Raman spectrum non-invasive blood glucose detection mouse

Abstract

The invention relates to the technical field of noninvasive blood glucose detection, in particular to a Raman spectrum noninvasive blood glucose detection mouse which comprises a signal acquisition module, a signal processing module, a noise suppression module and a data output module. The signal acquisition module adopts a continuous wavelength tunable laser and a dual-channel optical fiber probe, so that optical path crosstalk is remarkably reduced; the noise suppression module effectively removes background noise based on wavelet transform and a principal component analysis algorithm; the signal processing module improves the signal quality through a low-noise amplification circuit and a filtering design; and the data output module realizes efficient calculation and real-time display by using an embedded processor. According to the invention, the detection precision, the system stability and the anti-interference capability can be improved, and a reliable solution is provided for noninvasive blood glucose monitoring.

Description

Technical Field

[0001] The invention belongs to the technical field of biomedical detection, and in particular relates to a Raman spectroscopy non-invasive blood glucose detection mouse. Background Art

[0002] Non-invasive blood glucose monitoring technology, based on optical principles, has become a research hotspot due to its advantages of eliminating the need for blood sampling, ease of operation, and real-time monitoring. However, existing non-invasive blood glucose monitoring technologies still have some shortcomings in terms of accuracy, system stability, and applicable scenarios, which hinder their widespread clinical application.

[0003] The patent with publication number CN106264555B proposes a technical solution for transmitted light detection using light sources of multiple wavelengths (including characteristic wavelengths of blood glucose). The transmitted light is received by a photodetector and converted into an electrical signal, which is finally processed by the host to obtain the blood glucose value. This technology realizes the detection of blood glucose in a non-invasive manner and has the advantages of high precision, low cost and portability. However, this technical solution mainly relies on light sources and transmitted light signals of specific wavelengths, and is easily affected by the uneven depth of human tissue and the surface state of the skin, resulting in a decrease in the repeatability and stability of the test results. In addition, the solution does not fully consider the impact of differences in biological tissues between individuals on the absorption and scattering of light signals, which may further reduce the detection accuracy.

[0004] The above issues indicate that existing non-invasive blood glucose monitoring technologies still have room for improvement in terms of accuracy, system simplification, and anti-interference capabilities. Therefore, the present invention provides a Raman spectroscopy non-invasive blood glucose monitoring mouse. This mouse aims to improve detection accuracy and stability by optimizing spectral signal acquisition and processing methods, while simplifying the system architecture and enhancing anti-interference capabilities, thereby meeting the demand for efficient and reliable non-invasive blood glucose monitoring systems. Summary of the Invention

[0005] The purpose of the present invention is to provide a Raman spectroscopy non-invasive blood glucose detection mouse to solve the deficiencies of the existing non-invasive blood glucose detection technology proposed in the above background technology in terms of detection accuracy, system stability and anti-interference ability.

[0006] The technical solution of the present invention is as follows: it includes a signal acquisition module, a signal processing module, a noise suppression module and a data output module. The signal acquisition module is composed of a light source assembly, a fiber optic probe and a photoelectric converter. The light source assembly transmits excitation light to the surface of human skin through the fiber optic probe. The photoelectric converter receives Raman scattered light returned from the skin tissue and converts it into an electrical signal. The signal processing module contains a preamplifier circuit, a filter circuit and an analog-to-digital converter. The preamplifier circuit performs preliminary amplification on the weak signal output by the photoelectric converter. The filter circuit is used to remove high-frequency noise. The analog-to-digital converter converts the analog signal into a digital signal for subsequent processing. The noise suppression module uses a digital signal processing unit based on an adaptive algorithm, combined with a frequency domain analysis method, to separate the effective components and background noise in the Raman spectral signal. The data output module calculates the received digital signal through an embedded processor and displays the result in numerical form on the LCD screen.

