Method and system for detecting concentration of glucose solution based on near infrared spectrum

Abstract

The invention relates to the technical field of spectral analysis, and discloses a glucose solution concentration detection method and system based on a near infrared spectrum. The method comprises the following steps: acquiring transmission light intensity of a sample and a reference liquid by using a narrow-band near-infrared LED light source, calculating absorbance characteristics, screening a dynamic wavelength set based on sensitivity, and constructing a concentration prediction model to realize accurate detection of concentration. Compared with the prior art which depends on single wavelength or lossy detection, especially under the condition of temperature and humidity fluctuation or unstable light source power, the technical problem that high-precision concentration measurement cannot be realized is solved. High-precision glucose concentration detection under complex environmental conditions is realized, so that the problem of error accumulation caused by environmental interference is avoided, the detection precision, stability and environmental adaptability of the system are improved, and the reliability and accuracy of glucose solution concentration detection are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of spectral analysis, and in particular relates to a method and system for detecting the concentration of a glucose solution based on near-infrared spectroscopy. Background Art

[0002] At present, solution concentration detection methods based on near-infrared spectroscopy are widely used in the fields of food, medicine, environmental protection, etc., especially in the analysis of sugar solutions and drug components. However, the existing technology has shortcomings in many aspects. For example, most traditional near-infrared spectroscopy methods rely on the measurement of fixed wavelength points, which makes it limited in the use of spectral information. The fixed wavelength method has low sensitivity for samples with small concentration changes, and is greatly disturbed by environmental factors such as temperature, humidity and light source power fluctuations, resulting in the accuracy and stability of the detection results being affected. The existing technology can often only perform accurate measurements under relatively stable experimental conditions, but in actual industrial or experimental environments, these interference factors inevitably exist, thus limiting its application in more complex environments. In addition, many methods rely on linear models for concentration prediction, but because glucose solutions have more complex absorption characteristics in the near-infrared spectrum range, simple linear regression models cannot fully capture the nonlinear relationship between absorbance and concentration, resulting in low prediction accuracy. Existing methods mostly rely on a large amount of experimental data and debugging under specific environments, but lack sufficient adaptability and robustness for variable environmental conditions and complex samples. Therefore, a new detection method is urgently needed that can achieve high-precision and high-stability glucose solution concentration detection under the condition of fluctuations in environmental conditions such as temperature, humidity, and light source power. This method needs to be able to eliminate environmental interference through comprehensive analysis of multi-wavelength information and adapt to different experimental conditions, thereby improving the accuracy of concentration measurement and the robustness of the system. Summary of the invention

[0003] In view of the above-mentioned technical deficiencies, the purpose of the present invention is to propose a method for detecting the concentration of glucose solution based on near-infrared spectroscopy, aiming to solve the technical problem that high-precision concentration measurement cannot be achieved in the prior art due to reliance on single wavelength or lossy detection, especially under conditions of temperature, humidity fluctuations or unstable light source power.

[0004] In order to solve the above technical problems, the present invention adopts the following technical solution: The present invention provides a method for detecting the concentration of a glucose solution based on near infrared spectroscopy.

[0005] The method for detecting the concentration of glucose solution based on near infrared spectroscopy comprises:

[0006] Step S10: irradiate the glucose solution sample in the sample chamber by emitting a narrow-band LED light source with a central wavelength of λ, and collect the transmitted light intensity I of the glucose solution sample.sample (λ) and the transmitted light intensity of the reference liquid I reference (λ);

[0007] Step S20: Calculate the preliminary glucose solution sample transmittance and preliminary glucose solution sample absorbance, the calculation formula is:

[0008]

[0009] A λ0 =-log 10 (T λ0 )

[0010] Among them, T λ0 is the transmittance of the preliminary glucose solution sample, A λ0 is the absorbance of the preliminary glucose solution sample;

[0011] Repeat the measurement 10 times and take the average value to obtain the transmittance T of the glucose solution sample. λ and glucose solution sample absorbance A λ ;

[0012] Step S30: Calculate the sensitivity S(λ) of wavelength λ based on the absorbance data of the glucose solution sample obtained by 10 measurements:

[0013]

