Physiological signal monitoring system of contact lens based on embedded intelligent chip

Abstract

The invention discloses a physiological signal monitoring system of a contact lens based on an embedded intelligent chip, which relates to the field of physiological signal monitoring and comprises a contact lens and a monitoring bracelet, a glucose detection module is arranged in the contact lens, and the glucose detection module detects tears of a contact lens wearer and sends detected voltage data to the monitoring bracelet; the monitoring bracelet detects environment data; on the basis of the voltage data and the environment data, the blood glucose value of the current wearer is determined by means of the trained blood glucose recognition model. The glucose concentration in the tear of the wearer is detected through the contact lenses, blood detection is not needed, the user experience is better, and blood glucose detection is more convenient.

Description

Technical Field

[0001] The present invention relates to the technical field of physiological signal monitoring, and more particularly to a physiological signal monitoring system based on a contact lens with an embedded intelligent chip. Background Art

[0002] Diabetes is a chronic disease characterized by high blood sugar, which is caused by absolute or relative insulin deficiency and utilization disorder and seriously affects people's health. Diabetic patients need to strictly monitor and control their blood sugar to stabilize blood sugar and reduce or delay the occurrence of related complications.

[0003] However, existing blood sugar tests usually require piercing the skin and extracting blood samples for measurement. Although it can accurately detect the blood sugar of diabetics, the patient experience is not good, resulting in fewer blood sugar tests in daily life and inability to control their own blood sugar in real time. The blood sugar monitoring of diabetics is not effectively monitored and controlled, which is not conducive to the health of patients.

[0004] Therefore, how to provide a monitoring system with better user experience and more convenient blood glucose detection is an urgent problem that needs to be solved by technical personnel in this field. Summary of the invention

[0005] In view of this, the present invention provides a physiological signal monitoring system based on a contact lens with an embedded smart chip to solve the problems in the background technology.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] The present invention discloses a physiological signal monitoring system for contact lenses based on an embedded smart chip, comprising: a contact lens and a monitoring bracelet;

[0008] The contact lens has a built-in glucose detection module, which detects the tears of the contact lens wearer and sends the detected voltage data to the monitoring bracelet;

[0009] The monitoring bracelet detects environmental data; based on the voltage data and the environmental data, the trained blood glucose recognition model is used to confirm the current wearer's blood glucose value.

[0010] Further, the contact lens comprises: a corneal covering layer, a marginal area, a tear channel and a UV absorbing layer;

[0011] The corneal covering layer is the circular area in the center of the contact lens, which covers the surface of the eye when worn to perform refractive correction; the UV absorbing layer is located on the surface of the corneal covering layer away from the eye, and is used to absorb ultraviolet rays;

[0012] The edge area is an annular area adjacent to the corneal covering layer, and the glucose detection module is built in it; the detection electrode of the glucose detection module is located in the tear channel; the tear channel is located at the edge of the corneal covering layer, passes through the inner surface and the outer surface of the corneal covering layer, and tears enter the tear channel under capillary action.

[0013] Furthermore, when making contact lenses, polymethyl methacrylate is first poured onto the glucose detection module to obtain the edge area; silicone hydrogel is then poured into the edge area to obtain the corneal covering layer, and during pouring, the mold has protrusions at positions corresponding to the detection electrodes, and the tear channel on the corneal covering layer is obtained after pouring; finally, the UV absorbing layer is coated on the corneal covering layer, and the material of the UV absorbing layer is ethyl benzoylacrylate.

[0014] Further, the glucose detection module includes: a glucose detection unit, a control transmission unit, and a power supply unit;

[0015] The glucose detection unit includes a detection chip, a detection electrode, and a reference electrode; the reference electrode is located on the outer surface of the edge region; the surface of the detection electrode is coated with glucose oxidase, and when the tear contains glucose and enters the tear channel, the glucose in the tear is oxidized and decomposed under the action of the glucose oxidase and oxygen to generate hydroxide ions and electrons, and the detection electrode has a potential difference with the reference electrode under the action of the electrons; the detection chip detects the voltage signal between the detection electrode and the reference electrode, and transmits it to the control transmission unit;

[0016] The control transmission unit samples the voltage signal and obtains the voltage data, and sends the data to the monitoring bracelet through the antenna;

[0017] The power supply unit supplies power to the glucose detection unit and the control transmission unit respectively.