[0007] Furthermore, the light source assembly utilizes a continuously tunable laser with a wavelength range of 780nm to 950nm, capable of stimulating the characteristic Raman peak of glucose molecules. The fiber optic probe is designed as a dual-channel structure, with one channel transmitting the excitation light and the other collecting scattered light. The two channels are separated by an optical isolation film to reduce crosstalk from the excitation light. The photoelectric converter utilizes a highly sensitive avalanche photodiode with a response time of less than 1 nanosecond and a dynamic range of 120dB, ensuring efficient capture of weak Raman signals.

[0008] Furthermore, the preamplifier circuit in the signal processing module utilizes a low-noise operational amplifier with a gain of 100 and an input impedance greater than 10 megohms to match the output characteristics of the avalanche photodiode. The filtering circuit consists of two active low-pass filters connected in series, with the first stage having a cutoff frequency of 10 kHz and the second stage having a cutoff frequency of 1 kHz, gradually attenuating high-frequency interference signals. The analog-to-digital converter uses a 16-bit successive approximation ADC with a sampling rate of 1 MSPS to ensure high-precision digitization of the signal.

[0009] Furthermore, the core of the noise suppression module is a multi-scale decomposition algorithm based on wavelet transforms. This algorithm extracts characteristic information from different frequency bands by performing a multi-level decomposition of the original signal, while simultaneously suppressing noise components using adaptive thresholds. The frequency domain analysis method uses a fast Fourier transform (FFT) to convert the time domain signal into the frequency domain, identifying and removing frequency components unrelated to the Raman characteristic peaks. Furthermore, the noise suppression module integrates a dimensionality reduction technique based on principal component analysis (PCA) to reduce the impact of redundant information on subsequent calculations.

[0010] Furthermore, to evaluate the synergistic performance of the signal acquisition module and the noise suppression module, a laboratory test environment was designed. This test environment included a biomimetic material model simulating human tissue, a standard glucose solution, and a light source power adjustment device. The biomimetic material model consisted of a silica gel substrate with dispersed microspheres to simulate the optical scattering properties of skin tissue. The standard glucose solution concentration ranged from 0 mmol / L to 30 mmol / L in 2 mmol / L intervals to verify the system's detection linearity and sensitivity.

[0011] The biomimetic material model is prepared as follows: First, medical-grade silicone and a curing agent are mixed in a 10:1 mass ratio. Glass microspheres with diameters ranging from 1 to 5 μm are then added, with the volume fraction controlled between 5% and 10%. The mixture is poured into a mold and cured in a 60°C incubator for 24 hours. After curing, an ultrasonic cleaner is used to remove surface impurities, and a 50 μm-thick transparent protective film is applied to the surface to simulate the optical properties of the epidermis.

[0012] Furthermore, the embedded processor in the data output module uses an ARM Cortex-M4 architecture microcontroller with a main frequency of 168MHz and a built-in floating-point unit (FPU), which can efficiently perform complex mathematical operations. The LCD screen uses an OLED display with a resolution of 128×64 and supports a graphical interface. Users can switch display modes with a button to view real-time blood glucose values ​​or historical data curves. The embedded processor communicates with the analog-to-digital converter via the I2C bus, with a data transmission rate of 400kHz, ensuring real-time signal quality.

[0013] Furthermore, the signal acquisition module and signal processing module are connected via a shielded cable, with the shield grounded to reduce electromagnetic interference. The noise suppression module and data output module communicate via an SPI interface, with a 10MHz clock frequency and a data frame format of 8 data bits, 1 stop bit, and no parity check. Each module is powered by a unified power management unit, which provides three voltage outputs: +5V, +3.3V, and -5V, which are used to drive the light source assembly, signal processing circuit, and photoelectric converter, respectively.

[0014] The present invention provides a Raman spectroscopy non-invasive blood glucose detection mouse through improvement, which has the following improvements and advantages compared with the existing technology:

[0015] By utilizing a continuously wavelength-tunable laser and a dual-channel fiber optic probe, the signal acquisition module can excite Raman scattered light at multiple wavelengths, effectively reducing crosstalk from the excitation light and improving signal purity. The dual-channel design allows for more complete separation of excitation and scattered light, avoiding the optical path interference common in traditional single-channel probes.

