阅读说明：本技术 基于鸿蒙操作系统的臂戴式血糖仪 (Arm-wearing type blood glucose meter based on hongmeng operating system ) 是由 卢璐 刘鹤宁 于 2021-08-19 设计创作，主要内容包括：本发明提供一种基于鸿蒙操作系统的臂戴式血糖仪,属于人工智能技术领域。该血糖仪包括：基于鸿蒙操作系统的智能终端,与智能终端通信连接的数据采集模块；数据采集模块包括壳体、数据采集单元；壳体包括两个弧形部件和至少两个弹性部件,两个弧形部件的两端分别通过弹性部件连接；数据采集单元包括光源模组和光检测模组,光源模组用于发射照向检测对象手臂上指定位置的光线,光检测模组用于检测指定位置的反射光信息；智能终端获取数据采集模块输出的数据并输入训练好的人工智能模型,根据人工智能模型的输出确定检测对象的血糖值或血糖值范围,数据采集模块输出的数据包括反射光信息。本发明能够实现快速、便捷、准确且无损的血糖检测。(The invention provides an arm-worn blood glucose meter based on a Hongmon operating system, and belongs to the technical field of artificial intelligence. This blood glucose meter includes: the system comprises an intelligent terminal based on a Hongmon operating system and a data acquisition module in communication connection with the intelligent terminal; the data acquisition module comprises a shell and a data acquisition unit; the shell comprises two arc-shaped parts and at least two elastic parts, and two ends of the two arc-shaped parts are respectively connected through the elastic parts; the data acquisition unit comprises a light source module and a light detection module, the light source module is used for emitting light rays irradiating the designated position on the arm of the detection object, and the light detection module is used for detecting reflected light information of the designated position; the intelligent terminal obtains the data output by the data acquisition module and inputs the data into the trained artificial intelligence model, and the blood sugar value or the blood sugar value range of the detection object is determined according to the output of the artificial intelligence model, wherein the data output by the data acquisition module comprises reflected light information. The invention can realize rapid, convenient, accurate and nondestructive blood sugar detection.)

1. An arm-worn blood glucose meter based on the hongmeng operating system, comprising: the system comprises a data acquisition module and an intelligent terminal based on a Hongmon operating system, wherein the data acquisition module is in communication connection with the intelligent terminal;

the data acquisition module comprises a shell and a data acquisition unit; the shell comprises two arc-shaped parts and at least two elastic parts, and two ends of the two arc-shaped parts are respectively connected through the elastic parts; the data acquisition unit comprises a light source module and a light detection module, the light source module is used for emitting light rays irradiating to a designated position on the arm of the detection object, and the light detection module is used for detecting reflected light information of the designated position;

the intelligent terminal acquires the data output by the data acquisition module and inputs the data into a trained artificial intelligence model, and determines the blood sugar value or the blood sugar value range of the detection object according to the output of the artificial intelligence model, wherein the data output by the data acquisition module comprises the reflected light information.

2. The blood glucose meter of claim 1, wherein the light source module comprises parallel light diodes, and the parallel light diodes are uniformly disposed around the light detection module;

the parallel light diode comprises a tube core and a packaging body, wherein the packaging body is coated outside the tube core;

the package body comprises a light-proof part arranged around the side face of the tube core, the top of the light-proof part covers the edge part of the top of the tube core, the package body also comprises a light-absorbing part arranged at the bottom side of the tube core and a spherical collimating lens arranged at the top side of the tube core;

the die is obtained by forming an epitaxial layer on a substrate and patterning the epitaxial layer, and the die has a vertical side or a side with the top inclined to the center.

3. The blood glucose meter of claim 1, wherein the artificial intelligence model comprises: n first neural network models for respectively extracting the characteristics of the reflected light information, wherein N is a positive integer greater than 1; the N first neural network models comprise a neural network model used for extracting the features of the wave number dimension in the reflected light information and a neural network model used for extracting the features of the frequency domain dimension in the reflected light information;

the pyramid model is used for respectively integrating the first feature information of multiple layers extracted by the first neural network to obtain N pieces of second feature information;

a classifier model for outputting a probability that a blood glucose level of the detection target falls within each blood glucose level range, based on the N pieces of second feature information; the classifier model includes a plurality of classifiers that are integrated by a weighted summation.