[0018] Furthermore, the power supply unit includes an induction coil, an energy storage capacitor and a power supply control chip;

[0019] The monitoring bracelet is provided with a charging coil corresponding to the induction coil. When the monitoring bracelet is close to the contact lens, the power supply control chip is powered by wireless induction; the power supply control chip stores electric energy through the energy storage capacitor, and uses the electric energy stored in the energy storage capacitor to continuously power the glucose detection unit and the control transmission unit;

[0020] The two electrodes of the energy storage capacitor have the same structure, both of which are obtained by alternately connecting multiple groups of U-shaped structures in series, and the overall structure is a ring;

[0021] The centers of the annular structure, the induction coil, the annular area, and the corneal covering layer overlap.

[0022] Furthermore, the monitoring bracelet includes: a computing control module, a storage module, a wireless communication module, a touch screen, a temperature detection module, a humidity detection module, an air pressure detection module and a power supply module;

[0023] The detection probe of the temperature detection module is located on the side of the monitoring bracelet away from the wearer's skin, detects the air temperature, and sends it to the storage module for storage; the humidity detection module and the air pressure detection module detect the air humidity and air pressure respectively, and send them to the storage module for storage; the environmental data are air temperature, humidity and air pressure data;

[0024] The wireless communication module receives the voltage data and stores it in the storage module;

[0025] The calculation control module identifies the blood glucose value at each moment according to the air temperature, humidity and air pressure data and the voltage data using the trained blood glucose recognition model and stores the results in the storage module; the calculation control module selects the blood glucose value at the corresponding moment for display according to the instruction of the touch screen, and controls the power supply module to charge the contact lens through the charging coil according to the instruction of the touch screen;

[0026] The power supply module respectively supplies power to the calculation control module, the storage module, the wireless communication module, the touch screen, the temperature detection module, the humidity detection module and the air pressure detection module.

[0027] Furthermore, the blood sugar recognition model is a machine learning model, which is trained in the following way:

[0028] Step 1, simulating different temperature, humidity, and air pressure conditions, and using tears with different blood sugar values ​​to conduct simulation experiments to obtain corresponding contact lens voltage data;

[0029] Step 2: Using temperature, humidity, air pressure, and voltage as sample features and blood glucose as sample labels, the sample set is sorted and divided into a training set and a test set.

[0030] Step 3, establishing a machine learning model and using the training set for training;

[0031] Step 4, use the test set to verify the trained machine learning model, and obtain the final blood glucose recognition model after reaching the preset accuracy, otherwise return to step 3.

[0032] Furthermore, the training set is expressed as:

[0033] S={(xi ,y i ),x i ∈R N},i=1,2,…,n;

[0034] Among them, x i is the input feature vector, y i is the sample label, R N represents the feature domain, N is the feature dimension, and n is the total number of samples;

[0035] The recognition function of the machine learning model is expressed as:

[0036]

[0037] Among them, f(x) represents the recognition function, x represents the feature vector of the input sample, ω represents the weight vector, represents the kernel function, which maps the input sample to the high-dimensional feature space for linear regression solution, and b represents the variation;

[0038] The optimization objective function of the machine learning model is expressed as:

[0039]

[0040] Among them, minG(ω,b,e) represents the objective function, γ represents the penalty factor, and e i represents the fitting error;

[0041] A Lagrangian function is constructed, and a linear regression solution is performed on the optimization objective function according to the KKT condition to obtain the optimal solution of ω and b.

[0042] Furthermore, the regression function for solving the optimization objective function by linear regression is:

[0043]

[0044] Where f′(x) represents the regression function, λ i represents the Lagrange multiplier; K(x i ,x j ) represents the radial basis kernel function, and x j To x i The formula for the feature vector after mapping to the high-dimensional feature space is:

[0045]

[0046] Where σ represents the normalization parameter of the radial basis kernel function.