[0014] Where S(λ) is the sensitivity at wavelength λ, ΔA(λ) is the amplitude of absorbance change, and σ(λ) is the standard deviation of absorbance noise at wavelength λ;

[0015] Filter the 5-8 wavelengths with the highest sensitivity to obtain the filtered wavelength set λ optimal ;

[0016] Step S40: Based on the wavelength set λ selected in step S30 optimal , for the wavelength set λ optimal The absorbance of glucose solution sample at wavelength A λ Make corrections, the correction formula is:

[0017]

[0018] Among them, A ′ (λ) is the corrected absorbance of the glucose solution sample; f(T, H, P) is the multi-physics field calibration model, which is established based on the influence of temperature T, humidity H and light source power P;

[0019] According to the corrected glucose solution sample absorbance A ′ (λ) Construct a long set λ optimal The corresponding glucose solution sample absorbance characteristic set A′ (λ optimal );

[0020] Step S50: Set the light absorption feature set A ′ (λ optimal ) is converted into the absorption characteristic vector A ′ , build a concentration prediction model, the specific formula is:

[0021] C g =w T A ′ +b

[0022] Among them, C g is the predicted glucose concentration, A ′ is the absorption eigenvector, w T is the weight vector of the partial least squares regression model, and b is the bias term of the concentration prediction model;

[0023] The constructed concentration prediction model is applied, the absorbance characteristic vector of the glucose solution to be tested is taken as input, and the predicted glucose concentration is output.

[0024] Preferably, in step S30, ΔA(λ)=A max (λ)-A min (λ), where A max (λ) is the maximum absorbance at wavelength λ, A min (λ) is the minimum absorbance at wavelength λ.

[0025] Preferably, in step S40, T is measured in real time by a PT100 thermistor, in degrees Celsius; H is collected by a digital humidity sensor, in percentage; and P is detected by a built-in photodiode to measure the real-time emission power of the LED, in milliwatts.

[0026] Preferably, in step S40, the multi-physics field calibration model f(T, H, P) is constructed using a random forest regression method, and the training samples include absorbance data of glucose solution samples with a standard concentration gradient of 1%-5% glucose solution under different temperature, humidity and light source power conditions.

[0027] Preferably, in step S50, during the training process of the constructed concentration prediction model, the input includes a glucose solution with a standard concentration gradient of 1%-5%, and the output includes a standard absorption feature vector of the glucose solution with a standard concentration gradient of 1%-5% processed and converted from step S10 to step S40.

[0028] Preferably, in step S50, a k-fold cross-validation method is used in the training process of the constructed concentration prediction model.

[0029] Preferably, in step S10, the optical path length of the sample chamber is d=1 mm, the light intensity signal is collected by a high-sensitivity InGaAs detector, and the signal output range is 0.5-2.0V.

[0030] The present invention also provides a glucose solution concentration detection system based on near infrared spectroscopy, comprising:

[0031] The information collection module is used to irradiate the glucose solution sample in the sample chamber by emitting a narrow-band LED light source with a central wavelength of λ, and collect the transmitted light intensity I of the glucose solution sample. sample (λ) and the transmitted light intensity of the reference liquid I reference (λ);

[0032] The transmittance and absorbance calculation module is used to calculate the transmittance and absorbance of the preliminary glucose solution sample. The calculation formula is:

[0033]

[0034] A λ0 =-log 10 (T λ0 )

[0035] Among them, T λ0 is the transmittance of the preliminary glucose solution sample, A λ0 is the absorbance of the preliminary glucose solution sample;

[0036] Repeat the measurement 10 times and take the average value to obtain the transmittance T of the glucose solution sample. λ and glucose solution sample absorbance A λ ;

[0037] The wavelength sensitivity calculation module is used to calculate the sensitivity S(λ) of the wavelength λ based on the absorbance data of the glucose solution sample obtained by 10 measurements:

[0038]

[0039] Where S(λ) is the sensitivity at wavelength λ, ΔA(λ) is the amplitude of absorbance change, and σ(λ) is the standard deviation of absorbance noise at wavelength λ;

[0040] Filter the 5-8 wavelengths with the highest sensitivity to obtain the filtered wavelength set λ optimal ;