[0016] By employing a noise suppression algorithm based on wavelet transform and principal component analysis, the noise suppression module achieves efficient denoising of Raman spectral signals, significantly reducing the impact of background noise on detection results. The introduction of an adaptive threshold allows the system to dynamically adjust noise suppression parameters based on varying signal strengths, thereby maximizing noise removal while maintaining signal integrity.

[0017] By employing a highly sensitive avalanche photodiode and a low-noise preamplifier circuit, the signal acquisition module significantly enhances its ability to capture weak Raman signals. This expanded dynamic range enables the system to maintain high accuracy across a wider range of signal strengths. A two-stage active low-pass filter design further optimizes signal quality, providing a reliable foundation for subsequent processing.

[0018] By utilizing an embedded processor based on the ARM Cortex-M4 architecture and a Fast Fourier Transform algorithm, the data output module efficiently processes complex signals. This increased computing speed enables the system to display real-time blood glucose levels. The graphical interface enhances the user experience, and the ability to store historical data curves facilitates long-term monitoring.

[0019] By using shielded cables and a unified power management unit, electromagnetic compatibility within the system is effectively guaranteed, significantly improving signal transmission stability between modules. The high-speed communication capability of the SPI interface ensures real-time data transmission, avoiding detection errors caused by communication delays. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a block diagram of the overall system structure of the present invention, showing the connection relationship and main functions of the signal acquisition module, signal processing module, noise suppression module and data output module.

[0021] Figure 2 This is a structural diagram of the signal acquisition module, which includes a light source assembly, a fiber optic probe, and a photoelectric converter. The fiber optic probe adopts a dual-channel design, which is used for excitation light transmission and scattered light collection respectively.

[0022] Figure 3 This is the circuit schematic of the signal processing module, which includes a preamplifier circuit, a filter circuit, and an analog-to-digital converter. It focuses on the process of signal amplification and the design of the filter circuit.

[0023] Figure 4 This is the algorithm flow chart of the noise suppression module, which describes the noise separation and suppression process based on wavelet transform and principal component analysis, as well as the application of frequency domain analysis method.

[0024] Figure 5This is a schematic diagram of the display interface of the data output module, showing the real-time blood glucose value display and the graphical presentation of historical data curves on the LCD screen. DETAILED DESCRIPTION

[0025] The present invention provides a Raman spectroscopy non-invasive blood glucose detection mouse, the overall structure of which is as follows: Figure 1 As shown, the system comprises a signal acquisition module, a signal processing module, a noise suppression module, and a data output module. The signal acquisition module contacts the human skin surface via a fiber optic probe. Excitation light from the light source assembly is transmitted through the fiber optic probe to the skin surface, triggering Raman scattering. The photoelectric converter receives the returned scattered light and converts it into an electrical signal. The signal processing module is connected to the signal acquisition module via a shielded cable. A preamplifier circuit initially amplifies the weak electrical signal output by the photoelectric converter. A filter circuit further removes high-frequency interference. The analog-to-digital converter converts the analog signal into a digital signal for subsequent processing. The noise suppression module communicates with the signal processing module via an SPI interface and uses an algorithm based on wavelet transform and principal component analysis to separate and suppress noise from the digital signal. The data output module connects to the analog-to-digital converter via an I2C bus. An embedded processor calculates the received digital signal and displays the result on an LCD screen.

[0026] The specific structure of the signal acquisition module is as follows Figure 2 As shown, the light source assembly uses a continuous wavelength tunable laser with a wavelength range of 780nm to 950nm, which can cover the characteristic Raman peak region of glucose molecules. The fiber optic probe is designed as a dual-channel structure, one channel for transmitting excitation light and the other for collecting scattered light. The two channels are separated by an optical isolation film to reduce crosstalk. The photoelectric converter uses a high-sensitivity avalanche photodiode with a response time of less than 1 nanosecond and a dynamic range of 120dB. In actual assembly, the light source assembly is connected to one end of the fiber optic probe through a fiber optic coupler, and the other end of the fiber optic probe is tightly fitted with the photosensitive surface of the photoelectric converter to ensure efficient transmission and reception of optical signals. The output end of the photoelectric converter is connected to the input end of the signal processing module through a shielded cable, and the shielding layer is grounded to reduce electromagnetic interference.