4. The glucometer according to claim 3, wherein the intelligent terminal is further configured to preprocess the data output by the data acquisition module, the preprocessing includes performing polynomial least squares fitting on the data within the moving window by establishing a filter function using least squares fitting coefficients, and the expression of the polynomial fitting is:

z^*(i)＝a₀+a₁i+a₂i²+…+a_bi^b

wherein z is^*(i) A fitting value a obtained by the position of a central point after a multi-time fitting curve is established for the Savitzky-Golay convolution smoothing method₀，a₁，a₂...a_bIs calculated by the following formula:

wherein, the data in the moving window is z (i), i ═ M, …,0, …, M, μ ═ 0,1,2 …, b.

5. The blood glucose meter of claim 3, wherein the classifier model is constructed by:

obtaining l training samples: { (x)₁，y₁)，(x₂，y₂)，…，(x_l，y_l)}，x_jFor N of said second characteristic information, y_jInitializing a weight value for the blood sugar range with the total number of the blood sugar ranges of K

Let T be 1,2, …, T be the maximum training times;

according to the weight value w^tSelecting a training sample;

carrying out classification and identification on the blood sugar value range of the training sample;

and (3) making K equal to 1,2, … and K, circularly calculating the weight sum of samples in each blood sugar range:judging whether the weight sum of the samples classified correctly in each blood sugar value range is larger than the weight sum of the samples classified into other blood sugar value rangesIf so, performing next circulation, otherwise, turning to a step of performing classification identification on the blood sugar value range of the sample and restarting calculation;

calculate h_tFalse error rate of (2):

order to

Calculate new weight vector:

normalizationObtaining the classifier model as follows:

6. the blood glucose meter of claim 3 or 5, wherein the training samples of the artificial intelligence model comprise: and reflected light information of the designated position of the detection object acquired by the data acquisition module and a blood glucose value range to which the blood glucose value of the detection object acquired by a blood sampling detection mode belongs when the reflected light information is acquired are utilized.

7. The blood glucose meter according to claim 1, wherein the smart terminal is further configured to receive a first blood glucose value range input by a user, the first blood glucose value range being a blood glucose value range to which a blood glucose value of the test object obtained by a blood sampling test method when the blood glucose meter is used belongs;

the intelligent terminal is further used for carrying out transfer learning training on the artificial intelligence model based on the first blood sugar value range.

8. The blood glucose meter of claim 7, wherein the intelligent terminal is configured to obtain the data output by the data collection module when the blood glucose meter is used, a first probability value that the blood glucose value of the test object output by the artificial intelligence model belongs to the first blood glucose value range, and a second probability value that the blood glucose value belongs to a second blood glucose value range, and the second blood glucose value range is the blood glucose value range output when the blood glucose meter is used;

the intelligent terminal is further used for determining a parameter adjustment initial step according to a difference value between the first probability value and the second probability value;

the intelligent terminal is further used for determining a weight value corresponding to parameter adjustment in each neural network model according to the influence degree of each neural network model in the artificial intelligence model on an output result, and determining a parameter adjustment step length of each neural network model according to the weight value and the initial step length;

the intelligent terminal is further used for adjusting parameters in the corresponding neural network model according to the parameter adjustment step length to obtain an adjusted artificial intelligence model, inputting data output by the data acquisition module when the glucometer is used into the adjusted artificial intelligence model to obtain probability values of blood glucose values of a detected object belonging to all blood glucose value ranges, judging whether conditions for determining the first blood glucose value range as the blood glucose value range of the detected object are met, and if not, continuing to adjust the parameters in the corresponding neural network model according to the parameter adjustment step length until the probability values of the blood glucose values output by the adjusted artificial intelligence model can determine that the first blood glucose value range is the blood glucose value range of the detected object.

9. The blood glucose meter of claim 1, wherein the data acquisition unit further comprises at least one of: a temperature sensor, a skin component detection sensor;

the data output by the data acquisition module further comprises data output by the temperature sensor and/or the skin composition detection sensor.