[0041] Absorption characteristic calibration module, for use based on the wavelength set λ screened in step S30 optimal , for the wavelength set λ optimal The absorbance of glucose solution sample at wavelength A λ Make corrections, the correction formula is:

[0042]

[0043] Among them, A ′ (λ) is the corrected absorbance of the glucose solution sample; f(T, H, P) is the multi-physics field calibration model, which is established based on the influence of temperature T, humidity H and light source power P;

[0044] According to the corrected glucose solution sample absorbance A ′ (λ) Construct a long set λ optimal The corresponding glucose solution sample absorbance characteristic set A ′ (λ optimal );

[0045] The concentration prediction module is used to transform the absorption feature set A ′ (λ optimal ) is converted into the absorption characteristic vector A ′ , build a concentration prediction model, the specific formula is:

[0046] C g =w T A ′ +b

[0047] Among them, C g is the predicted glucose concentration, A ′ is the absorption eigenvector, w T is the weight vector of the partial least squares regression model, and b is the bias term of the concentration prediction model;

[0048] The constructed concentration prediction model is applied, the absorbance characteristic vector of the glucose solution to be tested is taken as input, and the predicted glucose concentration is output.

[0049] The present invention also provides a device for detecting the concentration of a glucose solution based on near-infrared spectroscopy, comprising a memory, a processor, and a detection program for the concentration of a glucose solution based on near-infrared spectroscopy stored in the memory and executable on the processor. When the detection program for the concentration of a glucose solution based on near-infrared spectroscopy is executed by the processor, the detection method for the concentration of a glucose solution based on near-infrared spectroscopy is implemented.

[0050] The present invention also provides a computer program product, including a program for detecting the concentration of a glucose solution based on near-infrared spectroscopy. When the program for detecting the concentration of a glucose solution based on near-infrared spectroscopy is executed by a processor, the method for detecting the concentration of a glucose solution based on near-infrared spectroscopy is implemented.

[0051] The beneficial effects of the present invention are as follows: compared with the prior art which relies on single wavelength or lossy detection, especially under conditions of temperature, humidity fluctuations or unstable light source power, and cannot achieve technical problems of high-precision concentration measurement, the present application realizes high-precision glucose concentration detection under complex environmental conditions through dynamic screening of sensitive wavelengths and construction of concentration prediction models, thereby avoiding the problem of error accumulation caused by environmental interference, improving the detection accuracy, stability and environmental adaptability of the system, and improving the reliability and accuracy of glucose solution concentration detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] 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 some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0053] Figure 1 The figure is a flow chart of a first embodiment of a method for detecting the concentration of a glucose solution based on near infrared spectroscopy according to the present invention.

[0054] Figure 2 The present invention is a schematic diagram of a device for detecting the concentration of a glucose solution based on near infrared spectroscopy. DETAILED DESCRIPTION

[0055] 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.

[0056] Embodiment 1: Figure 1 , which is a schematic flow chart of a first embodiment of a method for detecting the concentration of a glucose solution based on near-infrared spectroscopy according to the present invention, and provides a first embodiment of a method for detecting the concentration of a glucose solution based on near-infrared spectroscopy according to the present invention.

[0057] In a first embodiment, the method for detecting the concentration of a glucose solution based on near infrared spectroscopy includes:

[0058] Step S10: irradiate the glucose solution sample in the sample chamber by emitting a narrow-band LED light source with a central wavelength of λ, and collect the transmitted light intensity I of the glucose solution sample. sample (λ) and the transmitted light intensity of the reference liquid I reference (λ);

[0059] It should be noted that the central wavelength of the narrowband LED light source used in this step selects the characteristic absorption band of the glucose solution within the near-infrared spectrum, and has strong absorption characteristics. This wavelength presents a significant absorption peak in the glucose solution, which can effectively improve the detection sensitivity, and is particularly suitable for accurate measurement when the concentration is low. In order to optimize the signal acquisition effect, the half-width (FWHM) of the light source is set to 120nm to ensure accurate wavelength selection and reduce interference from other irrelevant spectral signals.

