Optical fiber Raman spectrum enhanced rapid detection system for aflatoxin in traditional Chinese medicinal materials

Abstract

The invention discloses an optical fiber Raman spectrum enhanced rapid detection system for traditional Chinese medicinal material aflatoxin, and belongs to the technical field of safety detection. The system comprises an optical excitation module, a sampling probe module, a signal enhancement module, a spectrum acquisition module and an intelligent analysis module, a dynamic adaptive surface enhanced Raman scattering (SERS) technology is innovatively adopted, signal enhancement is realized through a specially-made MXene composite substrate, and sample acquisition and processing are completed in combination with a photonic crystal fiber probe and a micro-fluidic chip. And the intelligent analysis module effectively separates target signals from complex matrix interference by applying a deep learning algorithm, so that rapid identification and quantitative analysis of aflatoxin are realized. The whole system integrates optical detection, a microfluidic technology and an artificial intelligence algorithm, has the characteristics of high detection speed, simplicity and convenience in operation, reliable result and the like, is suitable for on-site rapid screening of production, circulation and supervision links of the traditional Chinese medicinal materials, and provides an innovative technical solution for quality safety control of the traditional Chinese medicinal materials.

Description

Technical Field

[0001] The present invention relates to the technical field of safety detection, and in particular to a rapid detection system for aflatoxin in traditional Chinese medicines enhanced by optical fiber Raman spectroscopy. Background Art

[0002] Aflatoxins (AFs), potent carcinogenic contaminants, are easily grown during the storage of traditional Chinese medicines, posing a serious threat to medication safety. Current mainstream detection methods, such as high-performance liquid chromatography (HPLC), require complex pretreatment (solid-phase extraction, derivatization, etc.), take several hours, and rely on large laboratory equipment, making them difficult to meet the needs of rapid on-site testing. While surface-enhanced Raman spectroscopy (SERS) has shown promise in rapid testing in recent years, its practical application remains limited by the following bottlenecks:

[0003] Insufficient sensitivity and specificity: Traditional gold / silver nanostructured substrates have low enhancement factors (usually <10 6 ), and lack of targeted recognition mechanism, it is easily interfered by coexisting components such as flavonoids and polysaccharides in Chinese medicinal materials, resulting in missed detection of trace AFs (<1ppb);

[0004] Complex matrix interference: Conventional spectral analysis methods have difficulty distinguishing AFs characteristic peaks from background signals, especially when multiple components coexist, resulting in severe peak overlap and significant quantitative errors;

[0005] Poor device portability: Existing fiber optic probes are bulky and cannot achieve in-situ detection of herbal morphology adaptation. Furthermore, multi-channel spectral acquisition is inefficient, making it difficult to simultaneously analyze multiple AFs subtypes.

[0006] Lack of data credibility: The test results lack a tamper-proof evidence storage mechanism, making it difficult to meet the traceability needs of cross-border trade.

[0007] To address the above issues, there is an urgent need to develop an innovative system that integrates highly sensitive SERS substrates, intelligent spectral analysis, and portable detection to achieve on-site accurate and rapid detection and reliable traceability of multiple AFs in traditional Chinese medicines. Summary of the Invention

[0008] The main purpose of the present invention is to provide a fiber optic Raman spectroscopy enhanced rapid detection system for aflatoxin in traditional Chinese medicine, which can effectively solve the problems mentioned in the background technology.

[0009] To achieve the above object, the technical solution adopted by the present invention is:

[0010] The fiber-optic Raman spectroscopy enhanced rapid detection system for aflatoxin in traditional Chinese medicines includes: an optical excitation module, a sampling probe module, a signal enhancement module, a spectrum acquisition module, and an intelligent analysis module; among which:

[0011] The signal enhancement module includes a dynamic adaptive SERS substrate composed of a MXene-Au@SiO2 core-shell nanoflower structure. The surface of the MXene two-dimensional material is modified with aflatoxin aptamer probes, and hotspot dynamic regulation is achieved through pH-responsive hydrogel.