10. The blood glucose meter of claim 1, wherein the data acquisition module further comprises a pressure sensor disposed adjacent to the data acquisition unit, and the intelligent terminal determines whether the pressure between the data acquisition unit and the arm of the detection object satisfies a preset condition according to the pressure information acquired by the pressure sensor, and outputs a prompt message when the pressure does not satisfy the preset condition.

Technical Field

The invention relates to the technical field of artificial intelligence, in particular to an arm-worn glucometer based on a Hongmon operating system.

Background

Diabetes is a syndrome of metabolic disorders of a human body caused by hypofunction of pancreatic islets of langerhans and insulin resistance due to the action of factors such as heredity and environment on the body, and is clinically characterized by hyperglycemia. The long-standing hyperglycemia causes chronic damage to various tissues and organs of a human body, particularly eyes, kidneys, hearts, blood vessels, nerves and the like, so that functional disorder is gradually generated, and the health of the human body is seriously threatened. For diabetic patients, frequent monitoring of blood glucose is very important, since it can help patients to control blood glucose levels and prevent complications. However, the conventional blood sugar detection method is still a venous blood drawing detection method or a fingertip blood detection method, and the method brings certain pain to patients, so that the change of blood sugar cannot be monitored at any time, and the infection chance is increased.

Disclosure of Invention

Therefore, the technical problem to be solved by the embodiment of the invention is to overcome the defect that a diabetic patient cannot monitor the blood sugar change in real time and further cannot control the blood sugar well due to invasive blood sugar detection in the prior art, so that the arm-worn blood glucose meter based on the Hongmon operating system is provided.

Therefore, the invention provides an arm-worn blood glucose meter based on Hongmon operating system, which comprises: the system comprises a data acquisition module and an intelligent terminal based on a Hongmon operating system, wherein the data acquisition module is in communication connection with the intelligent terminal;

the data acquisition module comprises a shell and a data acquisition unit; the shell comprises two arc-shaped parts and at least two elastic parts, and two ends of the two arc-shaped parts are respectively connected through the elastic parts; the data acquisition unit comprises a light source module and a light detection module, the light source module is used for emitting light rays irradiating to a designated position on the arm of the detection object, and the light detection module is used for detecting reflected light information of the designated position;

the intelligent terminal acquires the data output by the data acquisition module and inputs the data into a trained artificial intelligence model, and determines the blood sugar value or the blood sugar value range of the detection object according to the output of the artificial intelligence model, wherein the data output by the data acquisition module comprises the reflected light information.

Optionally, the light source module comprises parallel light diodes, and the parallel light diodes are uniformly arranged around the light detection module;

the parallel light diode comprises a tube core and a packaging body, wherein the packaging body is coated outside the tube core;

the package body comprises a light-proof part arranged around the side face of the tube core, the top of the light-proof part covers the edge part of the top of the tube core, the package body also comprises a light-absorbing part arranged at the bottom side of the tube core and a spherical collimating lens arranged at the top side of the tube core;

the die is obtained by forming an epitaxial layer on a substrate and patterning the epitaxial layer, and the die has a vertical side or a side with the top inclined to the center.

Optionally, the artificial intelligence model includes: n first neural network models for respectively extracting the characteristics of the reflected light information, wherein N is a positive integer greater than 1; the N first neural network models comprise a neural network model used for extracting the features of the wave number dimension in the reflected light information and a neural network model used for extracting the features of the frequency domain dimension in the reflected light information;

the pyramid model is used for respectively integrating the first feature information of multiple layers extracted by the first neural network to obtain N pieces of second feature information;

a classifier model for outputting a probability that a blood glucose level of the detection target falls within each blood glucose level range, based on the N pieces of second feature information; the classifier model includes a plurality of classifiers that are integrated by a weighted summation.

Optionally, the intelligent terminal is further configured to preprocess the data output by the data acquisition module, where the preprocessing includes establishing a filter function by using a least square fitting coefficient, and performing polynomial least square fitting on the data in the moving window, where an expression of the polynomial fitting is:

z^*(i)＝a₀+a₁i+a₂i²+...+a_bi^b

wherein z is^*(i) A fitting value a obtained by the position of a central point after a multi-time fitting curve is established for the Savitzky-Golay convolution smoothing method₀，a₁，a₂…a_bIs calculated by the following formula:

wherein, the data in the moving window is z (i), i ═ M, …,0, …, M, μ ═ 0,1,2 …, b.