[0060] It is understood that the transmitted light intensity and the transmitted light intensity of the reference liquid are collected by an InGaAs photodetector and digitized by a 16-bit analog-to-digital converter. These measurement results can provide high-precision data output to ensure the stability of the experiment and the reliability of the data. Since slight fluctuations in environmental conditions such as temperature and humidity during the experiment may have a certain impact on the transmitted light intensity, this method uses multiple measurements and takes the average value to eliminate the influence of accidental errors and instrument fluctuations, thereby ensuring a more stable and accurate transmitted light intensity value.

[0061] It should be understood that the reference liquid, as a control sample, does not have the light absorption characteristics of the glucose solution, so it can be used as a benchmark data to help eliminate the influence of light source power fluctuations, optical system errors and other environmental factors. By comparing with the reference liquid, the transmitted light intensity data can reflect the actual absorbance of the sample.

[0062] For example, in the experiment, when testing a glucose solution sample with a concentration of 3% (3000mg / dL), the system can irradiate with a light source with a wavelength of 1630nm, collect the transmitted light intensity of the sample and the reference liquid, and then calculate the absorbance of the glucose solution at this wavelength. Through multiple measurements of samples of different concentrations, the absorbance data of the samples can be accurately obtained, and a reliable data basis can be provided for subsequent sensitivity analysis, wavelength screening and concentration prediction models.

[0063] Step S20: Calculate the preliminary glucose solution sample transmittance and preliminary glucose solution sample absorbance, the calculation formula is:

[0064]

[0065] A λ0 =-log 10 (T λ0 )

[0066] Among them, T λ0 is the transmittance of the preliminary glucose solution sample, A λ0 is the absorbance of the preliminary glucose solution sample;

[0067] Repeat the measurement 10 times and take the average value to obtain the transmittance T of the glucose solution sample. λ and glucose solution sample absorbance A λ ;

[0068] It should be noted that the transmittance is calculated by measuring the transmitted light intensity of the glucose solution sample and the transmitted light intensity of the reference liquid. Since the reference liquid does not contain glucose, its transmitted light intensity is used to correct the influence of light source fluctuations and environmental factors; absorbance is calculated by transmittance. Absorbance is the quantification of the absorbance of light and is often used to characterize the sample's absorption capacity for light of different wavelengths. In the concentration detection of glucose solution, there is a certain linear relationship between the absorbance value and the glucose concentration.

[0069] It can be understood that in this step, the transmittance and absorbance are calculated by measuring the values ​​at each wavelength point. In order to reduce measurement errors, the transmitted light intensity and absorbance at each wavelength will be repeatedly measured multiple times because the detector may be affected by random noise or light source instability. By averaging multiple measurements, the system can still provide accurate transmittance and absorbance values ​​even when there is a lot of noise interference, ensuring that subsequent wavelength sensitivity analysis and concentration prediction can be based on high-quality input data.

[0070] It should be understood that absorbance is a logarithmic function of transmittance, and according to the Lambert-Beer law, the absorbance of a sample is proportional to the concentration of the solution, and therefore, accurate measurement of transmittance is highly correlated with the prediction of subsequent glucose concentration.

[0071] Step S30: Calculate the sensitivity S(λ) of wavelength λ based on the absorbance data of the glucose solution sample obtained by 10 measurements:

[0072]

[0073] Where S(λ) is the sensitivity at wavelength λ, ΔA(λ) is the amplitude of absorbance change, and σ(λ) is the standard deviation of absorbance noise at wavelength λ;

[0074] Filter the 5-8 wavelengths with the highest sensitivity to obtain the filtered wavelength set λ optimal ;

[0075] It should be noted that the definition of sensitivity is: Sensitivity S(λ) is calculated based on the absorbance variation range ΔA(λ) and measurement noise ΔA(λ), and is used to evaluate the responsiveness of a certain wavelength point to concentration changes. In actual operation, wavelengths with higher sensitivity have a more obvious response to concentration changes, and therefore have higher measurement accuracy; the absorbance variation range ΔA(λ) represents the maximum variation range of the absorbance value under different concentrations of glucose solution. By measuring standard glucose solution samples (e.g., 1%, 3%, and 5% concentrations), the absorbance variation range ΔA(λ) at each wavelength can be obtained, and then the sensitivity of the wavelength to concentration changes can be determined.