[0012] The sampling probe module uses a 7-channel spiral integrated photonic crystal fiber array, including a central excitation fiber and six collection fibers arranged at golden section angles. The excitation fiber is a hollow photonic bandgap fiber, and the end of the collection fiber is integrated with a micro filter array.

[0013] The intelligent analysis module has a built-in dual-channel deep residual network (DRN-SERS) algorithm, which includes a background spectrum simulation channel based on the adversarial generative network (GAN) and a feature peak analysis channel based on the attention mechanism.

[0014] The dynamic adaptive SERS substrate achieves specific recognition of AFs through aptamer probes, and combined with the spatially resolved light field control of the photonic crystal fiber array, can directly obtain high signal-to-noise ratio signals on the surface of complex medicinal materials, avoiding the damage to the samples caused by traditional in vitro detection.

[0015] Preferably, the method for constructing the dynamic adaptive SERS substrate includes:

[0016] Using Ti3C2Tx MXene nanosheets as substrate, Au nanoflower clusters were grown on their surface by a seed growth method, with a nanopetal spacing of 2-5nm;

[0017] The surface of Au nanoflowers is coated with a mesoporous SiO2 shell with a pore diameter of 3-8nm and a shell thickness of 20-50nm;

[0018] A pH-responsive poly (N-isopropylacrylamide) (PNIPAM) hydrogel was immobilized on the surface of the SiO2 shell, and AFB1 / AFB2 / AFG1 / AFG2 aptamer probes were covalently linked to its interior;

[0019] When the pH drops to 5.0, the hydrogel shrinks by 65%-80%, forcing the target molecules into the plasmonic hotspot regions between the Au nanoflowers.

[0020] The MXene-Au@SiO2 core-shell structure selectively filters macromolecular interferents through the mesoporous shell, and its three-dimensional nanogaps form a uniform electromagnetic field distribution, significantly improving detection repeatability and reducing background interference.

[0021] Preferably, the specific structure of the 7-channel spiral integrated photonic crystal fiber array is:

[0022] The central excitation fiber is a hollow-core photonic bandgap fiber with a core diameter of 50 μm and a cladding containing a periodic array of air holes (hole spacing Λ = 2.1 μm, aperture d = 1.8 μm);

[0023] Six collecting optical fibers surround the central optical fiber with a golden section spiral angle of 61.8°. The diameter of each optical fiber is 200 μm and the numerical aperture NA is 0.22.

[0024] The probe end is integrated with a PDMS microfluidic chip, which includes an injection channel (500 μm in width), a SERS substrate reaction chamber (10 μL in volume), and a waste liquid pool. The chip surface is equipped with a serpentine mixing microstructure.

[0025] The overall diameter of the probe is ≤3mm, the bending radius is ≥5mm, and the working distance is adjustable from 2-10mm.

[0026] The collection optical fibers arranged at the golden section angle optimize the Raman scattered light collection efficiency, and the serpentine mixing structure of the integrated microfluidic chip prolongs the contact time between the sample and the substrate, ensuring the effective capture of low-concentration AFs.

[0027] Preferably, the implementation of the dual-channel deep residual network (DRN-SERS) algorithm includes:

[0028] The background simulation part of channel 1 uses a generative adversarial network (GAN) to construct a Chinese medicinal material matrix spectrum library. The generator consists of a 5-layer fully connected network. The input is the medicinal material type, water content, and polysaccharide / alkaloid content parameters, and the output is the corresponding background spectrum.

[0029] The feature parsing part of channel 2 uses the attention mechanism residual block, focusing on 550cm -1 (coumarin ring deformation vibration), 780cm -1 (cyclopentanone C=O stretching vibration), 1250cm -1 (Bisfuran ring vibration) characteristic peak corresponds to the target molecular vibration mode, and the residual block cascade depth is 15 layers;

[0030] The network weights are dynamically updated through the transfer learning framework, supporting rapid adaptation after users upload ≤100 sets of new medicinal material spectral data, with a training time of ≤5 minutes.

[0031] The dual-channel network separates background interference from target signals through physical mechanisms, improving the distinction between overlapping AFB1 / AFB2 characteristic peaks and enhancing the accuracy of quantitative analysis.