[0047] It can be known from the above technical solutions that, compared with the prior art, the present invention discloses a physiological signal monitoring system for contact lenses based on embedded smart chips. Among them, the contact lenses are designed with special edge areas and tear channels, which can easily collect tear samples and introduce them into the glucose detection module for detection; by integrating the glucose detection module on the contact lenses, the glucose content in the wearer's tears can be detected, and then by monitoring the environmental data on the bracelet and the trained machine learning model, the wearer's current blood sugar value can be accurately calculated; the monitoring bracelet integrates a variety of environmental sensors such as temperature, humidity, and air pressure, which can comprehensively collect environmental data, provide more accurate input features for the blood sugar recognition model, and realize continuous and non-invasive detection of the wearer's blood sugar value. The present invention detects the glucose concentration in the wearer's tears through contact lenses, without the need for blood testing, so the user experience is better and blood sugar testing is more convenient. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0049] Figure 1 Schematic diagram of the structure of a contact lens according to an embodiment of the present invention.

[0050] In the figure, 1. edge area; 2. corneal covering layer; 3. glucose detection module; 31. control transmission unit; 32. power supply unit; 33. glucose detection unit; 34. reference electrode; 35. detection electrode; 36. induction coil; 37. energy storage capacitor. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] The embodiment of the present invention discloses a physiological signal monitoring system based on a contact lens with an embedded smart chip, comprising: a contact lens and a monitoring bracelet;

[0053] The contact lens has a built-in glucose detection module 3, which detects the tears of the contact lens wearer and sends the detected voltage data to the monitoring bracelet;

[0054] The monitoring bracelet detects environmental data; based on voltage data and environmental data, the trained blood glucose recognition model is used to confirm the current wearer's blood glucose value.

[0055] In a specific embodiment, Figure 1 As shown, the contact lens includes: a corneal covering layer 2 (blank area in the figure), a peripheral area 1 (blank area and yellow area in the figure), a tear channel and a UV absorbing layer;

[0056] The corneal covering layer 2 is the circular area in the center of the contact lens, which covers the surface of the eye when worn to perform refractive correction; the UV absorption layer is located on the surface of the corneal covering layer 2 away from the eye, and is used to absorb ultraviolet rays;

[0057] The edge area 1 is an annular area adjacent to the corneal covering layer 2, and has a built-in glucose detection module 3; the detection electrode 35 of the glucose detection module 3 is located in the tear channel; the tear channel is located at the edge of the corneal covering layer 2, passes through the inner and outer surfaces of the corneal covering layer 2, and tears enter the tear channel under capillary action.

[0058] In a specific embodiment, when making contact lenses, polymethyl methacrylate is first poured onto the glucose detection module 3 to obtain the edge region 1; polymethyl methacrylate is a hard, insulating, hydrophobic material that can effectively ensure the insulation between the electrical components in the glucose detection module 3, and only the reference electrode 34 and the detection electrode 35 in the glucose detection module 3 are not insulated from the outside. Then, silicone hydrogel is poured into the edge region 1 to obtain the corneal covering layer 2, and during the pouring, the mold has a protrusion at the corresponding position of the detection electrode 35, and after pouring, the tear channel on the corneal covering layer 2 is obtained. Finally, a UV absorption layer is coated on the corneal covering layer 2, and the material of the UV absorption layer is ethyl benzoylacrylate.

[0059] In a specific embodiment, the glucose detection module 3 includes: a glucose detection unit 33, a control transmission unit 31, and a power supply unit 32;

[0060] The glucose detection unit 33 includes a detection chip, a detection electrode 35, and a reference electrode 34; the reference electrode 34 is located on the outer surface of the edge region 1, and contacts the eyelid when the wearer blinks. It can be considered that there is a short grounding, ensuring that the potential of the reference electrode 34 is 0; and because the edge region 1 is hydrophobic, there is generally no tear on the surface of the edge region 1, avoiding the problem of forming a loop between the detection electrode 35 and the reference electrode 34 through the tear, which affects the detection accuracy. The surface of the detection electrode 35 is coated with glucose oxidase. When the tear contains glucose and enters the tear channel, the glucose in the tear is oxidized and decomposed under the action of glucose oxidase and oxygen to produce hydroxide ions and electrons. The detection electrode 35 has a potential difference with the reference electrode 34 under the action of electrons; wherein, the detection electrode 35 is foamed silver, which can effectively increase the contact area between the electrode and the tear and air, and increase the reaction rate of glucose oxidation and decomposition, thereby increasing the intensity of the point signal; the detection chip detects the voltage signal between the detection electrode 35 and the reference electrode 34, and transmits it to the control transmission unit 31.