[0076] It can be understood that the noise standard deviation σ(λ) is a parameter that describes the measurement error of each wavelength, representing the range of variation of the absorbance value in multiple repeated measurements. Due to fluctuations in experimental equipment and environment, the measurement results will inevitably be affected by noise, and the standard deviation reflects the stability of the measurement. By measuring multiple times and taking the average, the random error in a single measurement can be reduced, thereby improving the reliability of the sensitivity calculation.

[0077] It should be understood that 5 to 8 wavelength points with the highest sensitivity are selected based on the sensitivity of each wavelength. These wavelength points are the key bands that best reflect concentration changes, which can improve the accuracy and robustness of subsequent concentration prediction models. By reducing the number of wavelength points, the computational complexity of the model is reduced, while avoiding the influence of irrelevant wavelengths on concentration prediction.

[0078] For example, in the experiment, 1%, 3%, and 5% glucose solutions were measured 10 times, and the absorbance data at different wavelengths were calculated. At λ=1630nm, assuming that the maximum absorbance is 0.85 and the minimum absorbance is 0.70, then ΔA(1630)=0.85-0.70=0.15. By calculating the noise standard deviation σ(1630) of the absorbance as 0.02, S(1630)=0.15 / 0.02=7.5. Similar measurements were performed at wavelengths of 1600nm, 1640nm, and 1650nm. Finally, the five wavelengths with the highest sensitivity were screened out, such as λ1=1630nm, λ2=1600nm, …, λ5=1650nm. These wavelengths will be used in the subsequent concentration prediction model.

[0079] Step S40: Based on the wavelength set λ selected in step S30 optimal , for the wavelength set λ optimal The absorbance of glucose solution sample at wavelength A λ Make corrections, the correction formula is:

[0080]

[0081] Among them, A ′ (λ) is the corrected absorbance of the glucose solution sample; f(T, H, P) is the multi-physics field calibration model, which is established based on the influence of temperature T, humidity H and light source power P;

[0082] According to the corrected glucose solution sample absorbance A ′ (λ) Construct a long set λ optimal The corresponding glucose solution sample absorbance characteristic set A ′ (λ optimal );

[0083] It should be noted that in actual measurements, the absorbance of the sample is not only related to its concentration, but also significantly affected by the temperature, humidity and light source power fluctuations in the experimental environment. Changes in temperature may cause changes in the sample's refractive index, thereby causing a shift in the absorption characteristics; changes in humidity will affect the intensity of scattered light in the light path; and instability in the light source power will directly lead to changes in the intensity of the transmitted light, thereby affecting the accuracy of the absorbance. Therefore, in order to eliminate the interference of these factors, this method establishes a multi-physics field calibration model, and corrects the absorbance value by collecting data on temperature, humidity and light source power in real time. The corrected absorbance can more truly reflect the concentration characteristics of the glucose solution without being affected by environmental fluctuations.

[0084] It is understandable that in the multi-physics field calibration model, temperature, humidity and light source power are the main influencing factors. Among them, temperature is measured by a thermistor, humidity is collected by a digital humidity sensor, and light source power is monitored by a photodiode. These real-time collected environmental parameters are input into the calibration model, and the model will calculate a correction value based on different environmental conditions to compensate for environmental interference in the original absorbance.

[0085] It should be understood that the corrected absorbance is obtained by subtracting the environmental correction value calculated by the calibration model from the original absorbance. The corrected absorbance can more accurately reflect the optical properties of the sample, thereby improving the accuracy of subsequent analysis and prediction. At the same time, the corrected absorbance feature set contains the calibrated absorbance of all key wavelengths in the screening wavelength set, which is the input data of the concentration prediction model. By using only the data of sensitive wavelength points, the calculation efficiency and accuracy of the concentration prediction model are improved.

[0086] For example, in the experiment, the temperature of the sample chamber was 25.5°C, the humidity was 45%, and the light source power was 4.95mW. These environmental parameters will cause a slight deviation in the absorbance measurement. The original absorbance value is 0.82, the environmental correction value calculated by the calibration model is 0.02, and the corrected absorbance is 0.80. This calibration process is repeated for each wavelength in the screening wavelength set, thereby constructing a calibrated absorbance feature set, which provides reliable data input for subsequent concentration predictions.