Optionally, the classifier model is constructed in the following manner:

obtaining l training samples: { (x)₁,y₁),(x₂,y₂),…,(x_l,y_l)}，x_jFor N of said second characteristic information, y_jInitializing a weight value for the blood sugar range with the total number of the blood sugar ranges of K

Let T be 1,2, …, T be the maximum training times;

according to the weight value w^tSelecting a training sample;

carrying out classification and identification on the blood sugar value range of the training sample;

and (3) making K equal to 1,2, … and K, circularly calculating the weight sum of samples in each blood sugar range:judging whether the weight sum of the samples classified correctly in each blood sugar value range is larger than the weight sum of the samples classified into other blood sugar value rangesIf so, performing next circulation, otherwise, turning to a step of performing classification identification on the blood sugar value range of the sample and restarting calculation;

calculate h_tFalse error rate of (2):

order to

Calculate new weight vector:

normalizationObtaining the classifier model as follows:

optionally, the training samples of the artificial intelligence model include: and reflected light information of the designated position of the detection object acquired by the data acquisition module and a blood glucose value range to which the blood glucose value of the detection object acquired by a blood sampling detection mode belongs when the reflected light information is acquired are utilized.

Optionally, the intelligent terminal is further configured to receive a first blood glucose value range input by a user, where the first blood glucose value range is a blood glucose value range to which a blood glucose value of the detection object obtained by a blood sampling detection method when the blood glucose meter is used belongs;

the intelligent terminal is further used for carrying out transfer learning training on the artificial intelligence model based on the first blood sugar value range.

Optionally, the intelligent terminal is configured to obtain data output by the data acquisition module when the blood glucose meter is used, a first probability value that a blood glucose value of the detection object output by the artificial intelligence model belongs to the first blood glucose value range, and a second probability value that the blood glucose value belongs to a second blood glucose value range, where the second blood glucose value range is the blood glucose value range output when the blood glucose meter is used;

the intelligent terminal is further used for determining a parameter adjustment initial step according to a difference value between the first probability value and the second probability value;

the intelligent terminal is further used for determining a weight value corresponding to parameter adjustment in each neural network model according to the influence degree of each neural network model in the artificial intelligence model on an output result, and determining a parameter adjustment step length of each neural network model according to the weight value and the initial step length;

the intelligent terminal is further used for adjusting parameters in the corresponding neural network model according to the parameter adjustment step length to obtain an adjusted artificial intelligence model, inputting data output by the data acquisition module when the glucometer is used into the adjusted artificial intelligence model to obtain probability values of blood glucose values of a detected object belonging to all blood glucose value ranges, judging whether conditions for determining the first blood glucose value range as the blood glucose value range of the detected object are met, and if not, continuing to adjust the parameters in the corresponding neural network model according to the parameter adjustment step length until the probability values of the blood glucose values output by the adjusted artificial intelligence model can determine that the first blood glucose value range is the blood glucose value range of the detected object.

Optionally, the data acquisition unit further includes at least one of: a temperature sensor, a skin component detection sensor;

the data output by the data acquisition module further comprises data output by the temperature sensor and/or the skin composition detection sensor.

Optionally, the data acquisition module further includes a pressure sensor disposed adjacent to the data acquisition unit, and the intelligent terminal determines whether the pressure between the data acquisition unit and the arm of the detection object satisfies a preset condition according to the pressure information acquired by the pressure sensor, and outputs a prompt message when the pressure does not satisfy the preset condition.

The technical scheme of the embodiment of the invention has the following advantages:

the blood glucose meter provided by the embodiment of the invention can realize rapid, convenient, accurate and nondestructive blood glucose detection so as to monitor the change of blood glucose at any time and achieve the aim of effectively controlling blood glucose. In addition, the elastic component can enable the data acquisition module to be relatively fixed with the arm of the detection object when being worn on the arm of the detection object, and the influence of shaking on the accuracy of the detection result is avoided. The elastic component is used with the cooperation of arc part, can also make light source module and light detection module all hug closely the arm of detection object, increase the laminating degree with the arm, avoid external light and skin surface reverberation to detect accurate influence.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.