[0061] After the control transmission unit 31 samples the voltage signal, it obtains the voltage data and sends it to the monitoring bracelet through the antenna;

[0062] The power supply unit 32 supplies power to the glucose detection unit 33 and the control transmission unit 31 respectively.

[0063] In a specific embodiment, the power supply unit 32 includes an induction coil 36, an energy storage capacitor 37 and a power supply control chip;

[0064] The monitoring bracelet is provided with a charging coil corresponding to the induction coil 36. When the monitoring bracelet is close to the contact lens, the power supply control chip is powered by wireless induction; the power supply control chip stores electric energy through the energy storage capacitor 37, and uses the electric energy stored in the energy storage capacitor 37 to continuously power the glucose detection unit 33 and the control transmission unit 31. The power supply system built into the contact lens obtains electric energy from the monitoring bracelet by wireless induction, and can continuously provide power support for glucose detection and data transmission. When in use, the wearer charges the contact lens through the monitoring bracelet, and the glucose detection module 3 automatically starts detection and data transmission, which is simple and convenient; and the capacitor power supply has a long service life and stability, avoiding the risks of built-in chemical batteries.

[0065] The two electrodes of the energy storage capacitor 37 have the same structure, both of which are obtained by alternately connecting multiple groups of U-shaped structures in series, and the overall structure is a ring structure;

[0066] The centers of the annular structure, the induction coil 36 , the annular area, and the corneal covering layer 2 overlap.

[0067] In a specific embodiment, the monitoring bracelet includes: a computing control module, a storage module, a wireless communication module, a touch screen, a temperature detection module, a humidity detection module, an air pressure detection module and a power supply module;

[0068] The detection probe of the temperature detection module is located on the side of the monitoring bracelet away from the wearer's skin, detects the air temperature, and sends it to the storage module for storage; the humidity detection module and the air pressure detection module detect the air humidity and air pressure respectively, and send them to the storage module for storage; the environmental data are air temperature, humidity and air pressure data;

[0069] The wireless communication module receives the voltage data and stores it in the storage module;

[0070] The calculation control module uses the trained blood sugar recognition model to identify the blood sugar value at each moment according to the air temperature, humidity and air pressure data, as well as the voltage data, and stores it in the storage module; the calculation control module selects the blood sugar value at the corresponding moment for display according to the instruction of the touch screen, and controls the power supply module to charge the contact lens through the charging coil according to the instruction of the touch screen;

[0071] The power supply modules respectively supply power to the computing control module, the storage module, the wireless communication module, the touch screen, the temperature detection module, the humidity detection module and the air pressure detection module.

[0072] In a specific embodiment, the blood glucose recognition model is a machine learning model, which is trained in the following way:

[0073] Step 1, simulating different temperature, humidity, and air pressure conditions, and using tears with different blood sugar values ​​to conduct simulation experiments to obtain corresponding contact lens voltage data;

[0074] Step 2: Using temperature, humidity, air pressure, and voltage as sample features and blood glucose as sample labels, the sample set is sorted and divided into a training set and a test set.

[0075] Step 3: Build a machine learning model and train it using the training set;

[0076] Step 4: Use the test set to verify the trained machine learning model. When the preset accuracy is reached, the final blood glucose recognition model is obtained, otherwise return to step 3.

[0077] Specifically, since blood sugar detection is based on the voltage generated by the glucose oxidation reaction, the reaction voltage will not only be affected by the glucose content in the tears, but also by environmental factors. For example, the ambient temperature and humidity will affect the evaporation of tears and thus affect the glucose concentration; the reaction rate will be affected by the reaction temperature. Assuming that the human body temperature remains unchanged, the actual temperature at the detection electrode 35 affected by the ambient temperature must also be considered; and the reaction rate is also affected by the oxygen content involved in the reaction. It is generally believed that the oxygen content in the air is mainly affected by air pressure. By comprehensively considering the effects of temperature, humidity, and air pressure on the reaction, and combining the potential difference between the detection electrode 35 and the reference electrode 34, a machine learning model is constructed to identify the glucose content in tears. Compared with simply determining the glucose content based on the potential difference between the two electrodes, it has higher accuracy.