[0027] The circuit principle of the signal processing module is as follows Figure 3As shown, the preamplifier circuit uses a low-noise operational amplifier with a gain of 100 and an input impedance greater than 10 megohms to match the output characteristics of the avalanche photodiode. The output of the preamplifier circuit is connected to the input of the filtering circuit, which consists of two active low-pass filters connected in series. The first stage has a cutoff frequency of 10 kHz, and the second stage has a cutoff frequency of 1 kHz, gradually attenuating high-frequency interference signals. The output of the filtering circuit is connected to the input of the analog-to-digital converter (ADC), a 16-bit successive approximation register (SAR) ADC with a sampling rate of 1 MSPS. The digital output of the ADC is connected to the input of the noise suppression module via an SPI interface. The clock frequency is set to 10 MHz, and the data frame format is 8 data bits, 1 stop bit, and no parity check.

[0028] The algorithm flow of the noise suppression module is as follows Figure 4 As shown, the original signal is first decomposed at multiple levels using wavelet transform to extract characteristic information from different frequency bands, while noise components are simultaneously suppressed using an adaptive threshold. The adaptive threshold is dynamically adjusted based on signal strength to ensure that effective signal components are retained while noise is removed. The frequency domain analysis method uses fast Fourier transform to convert the time domain signal into the frequency domain, identifying and eliminating frequency components unrelated to the Raman characteristic peak. In addition, the noise suppression module integrates a dimensionality reduction technique based on principal component analysis. By performing eigenvalue decomposition on the signal matrix, the main components are extracted to reduce the impact of redundant information on subsequent calculations. The output of the noise suppression module is connected to the input of the data output module via an SPI interface, with a data transmission rate of 10MHz.

[0029] The display interface of the data output module is as follows: Figure 5 As shown, the embedded processor uses an ARM Cortex-M4 architecture microcontroller with a main frequency of 168MHz and a built-in floating-point unit to support complex mathematical operations. The embedded processor communicates with the analog-to-digital converter via the I2C bus, with a data transmission rate of 400kHz. The LCD screen uses an OLED display with a resolution of 128×64 and supports a graphical interface. Users can switch display modes by pressing a button to view real-time blood glucose values ​​or historical data curves. The embedded processor's internal memory is used to store historical data, supporting data recording for up to 30 days. The LCD driver circuit is connected to the embedded processor via the I2C bus, with a data transmission rate of 400kHz.

[0030] The shielded cable length between the signal acquisition module and the signal processing module is kept within 50 cm to minimize signal attenuation and electromagnetic interference. The SPI interface between the noise suppression module and the data output module uses differential signaling to ensure stable high-speed data transmission. Each module is powered by a unified power management unit, which provides three voltage outputs: +5V, +3.3V, and -5V, which are used to drive the light source assembly, signal processing circuit, and photoelectric converter, respectively. The output of the power management unit is connected to each module through an independent voltage stabilization circuit to ensure supply voltage stability.

[0031] The laboratory testing environment was designed to include a biomimetic material model simulating human tissue, a standard glucose solution, and a light source power control device. The biomimetic material model consists of a silicone base with dispersed microsphere particles to simulate the optical scattering properties of skin tissue. The biomimetic material model was prepared as follows: First, medical-grade silicone and a curing agent were mixed at a mass ratio of 10:1. Glass microspheres with diameters ranging from 1μm to 5μm were then added, with the microsphere volume fraction controlled between 5% and 10%. The mixture was poured into a mold and cured in a 60°C incubator for 24 hours. After curing, surface impurities were removed using an ultrasonic cleaner, and a 50μm-thick transparent protective film was applied to the surface to simulate the optical properties of the epidermis. The standard glucose solution concentration ranged from 0mmol / L to 30mmol / L in 2mmol / L increments to verify the detection linearity and sensitivity of the system.