FIG. 1 is a functional block diagram of one specific example of a hong meng operating system based arm-worn glucose meter in an embodiment of the present invention;

fig. 2 is a schematic structural diagram of a specific example of a data acquisition module in an embodiment of the present invention.

Detailed Description

The technical solutions of the present invention will be described clearly and completely with reference to the accompanying drawings, and it should be understood that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

In describing the present invention, it is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. The terms "comprises" and/or "comprising," when used in this specification, are intended to specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. The term "and/or" includes any and all combinations of one or more of the associated listed items. The terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," "outer," and the like are used in the orientation or positional relationship indicated in the drawings for convenience in describing the invention and for simplicity in description, and do not indicate or imply that the referenced device or element must have a particular orientation, be constructed and operated in a particular orientation, and are therefore not to be construed as limiting the invention. The terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. The terms "mounted," "connected," and "coupled" are to be construed broadly and may, for example, be fixedly coupled, detachably coupled, or integrally coupled; can be mechanically or electrically connected; the two elements can be directly connected, indirectly connected through an intermediate medium, or communicated with each other inside; either a wireless or a wired connection. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.

In addition, the technical features involved in the different embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

The present embodiment provides an arm-worn blood glucose meter based on the hongmeng operating system, as shown in fig. 1, including: the system comprises a data acquisition module 1 and an intelligent terminal 2 based on a Hongmon operating system, wherein the data acquisition module 1 is in communication connection with the intelligent terminal 2;

as shown in fig. 2, the data acquisition module 1 includes a housing 11, a data acquisition unit 12; the housing 11 comprises two arc-shaped parts 111 and at least two elastic parts 112, and two ends of the two arc-shaped parts 111 are connected through the elastic parts 112 respectively; the data acquisition unit 12 comprises a light source module and a light detection module, wherein the light source module is used for emitting light rays irradiating to a designated position on the arm of the detection object, and the light detection module is used for detecting reflected light information of the designated position;

the intelligent terminal 2 acquires the data output by the data acquisition module 1 and inputs the data into a trained artificial intelligence model, and determines the blood sugar value or the blood sugar value range of the detection object according to the output of the artificial intelligence model, wherein the data output by the data acquisition module 1 comprises the reflected light information.

The blood glucose meter provided by the embodiment can realize rapid, convenient, accurate and nondestructive blood glucose detection so as to monitor blood glucose changes at any time and achieve the purpose of effectively controlling blood glucose. In addition, the elastic component 112 can make the data acquisition module 1 be fixed relative to the arm of the detection object when being worn on the arm of the detection object, so as to avoid shaking to influence the accuracy of the detection result. Elastic component 112 and arc part 111 cooperation are used, can also make light source module and optical detection module all hug closely the arm of detection object, increase the laminating degree with the arm, avoid external light and skin surface reflection light to detect accurate influence.

Optionally, the data acquisition unit 12 is disposed on one of the arc-shaped components 111, an energy storage module (for example, including a storage battery) for supplying power to the data acquisition unit 12 may be disposed on the other arc-shaped component 111, and an electrical connection wire between the energy storage module and the data acquisition unit 12 may be disposed in the elastic component 112 in a penetrating manner.

The data acquisition module 1 comprises a first wireless communication module, the intelligent terminal 2 comprises a second wireless communication module, the first wireless communication module is used for sending the data acquired by the data acquisition unit 12 to the second wireless communication module, and the second wireless communication module is used for receiving the data sent by the first wireless communication module.

Optionally, the light source module comprises parallel light diodes, and the parallel light diodes are uniformly arranged around the light detection module;

the parallel light diode comprises a tube core and a packaging body, wherein the packaging body is coated outside the tube core;

the package body comprises a light-proof part arranged around the side face of the tube core, the top of the light-proof part covers the edge part of the top of the tube core, the package body also comprises a light-absorbing part arranged at the bottom side of the tube core and a spherical collimating lens arranged at the top side of the tube core;

the die is obtained by forming an epitaxial layer on a substrate and patterning the epitaxial layer, and the die has a vertical side or a side with the top inclined to the center.

