is formed, and the width of the step signal is related to the flow rate. The sweat flow rate extraction adopts a step signal edge recognition technology or a mutation point recognition technology. Wherein the recognition algorithm comprises the following steps:
s211, presetting a queue with the length of 2n, and setting a value of identification degree as k;
s212, emptying a queue;
S213, storing the collected data (the DC characteristic value after converting the AC signal into the DC signal) into a queue;
s214, when the queue is full, obtaining an average value m in the queue data;
recording L, wherein L is the number of values of which the queue 0~n part is smaller than (m-k);
recording R, wherein R is the number of values of which the n-2 n parts of the queue are larger than (m+k);
S215, acquiring a step edge signal and time, wherein if L is more than 2n f and R is more than 2n f (f is a matching factor), the step edge signal is a step edge signal, the current time is recorded as Tn, and the step S212 is skipped at the same time, otherwise, the step S213 is skipped;
S216, calculating step signal time T, wherein T=Tn-Tn-1, and T is the nth step edge signal time minus the (n-1) th step edge signal time;
s217, calculating sweat flow rate S, wherein S=V/T, V is the volume between two interdigital electrodes, sweat sensor patches of different types have different interdigital volumes among the interdigital electrodes, and the volume is calculated when the sweat sensor patches are designed;
s218, jumping to step S212.
Further, the invention also provides another sweat stream extraction method, which comprises the following steps:
s211, acquiring a step edge signal,
The calculation of the Teen _ err,
Teen_err=Teen_Vn-Teen_Vn-1;
Wherein Teen_Vn is the corresponding converted nth DC characteristic value of the high-frequency oscillation signal injection of the interdigital electrode, and n is an integer.
It is determined that the number of the cells to be processed,
Teen_err>Teen_K;
Wherein Teen_K is a set value (preset value).
If Teen_err > Teen_K, it indicates that a step edge signal is present at Teen_tn, and the corresponding time point Teen_tn of the current step edge signal is recorded.
S212 the sweat flow rate S is calculated,
S=VOL÷(Teen_tn-Teen_tn-1);
Wherein VOL is the volume between two interdigital electrodes, tentn is the time point corresponding to the step edge signal of tenvn, and tentn-1 is the time point corresponding to the last step edge signal.
As shown in fig. 6, the method of sweat concentration calculation is as follows:
s221 establishes a sweat concentration linear fit equation g,
Wherein f is a unitary linear regression model function, g is a linear fitting equation;
Taking n points for the sweat concentration interval, wherein the i-th point concentration value;
Is in combination withThe direct current characteristic value of the fixed electrode high-frequency signal corresponding to the concentration;
s222, sweat concentration conversion, calculating sweat concentration d,
d=g(v);
The sweat concentration can be obtained by substituting the direct current value v converted from the high frequency signal of the fixed electrode into the g function.
As shown in fig. 6, the sweat loss calculation further includes the steps of:
s231, calculating and obtaining the surface area BSA of the human body,
BSA=0.0061*H+0.0124*W-0.0099;
H is height (unit: cm), W is weight (unit: kg);
BSA is human body surface area (unit: m2).
S232, calculating and obtaining the whole body sweating rate V Whole body,
V Whole body＝a*V Local area+b;
V Whole body is in L/(min cm2);
the parameter a, b depends on the position of the sensor, and V Local area is calculated by multiplying the extracted sweat flow rate by the flow passage sectional area coefficient of the sweat sensor patch, wherein the flow passage sectional area coefficient is calculated when the sensor is designed.
S233, calculating and obtaining the loss amount M sweat liquid of whole body sweat,
U Whole body＝BSA*10000*V Whole body is t, t is time, and the unit is min;
M sweat liquid＝U Whole body*p sweat liquid;
Wherein p sweat liquid＝1g/cm3;p sweat liquid is sweat density.
S234, calculating and obtaining the whole body dehydration percentage W Whole body,
M Loss of＝M sweat liquid;
W Whole body＝(M Loss of/M weight of body)*100%;
Wherein M weight of body is the body weight of the user, and is configured through a software module, is configured through APP software of the mobile terminal, and is communicated with the wearable sweat detection equipment through the wireless communication module, and the configuration data is transmitted to the wearable sweat detection equipment.
As shown in fig. 6, the electrolyte loss amount calculation further includes the steps of:
s241, calculating and obtaining the whole body electrolyte concentration C Whole body,
C Whole body＝a*C Local area+b;
The unit of C Whole body is mmol/L;
wherein the parameters a, b depend on the position of the sensor attachment.
S242 the whole body electrolyte loss rate Vc Whole body is calculated,
Vc Whole body＝C Whole body*V Whole body*BSA*10000。
S243, calculating the loss of the whole body electrolyte Mc Whole body,
Mc Whole body＝Vc Whole body*MNaCl*t;
T is movement time, and the unit is min;
mNaCl molar mass in g/mol=mg/mmol, in one embodiment molar mass MNaCl is set as
58.5g/mol。
As shown in fig. 6, the water shortage degree determination logic is as follows:
when W Whole body is more than or equal to 2% and less than 3%, the software module can display and remind about thirst;
when W Whole body is more than or equal to 3% and less than 4%, the software module displays and reminds that the mind is restless and the appetite is poor;
When W Whole body% or more is less than 4%, the software module displays and reminds that the skin is flushed, the body temperature is raised and the patient is tired;
when W Whole body% or less and less than 8%, the software module displays and reminds of fever, dizziness and weakness and oliguria;
When W Whole body is more than or equal to 8% and less than 10%, the software module can display and remind of twitching and shock;
When W Whole body is more than or equal to 10%, the software module can display and remind that urine is not present and death occurs.