[0087] Step S50: Set the light absorption feature set A ′ (λ optimal ) is converted into the absorption characteristic vector A ′ , build a concentration prediction model, the specific formula is:

[0088] C g =w T A ′ +b

[0089] Among them, C g is the predicted glucose concentration, A ′ is the absorption eigenvector, w T is the weight vector of the partial least squares regression model, and b is the bias term of the concentration prediction model;

[0090] The constructed concentration prediction model is applied, the absorbance characteristic vector of the glucose solution to be tested is taken as input, and the predicted glucose concentration is output.

[0091] It should be noted that the absorbance feature set is composed of the corrected absorbance values ​​of each wavelength point in the screening wavelength set. The corrected absorbance feature set reflects the true absorbance characteristics of the glucose solution sample and is not affected by temperature, humidity or light source power fluctuations. These absorbance features are converted into absorbance feature vectors in wavelength order and used to train the concentration prediction model. The concentration prediction model is constructed using the partial least squares regression method, which can effectively handle the importance of each wavelength point in the absorbance feature vector to concentration prediction and suppress noise.

[0092] It is understandable that partial least squares regression is an algorithm widely used for regression analysis of high-dimensional input data. For this method, each element of the absorbance feature vector represents the absorbance data of a specific wavelength, and these wavelengths have been screened by step S30 to ensure their sensitivity to changes in glucose concentration. The model fits the mapping relationship between the absorbance feature and the concentration through the training data of the standard sample, and generates the weight and bias of the model based on the training results. The model weight reflects the contribution of each wavelength point to the concentration prediction, while the bias is used to adjust the overall offset of the predicted value.

[0093] It should be understood that in the concentration prediction process, the model inputs the absorbance characteristic vector of the sample to be tested, uses the weights and biases obtained through training, and outputs the corresponding concentration prediction value. The advantage of this method is that it can integrate data from multiple wavelength points to improve the prediction accuracy and has strong noise resistance. Since the input absorbance characteristic data has been calibrated by multiple physical fields and wavelength screening, the concentration prediction model has good stability and robustness, and is suitable for concentration detection under different environmental conditions.

[0094] For example, assuming that the screening wavelength set includes 1600nm, 1630nm and 1650nm, the corresponding corrected absorbance feature set is 0.80, 0.72 and 0.65. By arranging these corrected absorbance features in order of wavelength, an absorbance feature vector is formed. The vector is input into the concentration prediction model, and the model calculates the predicted concentration value based on the parameters obtained by training. For example, if the weights of the model obtained by fitting the standard sample training data are 1.5, 2.0 and 1.8 respectively, and the bias is 0.3, the concentration value calculated by the model is 3.39mg / dL. In this way, the model can quickly predict the concentration of the sample to be tested while maintaining high accuracy and stability.

[0095] In addition, the present invention provides a glucose solution concentration detection system based on near-infrared spectroscopy, which adopts a glucose solution concentration detection method based on near-infrared spectroscopy in the above embodiment, and can solve the technical problem of glucose solution concentration detection based on near-infrared spectroscopy. Compared with the prior art, the beneficial effects of the glucose solution concentration detection system based on near-infrared spectroscopy provided by the present invention are the same as the beneficial effects of the glucose solution concentration detection method based on near-infrared spectroscopy provided by the above embodiment, and other technical features of the glucose solution concentration detection system based on near-infrared spectroscopy are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0096] The present invention provides a detection device for the concentration of glucose solution based on near infrared spectroscopy. Figure 2, a glucose solution concentration detection device based on near infrared spectroscopy includes: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions that can be executed by at least one processor, and the instructions are executed by at least one processor so that at least one processor can execute a glucose solution concentration detection method based on near infrared spectroscopy in the above-mentioned embodiment 1. A glucose solution concentration detection device based on near infrared spectroscopy in the embodiment of the present invention may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. A glucose solution concentration detection device based on near infrared spectroscopy is only an example and should not bring any limitations to the functions and scope of use of the embodiments of the present invention. A detection device for the concentration of a glucose solution based on near infrared spectroscopy may include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of a detection device for the concentration of a glucose solution based on near infrared spectroscopy are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 can allow a detection device for the concentration of a glucose solution based on near-infrared spectroscopy to communicate wirelessly or wired with other devices to exchange data. Although a detection device for the concentration of a glucose solution based on near-infrared spectroscopy with various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have alternatively.