Specifically, after the die is obtained by patterning the epitaxial layer, the die may be bonded to a carrier wafer, and then a package covering the die and the bonding body of the carrier wafer is formed to obtain the parallel light diode.

Specifically, the top of the light-tight part only covers the edge part of the top of the tube core, so that a hole which can transmit light exists in the top of the light-tight part, and the spherical collimating lens can be embedded in the hole.

Further optionally, the parallel light diodes include at least two types, the light emitted by the at least two types of parallel light diodes is different, and each type of parallel light diode is uniformly arranged around the light detection module and is spaced from the other parallel light diodes. Alternatively, the parallel light diodes emit different light beams under different excitation electrical signals, for example, the light beams are different according to the excitation current or the excitation voltage.

Optionally, the artificial intelligence model includes: n first neural network models for respectively extracting the characteristics of the reflected light information, wherein N is a positive integer greater than 1; the N first neural network models comprise a neural network model used for extracting the features of the wave number dimension in the reflected light information and a neural network model used for extracting the features of the frequency domain dimension in the reflected light information;

the pyramid model is used for respectively integrating the first feature information of multiple layers extracted by the first neural network to obtain N pieces of second feature information;

a classifier model for outputting a probability that a blood glucose level of the detection target falls within each blood glucose level range, based on the N pieces of second feature information; the classifier model includes a plurality of classifiers that are integrated by a weighted summation.

Specifically, the first neural network model may include a VGG neural network model, a GoogleNet neural network model, and the like, and may further include a neural network formed by a multilayer weighted self-encoder.

Optionally, the intelligent terminal 2 is further configured to preprocess the data output by the data acquisition module 1, where the preprocessing includes establishing a filter function by using a least square fitting coefficient, and performing polynomial least square fitting on the data in the moving window, where an expression of the polynomial fitting is:

z^*(i)＝a₀+a₁i+a₂i²+...+a_bi^b

wherein z is^*(i) A fitting value a obtained by the position of a central point after a multi-time fitting curve is established for the Savitzky-Golay convolution smoothing method₀，a₁，a₂…a_bIs calculated by the following formula:

wherein, the data in the moving window is z (i), i ═ M, …,0, …, M, μ ═ 0,1,2 …, b.

The preprocessing in this embodiment is used to smooth the reflected light information.

In addition, the preprocessing further comprises performing first derivation, multivariate scattering correction and the like on the reflected light information.

Optionally, the classifier model is constructed in the following manner:

obtaining l training samples: { (x)₁,y₁),(x₂,y₂),…,(x_l,y_l)}，x_jFor N of said second characteristic information, y_jInitializing a weight value for the blood sugar range with the total number of the blood sugar ranges of K

Let T be 1,2, …, T be the maximum training times;

according to the weight value w^tSelecting a training sample;

carrying out classification and identification on the blood sugar value range of the training sample h_t:X→Y；

And (3) making K equal to 1,2, … and K, circularly calculating the weight sum of samples in each blood sugar range:judging whether the weight sum of the samples classified correctly in each blood sugar value range is larger than the weight sum of the samples classified into other blood sugar value rangesIf so, performing next circulation, otherwise, turning to a step of performing classification identification on the blood sugar value range of the sample and restarting calculation;

calculate h_tFalse error rate of (2):

order to

Calculate new weight vector:

normalizationObtaining the classifier model as follows:

wherein a plurality of said classifiers h_t(x) May include a minimum distance classifier, a linear discriminant classifier, a K-nearest neighbor classifier, a support vector machine, and/or an extreme learning machine, among others.

Optionally, the training samples of the artificial intelligence model include: reflected light information of the designated position of the test object acquired by the data acquisition module 1 and a blood glucose level range to which a blood glucose level of the test object acquired by a blood sampling test method belongs when the reflected light information is acquired are used.

In this embodiment, the labels of the training samples are a plurality of blood glucose value ranges.