[0078] In a specific embodiment, the training set is represented as:

[0079] S={(x i ,y i ),x i ∈R N},i=1,2,…,n;

[0080] Among them, x i is the input feature vector, y i is the sample label, R N represents the feature domain, N is the feature dimension, and n is the total number of samples;

[0081] The recognition function of the machine learning model is expressed as:

[0082]

[0083] Among them, f(x) represents the recognition function, x represents the feature vector of the input sample, ω represents the weight vector, represents the kernel function, which is used to map the input samples to the high-dimensional feature space for linear regression solution, and b represents the variation;

[0084] The optimization objective function of the machine learning model is expressed as:

[0085]

[0086] Among them, minG(ω,b,e) represents the objective function, γ represents the penalty factor, and e i represents the fitting error;

[0087] Construct the Lagrangian function, and perform linear regression to solve the optimization objective function according to the KKT condition to obtain the optimal solution of ω and b. The process of solving ω and b based on the optimization objective function using training samples is the training process.

[0088] Specifically, for the optimization problem of the constraints of the optimization objective function, the Lagrangian function is constructed to solve it and the Lagrangian multiplier λ is introduced. i , then:

[0089]

[0090] According to the KTT condition, the partial derivative of the above formula is optimized, the weight vector ω and the error vector b are eliminated, and the quadratic programming problem is transformed into a dual problem using the optimization theory, and the optimal solution of ω and b is obtained. The regression function for linear regression solution of the optimization objective function is:

[0091]

[0092] Where f′(x) represents the regression function, λ i represents the Lagrange multiplier; K(x i ,x j ) represents the radial basis kernel function, and x j To x i The formula for the feature vector after mapping to the high-dimensional feature space is:

[0093]

[0094] Where σ represents the normalization parameter of the radial basis kernel function.

[0095] In a specific embodiment, the normalization parameter σ and the penalty factor γ are optimized using a chaotic optimization algorithm.

[0096] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0097] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A physiological signal monitoring system for contact lenses based on an embedded smart chip, characterized in that: include: contact lenses and monitoring bracelets; The contact lens has a built-in glucose detection module, which detects the tears of the contact lens wearer and sends the detected voltage data to the monitoring bracelet; The monitoring bracelet detects environmental data; based on the voltage data and the environmental data, the trained blood glucose recognition model is used to confirm the current wearer's blood glucose value.

2. A physiological signal monitoring system based on contact lenses with embedded smart chips according to claim 1, characterized in that: The contact lens comprises: a corneal covering layer, a marginal area, a tear channel and a UV absorbing layer; The corneal covering layer is the circular area in the center of the contact lens, which covers the surface of the eye when worn to perform refractive correction; the UV absorbing layer is located on the surface of the corneal covering layer away from the eye, and is used to absorb ultraviolet rays; The edge area is an annular area adjacent to the corneal covering layer, and the glucose detection module is built in it; the detection electrode of the glucose detection module is located in the tear channel; the tear channel is located at the edge of the corneal covering layer, passes through the inner surface and the outer surface of the corneal covering layer, and tears enter the tear channel under capillary action.

3. The physiological signal monitoring system based on contact lenses with embedded smart chips according to claim 2, characterized in that: When making contact lenses, polymethyl methacrylate is first poured onto the glucose detection module to obtain the edge area; silicone hydrogel is then poured into the edge area to obtain the corneal covering layer, and during pouring, the mold has protrusions at positions corresponding to the detection electrodes, and the tear channel on the corneal covering layer is obtained after pouring; finally, the UV absorbing layer is coated on the corneal covering layer, and the material of the UV absorbing layer is ethyl benzoylacrylate.

4. The physiological signal monitoring system based on contact lenses with embedded smart chips according to claim 2, characterized in that: The glucose detection module includes: a glucose detection unit, a control transmission unit, and a power supply unit; The glucose detection unit includes a detection chip, a detection electrode, and a reference electrode; the reference electrode is located on the outer surface of the edge region; the surface of the detection electrode is coated with glucose oxidase, and when the tear contains glucose and enters the tear channel, the glucose in the tear is oxidized and decomposed under the action of the glucose oxidase and oxygen to generate hydroxide ions and electrons, and the detection electrode has a potential difference with the reference electrode under the action of the electrons; the detection chip detects the voltage signal between the detection electrode and the reference electrode, and transmits it to the control transmission unit; After the control transmission unit samples the voltage signal, the voltage data is obtained and sent to the monitoring bracelet via an antenna; The power supply unit supplies power to the glucose detection unit and the control transmission unit respectively.