[0097] The present invention also provides a computer program product, including a computer program, which implements the steps of the above-mentioned method for detecting the concentration of a glucose solution based on near-infrared spectroscopy when executed by a processor. The computer program product provided by the present invention can solve the technical problem of detecting the concentration of a glucose solution based on near-infrared spectroscopy. Compared with the prior art, the beneficial effects of the computer program product provided by the present invention are the same as the beneficial effects of the method for detecting the concentration of a glucose solution based on near-infrared spectroscopy provided in the above-mentioned embodiment, which will not be repeated here.

[0098] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through a communication device, or installed from the storage device 1003, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present invention are executed.

[0099] It should be understood that the various parts disclosed in the present invention can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0100] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for detecting the concentration of a glucose solution based on near infrared spectroscopy, characterized in that: Methods include: Step S10: irradiate the glucose solution sample in the sample chamber by emitting a narrow-band LED light source with a central wavelength of λ, and collect the transmitted light intensity I of the glucose solution sample. sample (λ) and the transmitted light intensity I of the reference liquid reference (λ); Step S20: Calculate the preliminary glucose solution sample transmittance and preliminary glucose solution sample absorbance, the calculation formula is: A λ0 =-log 10 (T λ0 ) Among them, T λ0 is the transmittance of the preliminary glucose solution sample, A λ0 is the absorbance of the preliminary glucose solution sample; Repeat the measurement 10 times and take the average value to obtain the transmittance T of the glucose solution sample. λ and glucose solution sample absorbance A λ ; Step S30: Calculate the sensitivity S(λ) of wavelength λ based on the absorbance data of the glucose solution sample obtained by 10 measurements: Where S(λ) is the sensitivity at wavelength λ, ΔA(λ) is the amplitude of absorbance change, and σ(λ) is the standard deviation of absorbance noise at wavelength λ; Filter the 5-8 wavelengths with the highest sensitivity to obtain the filtered wavelength set λ optimal ; Step S40: Based on the wavelength set λ selected in step S30 optomal , for the wavelength set λ optomal The absorbance of glucose solution sample at wavelength A λ Make corrections, the correction formula is: Among them, A ′ (λ) is the corrected absorbance of the glucose solution sample; f(T, H, P) is the multi-physics field calibration model, which is established based on the influence of temperature T, humidity H and light source power P; According to the corrected glucose solution sample absorbance A ′ (λ) Construct a long set λ optimal The corresponding glucose solution sample absorbance characteristic set A ′ (λ optimal ); Step S50: Set the absorption feature set A ′ (λ optimal ) is converted into the absorption characteristic vector A ′ , build a concentration prediction model, the specific formula is: C g =w T A ′ +b Among them, C g is the predicted glucose concentration, A ′ is the absorption eigenvector, w T is the weight vector of the partial least squares regression model, and b is the bias term of the concentration prediction model; The constructed concentration prediction model is applied, the absorbance characteristic vector of the glucose solution to be tested is taken as input, and the predicted glucose concentration is output.

2. A method for detecting the concentration of a glucose solution based on near infrared spectroscopy as claimed in claim 1, characterized in that: In step S30, ΔA(λ)=A max (λ)-A min (λ), where A max (λ) is the maximum absorbance at wavelength λ, A min (λ) is the minimum absorbance at wavelength λ.

3. A method for detecting the concentration of a glucose solution based on near infrared spectroscopy as claimed in claim 1, characterized in that: In step S40, T is measured in real time by a PT100 thermistor in degrees Celsius; H is collected by a digital humidity sensor in percentage; and P is detected by a built-in photodiode to measure the real-time LED transmission power in milliwatts (mW).

4. A method for detecting the concentration of a glucose solution based on near infrared spectroscopy as claimed in claim 1, characterized in that: In step S40, the multi-physics field calibration model f(T,H,P) is constructed using the random forest regression method, and the training samples include the absorbance data of glucose solution samples with a standard concentration gradient of 1%-5% glucose solution under different temperature, humidity and light source power conditions.