[0032] In actual application, the user places the fiber optic probe on the skin surface of a finger or arm. Excitation light emitted by the light source assembly is transmitted through the fiber optic probe to the skin surface, triggering Raman scattering. A photoelectric converter receives the returned scattered light and converts it into an electrical signal, which is then transmitted via a shielded cable to the signal processing module. A preamplifier circuit performs preliminary amplification on the electrical signal, a filter circuit removes high-frequency interference, and an analog-to-digital converter converts the analog signal into a digital signal. The digital signal is transmitted via the SPI interface to the noise suppression module, where it undergoes wavelet transform and principal component analysis to remove noise. The processed signal is then transmitted via the SPI interface to the data output module. An embedded processor calculates the received digital signal and displays the result on an LCD screen. The user can view real-time blood glucose values ​​or historical data curves on the LCD screen, achieving non-invasive blood glucose monitoring.

[0033] In order to better enable relevant personnel in this technical field to fully understand and implement the present invention, the specific implementation principle of the present invention is further supplemented below in combination with a specific application scenario.

[0034] During actual use, the user first places the fiber optic probe on the skin surface of a person's finger or arm. Excitation light from the light source assembly is transmitted to the skin surface through one channel of the fiber optic probe, where it interacts with glucose molecules in the skin tissue, triggering a Raman scattering effect. Because glucose molecules have specific Raman characteristic peaks, their scattered light contains information related to the glucose concentration. Another channel of the fiber optic probe is responsible for collecting this scattered light and transmitting it to a photoelectric converter. The photoelectric converter uses a highly sensitive avalanche photodiode that can quickly respond to and capture weak Raman scattering signals, while converting them into electrical signals for output.

[0035] The converted electrical signal is transmitted via a shielded cable to the signal processing module. During this process, the preamplifier circuit uses a low-noise operational amplifier to initially amplify the received weak electrical signal, with a gain set to 100x to match the output characteristics of the avalanche photodiode. The signal then enters the filtering circuit, which consists of two active low-pass filters connected in series. The first filter has a cutoff frequency of 10kHz to attenuate high-frequency interference; the second filter has a cutoff frequency of 1kHz to further optimize signal quality. The filtered signal is then transmitted to the analog-to-digital converter, which converts the analog signal into a digital signal at a sampling rate of 1MSPS, ensuring high-precision digitization.

[0036] The digital signal is transmitted to the noise suppression module via the SPI interface. This module first uses wavelet transform to perform multi-level decomposition of the signal, extracting characteristic information from different frequency bands. An adaptive threshold algorithm dynamically adjusts the noise suppression parameters based on signal strength, thereby removing background noise while retaining effective signal components. The signal is then converted from the time domain to the frequency domain using a fast Fourier transform (FFT), identifying and removing frequency components unrelated to the Raman characteristic peaks. Furthermore, dimensionality reduction techniques based on principal component analysis (PCA) further reduce redundant information and extract the main signal components, providing purer data for subsequent calculations.

[0037] The processed signal is transmitted to the data output module via the SPI interface. The embedded processor uses an ARM Cortex-M4 architecture microcontroller with a built-in floating-point unit (FPU), which efficiently performs complex mathematical operations. The processor calculates the received digital signal and displays the results numerically on the LCD screen. The LCD screen uses a 128×64 resolution OLED display and supports a graphical interface. Users can switch display modes with a button to view real-time blood glucose values ​​or historical data curves. The embedded processor's internal memory can store up to 30 days of historical data, allowing users to monitor blood glucose trends over the long term.

[0038] During the whole system running process, the connection mode and power supply design among the modules are optimized. Shielded cable is used to connect the signal acquisition module and the signal processing module, and the shield layer is grounded to reduce electromagnetic interference, while the cable length is controlled within 50 cm to avoid signal attenuation. Differential signal transmission is used between the noise suppression module and the data output module to ensure the stability of high-speed data transmission. The modules are powered by a unified power management unit, which provides +5V, +3.3V and -5V three voltage outputs, respectively, for driving the light source assembly, signal processing circuit and photoelectric converter. Independent voltage stabilizing circuit further ensures the stability of the power supply voltage of each module.