Optionally, the intelligent terminal 2 is further configured to receive a first blood glucose value range input by the user, where the first blood glucose value range is a blood glucose value range to which a blood glucose value of the detection target obtained by a blood sampling detection method when the blood glucose meter is used belongs;

the intelligent terminal 2 is further configured to perform transfer learning training on the artificial intelligence model based on the first blood glucose value range.

The user may or may not be the detection object.

In this embodiment, in order to ensure the reliability of the detection result of the blood glucose meter, before the blood glucose value range of a new detection object is detected, the artificial intelligence model may be subjected to the transfer learning training using the data of the new detection object, and the accuracy of the blood glucose meter may also be verified periodically or aperiodically in the use process of the new detection object, and if the accuracy is not satisfied, the transfer learning training may be performed again.

The device used for the initial training of the artificial intelligence model may not be the intelligent terminal 2. In addition, the intelligent terminal 2 may also obtain a newly optimized artificial intelligence model from another device, for example, the server may perform optimization training on the artificial intelligence model based on the newly collected data, and then push the artificial intelligence model to the intelligent terminal 2.

In addition, the intelligent terminal 2 is further configured to receive an operation input of a user, and generate a corresponding control signal according to the operation input, and send the control signal to the data acquisition module 1, so as to control the work of the data acquisition module 1, for example, start work (specifically, light emission of the light source module is controlled, and the light detection module detects reflected light after the light emission of the light source), stop work, suspend work, and the like, which are not detailed here.

Optionally, the intelligent terminal 2 is configured to obtain data output by the data acquisition module 1 when the blood glucose meter is used, a first probability value that a blood glucose value of the detection object output by the artificial intelligence model belongs to the first blood glucose value range, and a second probability value that the blood glucose value belongs to a second blood glucose value range, where the second blood glucose value range is a blood glucose value range output when the blood glucose meter is used;

the intelligent terminal 2 is further used for determining a parameter adjustment initial step according to the difference value between the first probability value and the second probability value;

the intelligent terminal 2 is further configured to determine a weight value corresponding to parameter adjustment in each neural network model according to the degree of influence of each neural network model in the artificial intelligence model on an output result, and determine a parameter adjustment step length of each neural network model according to the weight value and the initial step length;

the intelligent terminal 2 is further configured to adjust parameters in the corresponding neural network model according to the parameter adjustment step size to obtain an adjusted artificial intelligence model, input data output by the data acquisition module 1 when the blood glucose meter is used to the adjusted artificial intelligence model to obtain probability values that blood glucose values of a detection object belong to respective blood glucose value ranges, and determine whether a condition that the first blood glucose value range is determined to be the blood glucose value range of the detection object is satisfied, and if not, continue to adjust the parameters in the corresponding neural network model according to the parameter adjustment step size until the probability values of the blood glucose values output by the adjusted artificial intelligence model can determine that the first blood glucose value range is the blood glucose value range of the detection object.

Optionally, the data acquisition unit 12 further includes at least one of: a temperature sensor, a skin component detection sensor;

the data output by the data acquisition module 1 further comprises data output by the temperature sensor and/or the skin composition detection sensor.

That is, data output by the temperature sensor and/or the skin composition detection sensor also serves as an input to the artificial intelligence model.

The blood glucose meter provided by the embodiment also considers the influence of temperature, skin components and the like on blood glucose detection by utilizing reflected light, and improves the blood glucose detection precision.

Optionally, the data acquisition module 1 further includes a pressure sensor disposed adjacent to the data acquisition unit 12, and the intelligent terminal 2 determines whether the pressure between the data acquisition unit 12 and the arm of the detection object satisfies a preset condition according to the pressure information acquired by the pressure sensor, and outputs a prompt message when the pressure does not satisfy the preset condition.

Since the pressure may cause the blood flow and the tissue morphology in the arm to change, and these changes may affect the reflection of light, in this embodiment, the pressure applied by the data acquisition module may be detected during the detection process, so as to prevent the blood glucose detection accuracy from being affected by too much pressure.

It should be understood that the above examples are only for clarity of illustration and are not intended to limit the embodiments. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. And obvious variations or modifications therefrom are within the scope of the invention.