As shown in fig. 6, the water replenishment calculation formula is as follows:
Mw Loss of＝M sweat liquid-M Moisturizing。
As shown in fig. 1, the software module may be integrally installed in the wearable sweat detection device of the present invention, or in a computer, or in a mobile terminal such as a cell phone, tablet, smart watch, etc. In the invention, the mobile terminal is preferably adopted as a carrier of the software APP.
Furthermore, the communication module in the wearable sweat detection device of the invention preferably adopts a low-power consumption wireless Bluetooth communication chip, thereby forming a wireless communication module. The wearable sweat detection equipment sends detected and calculated data to the mobile terminal APP through the wireless communication module, and meanwhile, one data is backed up through the storage module inside the equipment, so that the equipment can be ensured not to lose when the wireless communication is abnormal. The sweat can be detected and analyzed by a user only by carrying the wearable sweat detection device, so that the carrying burden of the user can be greatly reduced, and the long-time data integrity can be ensured.
Further, as shown in fig. 2, the APP software of the mobile terminal further stores the transmitted data locally in a memory chip of the mobile terminal.
Meanwhile, the mobile terminal can also send corresponding configuration information to the wearable sweat detection equipment through the communication module, so that the working parameters of the wearable sweat detection equipment can be flexibly adjusted. If sweat sensor patches with different specifications are selected according to the movement types, the sweat sensor patches can be used only by reconfiguring the wearable sweat detection equipment through the mobile terminal APP, and compared with other sweat detection equipment which can only use sensors with one specification, the sweat sensor patches have the advantage that the convenience and the economy of use are greatly improved.
Further, the wearable sweat detection device transmits sweat data processed by the sweat signal analysis processing module to the software module through the communication module, and the software module performs visual analysis and display on the sweat data, so that curves of sweat concentration, sweat flow rate, human body water loss and electrolyte loss along with time change are displayed in real time, and the water shortage degree judgment logic is combined to remind the movement dehydration degree and the water supplement reminding. Meanwhile, the software module synchronizes the data into the cloud platform through the network, the cloud platform builds a big data model based on the trend to process the collected data, and the processed result (trend prediction) is fed back to the software module for display. Health reminders and advice, sports advice, and the like are made through the software module.
Further, the software module is installed in the smart phone, the smart watch and the tablet personal computer, and the smart phone, the smart watch and the tablet personal computer are used as carriers to perform visual display and display of data.
As shown in fig. 1, the cloud platform adopts a cloud computing platform for analysis of sweat big data and construction of big data models. The cloud platform builds a data model based on the trends, thereby predicting body states, and providing athletic advice.
Further, the trend prediction in the cloud platform comprises the following steps:
S51, calculating and obtaining the current whole body electrolyte estimation amount An,
An=f(Mc Whole body,M sweat liquid);
Wherein An is the current whole body electrolyte estimation quantity, n is a positive integer, and f is the whole body electrolyte estimation function, and the whole body electrolyte concentration is estimated according to the whole body electrolyte loss quantity and the whole body sweat loss quantity.
S52, calculating and obtaining the electrolyte loss rate Bs corresponding to the S motion type,
Bs=ma(Vc Whole body);
Where ma is the smoothing filter and Bs is the electrolyte loss rate corresponding to the s motion type.
S53, calculating and obtaining a whole body electrolysis trend value Cn,
Cn=pettitt(An...A1);
Wherein pettitt is a mutation detection algorithm, and An to A1 are previous data. Cn is a whole body electrolysis trend value calculated by pettitt functions, and pettitt functions are classical trend prediction algorithm functions.
S54 the acquired reference contrast deviation ratio Kn is calculated,
Kn=h(Cn);
Wherein the h function is a standard reference contrast function, and Kn is a reference contrast deviation ratio.
S55, constructing a data model, carrying out trend prediction,
When |kn| <10%, the electrolyte content trend is normal, and the body is healthy;
when 10% </kn| <30%, the electrolyte content trend is slightly abnormal, suggesting reasonable exercise and diet;
When 30% <|kn| <50%, the electrolyte content trend is moderate anomaly, and the body has potential health problem advice for examination;
When 50% <|kn|, the electrolyte content trend is heavy anomaly, and abnormal health advice examination is provided.
S56, calculating and acquiring recommended movement duration Tn,
Tn=g(Bs,Kn);
Wherein g is the reasonable movement duration Tn under the calculation k movement.
As shown in fig. 4, the sweat ac signal to sweat dc signal conversion formula is as follows:
Wherein x is the signal sampling point, and n is the sampling point.
Injecting a high-frequency sinusoidal signal into the sweat sensor patch to obtain an alternating current signal with impedance information, converting the alternating current signal into a direct current characteristic value Xrms through the root mean square formula, inputting the direct current characteristic value into a microprocessor of detection equipment for signal analysis, and obtaining sweat flow rate and sweat concentration value through the algorithm.
Furthermore, the invention also provides another conversion formula for converting sweat alternating current signals into sweat direct current signals:
wherein, ampout is the amplitude of the high-frequency injection sinusoidal signal;
The amplitude value of the input signal acquired for the nth time is set;
The amplitude ratio of the n-th input and the n-th output is set;
is the amplitude ratio after filtering and smoothing.
While the foregoing is directed to embodiments of the present invention, other and further details of the invention may be had by the present invention, it should be understood that the foregoing description is merely illustrative of the present invention and that no limitations are intended to the scope of the invention, except insofar as modifications, equivalents, improvements or modifications are within the spirit and principles of the invention.