5. The method for detecting the concentration of a glucose solution based on near infrared spectroscopy according to claim 1, characterized in that: In step S50, during the training process of the constructed concentration prediction model, a glucose solution with a standard concentration gradient of 1%-5% is input, and a standard absorption feature vector of the glucose solution with a standard concentration gradient of 1%-5% is output after being processed and converted from step S10 to step S40.

6. A method for detecting the concentration of a glucose solution based on near infrared spectroscopy as claimed in claim 1, characterized in that: In step S50, a k-fold cross validation method is used in the training process of the constructed concentration prediction model.

7. A method for detecting the concentration of a glucose solution based on near infrared spectroscopy as claimed in claim 1, characterized in that: In step S10, the optical path length of the sample chamber is d=1 mm, the light intensity signal is collected by an InGaAs detector, and the signal output range is 0.5-2.0V.

8. A glucose solution concentration detection system based on near infrared spectroscopy, characterized in that: The detection system of glucose solution concentration based on near infrared spectroscopy comprises: The information collection module is used to irradiate the glucose solution sample in the sample chamber by emitting a narrow-band LED light source with a central wavelength of λ, and collect the transmitted light intensity I of the glucose solution sample. sample (λ) and the transmitted light intensity I of the reference liquid reference (λ); The transmittance and absorbance calculation module is used to calculate the transmittance and absorbance of the preliminary glucose solution sample. The calculation formula is: A λ0 =-log 10 (T λ0 ) Among them, T λ0 is the transmittance of the preliminary glucose solution sample, A λ0 is the absorbance of the preliminary glucose solution sample; Repeat the measurement 10 times and take the average value to obtain the transmittance T of the glucose solution sample. λ and glucose solution sample absorbance A λ ; Repeat the measurement 10 times and take the average value to obtain the transmittance T of the glucose solution sample. λ and glucose solution sample absorbance A λ ; The wavelength sensitivity calculation module is used to calculate the sensitivity S(λ) of the wavelength λ based on the absorbance data of the glucose solution sample obtained by 10 measurements: Where S(λ) is the sensitivity at wavelength λ, ΔA(λ) is the amplitude of absorbance change, and σ(λ) is the standard deviation of absorbance noise at wavelength λ; Filter the 5-8 wavelengths with the highest sensitivity to obtain the filtered wavelength set λ 0ptimal ; Absorption characteristic calibration module, for use based on the wavelength set λ screened in step S30 optimal , for the wavelength set λ optimal The absorbance of glucose solution sample at wavelength A λ Make corrections, the correction formula is: Among them, A ′ (λ) is the corrected absorbance of the glucose solution sample; f(T, H, P) is the multi-physics field calibration model, which is established based on the influence of temperature T, humidity H and light source power P; According to the corrected glucose solution sample absorbance A ′ (λ) Construct a long set λ optimal The corresponding glucose solution sample absorbance characteristic set A ′ (λ optimal ); The concentration prediction module is used to transform the absorption feature set A ′ (λ optimal ) is converted into the absorption characteristic vector A ′ , build a concentration prediction model, the specific formula is: C g =w T A ′ +b Among them, C g is the predicted glucose concentration, A ′ is the absorption eigenvector, w T is the weight vector of the partial least squares regression model, and b is the bias term of the concentration prediction model; The constructed concentration prediction model is applied, the absorbance characteristic vector of the glucose solution to be tested is taken as input, and the predicted glucose concentration is output.

9. A glucose solution concentration detection device based on near infrared spectroscopy, characterized in that: The device for detecting the concentration of glucose solution based on near-infrared spectroscopy comprises: a memory, a processor, and a detection program for the concentration of glucose solution based on near-infrared spectroscopy stored in the memory and executable on the processor. When the detection program for the concentration of glucose solution based on near-infrared spectroscopy is executed by the processor, the method for detecting the concentration of glucose solution based on near-infrared spectroscopy described in any one of claims 1 to 7 is implemented.

10. A computer program product, characterized in that The computer program product includes a program for detecting the concentration of a glucose solution based on near-infrared spectroscopy, and when the program for detecting the concentration of a glucose solution based on near-infrared spectroscopy is executed by a processor, the method for detecting the concentration of a glucose solution based on near-infrared spectroscopy described in any one of Claims 1 to 7 is implemented.

Images