[0039] Through the above steps, the system realizes non-invasive detection of human blood glucose. The design of the double-channel optical fiber probe significantly reduces the crosstalk between the excitation light and the scattered light, improving the purity of the signal. The combination of high-sensitivity avalanche photodiode and low-noise preamplifier circuit enhances the ability to capture weak Raman signals, and the series design of two-stage active low-pass filter further optimizes the signal quality. The noise suppression algorithm based on wavelet transform and principal component analysis effectively removes background noise, ensuring the accuracy of the detection results. Finally, the embedded processor provides real-time, intuitive blood glucose monitoring experience for users through efficient mathematical operations and graphical interface display.

[0040] The contents not described in detail in the specification are all prior art known to those skilled in the art, and the model parameters of each electric appliance are not specifically limited, and conventional equipment can be used. In the technical solution, the electric appliance control elements not mentioned belong to the prior art, so they are not shown in the figure and will not be described here.

[0041] The above only describes the preferred embodiments of the present application and is not used to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A Raman spectroscopy non-invasive blood glucose detection mouse, characterized by: It includes a signal acquisition module, a signal processing module, a noise suppression module and a data output module. The signal acquisition module is composed of a light source component, an optical fiber probe and a photoelectric converter. The light source component transmits excitation light to the surface of human skin through the optical fiber probe. The photoelectric converter receives Raman scattered light returned from the skin tissue and converts it into an electrical signal. The signal processing module contains a preamplifier circuit, a filter circuit and an analog-to-digital converter. The preamplifier circuit preliminarily amplifies the electrical signal output by the photoelectric converter. The filter circuit is used to remove high-frequency interference. The analog-to-digital converter converts the analog signal into a digital signal. The noise suppression module uses an algorithm based on wavelet transform and principal component analysis to separate and suppress noise in the digital signal. The data output module calculates the received digital signal through an embedded processor and outputs the result to a liquid crystal display.

2. The Raman spectroscopy non-invasive blood glucose detection mouse according to claim 1, characterized in that: The light source component adopts a continuous wavelength tunable laser, and its wavelength range is 780nm to 950nm.

3. The Raman spectroscopy non-invasive blood glucose detection mouse according to claim 1, characterized in that: The optical fiber probe is designed as a dual-channel structure, wherein one channel is used to transmit excitation light and the other channel is used to collect scattered light, and the two channels are separated by an optical isolation film.

4. The Raman spectroscopy non-invasive blood glucose detection mouse according to claim 1, characterized in that: The photoelectric converter uses a high-sensitivity avalanche photodiode, whose response time is less than 1 nanosecond and dynamic range reaches 120dB.

5. The Raman spectroscopy non-invasive blood glucose detection mouse according to claim 1, characterized in that: The preamplifier circuit adopts a low-noise operational amplifier, the gain is set to 100 times, and the input impedance is greater than 10 megohms.

6. The Raman spectroscopy non-invasive blood glucose detection mouse according to claim 1, characterized in that: The filtering circuit is composed of two stages of active low-pass filters connected in series, the cut-off frequency of the first stage is 10 kHz, and the cut-off frequency of the second stage is 1 kHz.

7. The Raman spectroscopy non-invasive blood glucose detection mouse according to claim 1, characterized in that: The analog-to-digital converter uses a successive approximation ADC with a 16-bit resolution and a sampling rate of 1 MSPS.

8. The Raman spectroscopy non-invasive blood glucose detection mouse according to claim 1, characterized in that: The noise suppression module decomposes the original signal at multiple levels through wavelet transformation, extracts characteristic information of different frequency bands, and suppresses noise components using an adaptive threshold.

9. The Raman spectroscopy non-invasive blood glucose detection mouse according to claim 1, characterized in that: The noise suppression module uses fast Fourier transform to convert the time domain signal into the frequency domain signal, and identifies and eliminates the frequency components unrelated to the Raman characteristic peak.

10. The Raman spectroscopy non-invasive blood glucose detection mouse according to claim 1, characterized in that: The embedded processor in the data output module is a single-chip microcomputer with an ARM Cortex-M4 architecture, a main frequency of 168MHz, and a built-in floating-point operation unit.

Images