5. The physiological signal monitoring system based on contact lenses with embedded smart chips according to claim 4, characterized in that: The power supply unit includes an induction coil, an energy storage capacitor and a power supply control chip; The monitoring bracelet is provided with a charging coil corresponding to the induction coil. When the monitoring bracelet is close to the contact lens, the power supply control chip is powered by wireless induction; the power supply control chip stores electric energy through the energy storage capacitor, and uses the electric energy stored in the energy storage capacitor to continuously power the glucose detection unit and the control transmission unit; The two electrodes of the energy storage capacitor have the same structure, both of which are obtained by alternately connecting multiple groups of U-shaped structures in series, and the overall structure is a ring; The centers of the annular structure, the induction coil, the annular area, and the corneal covering layer overlap.

6. The physiological signal monitoring system based on contact lenses with embedded smart chips according to claim 1, characterized in that: The monitoring bracelet includes: a computing control module, a storage module, a wireless communication module, a touch screen, a temperature detection module, a humidity detection module, an air pressure detection module and a power supply module; The detection probe of the temperature detection module is located on the side of the monitoring bracelet away from the wearer's skin, detects the air temperature, and sends it to the storage module for storage; the humidity detection module and the air pressure detection module detect the air humidity and air pressure respectively, and send them to the storage module for storage; the environmental data are air temperature, humidity and air pressure data; The wireless communication module receives the voltage data and stores it in the storage module; The calculation control module identifies the blood glucose value at each moment according to the air temperature, humidity and air pressure data and the voltage data using the trained blood glucose recognition model and stores the results in the storage module; the calculation control module selects the blood glucose value at the corresponding moment for display according to the instruction of the touch screen, and controls the power supply module to charge the contact lens through the charging coil according to the instruction of the touch screen; The power supply module respectively supplies power to the calculation control module, the storage module, the wireless communication module, the touch screen, the temperature detection module, the humidity detection module and the air pressure detection module.

7. The physiological signal monitoring system based on contact lenses with embedded smart chips according to claim 1, characterized in that: The blood sugar recognition model is a machine learning model, which is trained in the following way: Step 1, simulating different temperature, humidity, and air pressure conditions, and using tears with different blood sugar values ​​to conduct simulation experiments to obtain corresponding contact lens voltage data; Step 2: Using temperature, humidity, air pressure, and voltage as sample features and blood glucose as sample labels, the sample set is sorted and divided into a training set and a test set. Step 3, establishing a machine learning model and using the training set for training; Step 4, use the test set to verify the trained machine learning model, and obtain the final blood glucose recognition model after reaching the preset accuracy, otherwise return to step 3.

8. The physiological signal monitoring system based on contact lenses with embedded smart chips according to claim 7, characterized in that: The training set is represented as: S={(x i ,y i ),x i ∈R N },i=1,2,…,n; Among them, x i is the input feature vector, y i is the sample label, R N represents the feature domain, N is the feature dimension, and n is the total number of samples; The recognition function of the machine learning model is expressed as: Among them, f(x) represents the recognition function, x represents the feature vector of the input sample, ω represents the weight vector, represents the kernel function, which maps the input sample to the high-dimensional feature space for linear regression solution, and b represents the variation; The optimization objective function of the machine learning model is expressed as: Among them, minG(ω,b,e) represents the objective function, γ represents the penalty factor, and e i represents the fitting error; A Lagrangian function is constructed, and a linear regression solution is performed on the optimization objective function according to the KKT condition to obtain the optimal solution of ω and b.

9. According to the physiological signal monitoring system of contact lenses with embedded smart chips in claim 8, the regression function for solving the optimization objective function by linear regression is: in, f′(x) represents the regression function, λ i represents the Lagrange multiplier; K(x i ,x j ) represents the radial basis kernel function, and x j To x i The formula for the feature vector after mapping to the high-dimensional feature space is: Where σ represents the normalization parameter of the radial basis kernel function.

Images