[0032] Preferably, the optical excitation module comprises:

[0033] 785nm semiconductor laser (output power 0-500mW continuously adjustable) equipped with an acousto-optic tunable filter (AOTF) and a power density closed-loop controller;

[0034] The laser transmission optical path is connected by a polarization-maintaining fiber, and the end is coupled to the central excitation fiber through a gradient refractive index lens (GRIN lens).

[0035] The power density closed-loop control module dynamically adjusts the laser power according to the color of the medicinal materials to avoid sample carbonization and achieve non-destructive testing.

[0036] Preferably, the spectrum acquisition module includes:

[0037] Echelle grating spectrometer: spectral range 200-2000cm -1 , resolution 0.5cm -1 , equipped with deep-cooled EMCCD detector: cooling temperature -70℃, quantum efficiency ≥90%@600-800nm;

[0038] The output ends of the six collecting fibers are connected to the spectrometer entrance through a fiber bundle-slit coupler, with a spatial resolution of 8 spectra / mm.

[0039] The deep-cooled EMCCD detector combined with high spatial resolution design supports simultaneous analysis of fingerprint peak differences of multiple AFs in a single acquisition, ensuring the reliability of multi-component detection.

[0040] Preferably, the signal enhancement module further comprises:

[0041] A micro electroosmotic pump (flow rate 0.1-10 μL / min adjustable) drives the sample solution to circulate in the microfluidic chamber;

[0042] Pulsed laser activation unit: wavelength 532nm, pulse width 10ms, single pulse energy 5mJ, used for in situ excitation of SERS substrate hot spots;

[0043] Temperature controlled stage: temperature range 4-40℃, accuracy ±0.5℃, reaction temperature is adjusted by Peltier element.

[0044] The pulsed laser activation strategy maintains the stability of the substrate hotspot, and combined with a micro-electrosmotic pump to drive sample circulation, the enrichment efficiency of low-concentration AFs is improved.

[0045] Preferably, the intelligent analysis module further integrates:

[0046] Blockchain spectral database, using IPFS + Ethereum dual-chain structure to store spectral fingerprints and test results;

[0047] 5G communication unit, supports real-time upload of detection data to the cloud and generates blockchain evidence;

[0048] The embedded touch screen (resolution 1920×1080) displays AFs three-dimensional imaging (spatial resolution 50μm) and quantitative results (B1 / B2 / G1 / G2 separation ≥1.5).

[0049] The blockchain spectral database generates a unique hash fingerprint for each batch of medicinal materials, supporting regulatory authorities to achieve reliable traceability of production and storage records through spectral maps.

[0050] Preferably, the detection method of the system comprises the following steps:

[0051] S1. The medicinal material powder was subjected to microwave-assisted extraction (50% ethanol solution, 80°C for 30 seconds) to obtain the test solution;

[0052] S2: The microfluidic chip automatically executes the "injection-enrichment-detection-cleaning" cycle. The electroosmotic pump drives the test solution to flow through the SERS substrate, and the aptamer probe specifically captures the AFs molecules.

[0053] S3, adjusting the pH to 5.0 to shrink the hydrogel, triggering local plasmon resonance with a pulsed laser, and synchronously collecting multi-channel Raman signals;

[0054] S4. The DRN-SERS algorithm removes background interference and outputs the concentration values ​​and confidence levels (RSD ≤ 3%) of the four AFs.

[0055] The microwave-assisted extraction step significantly shortens the sample pre-processing time, and the DRN-SERS algorithm quickly adapts to new medicinal materials through transfer learning, reducing the complexity of model training.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] The present invention realizes the specific capture and signal enhancement of aflatoxin molecules through a dynamically adaptive MXene-Au@SiO2 core-shell SERS substrate. Combining a multi-channel photonic crystal fiber probe with an AI spectral decoupling algorithm, it overcomes the problem of precise stripping of complex matrix interference from traditional Chinese medicines. It can quickly complete in-situ, non-destructive, and multi-component simultaneous detection of four AFs, combining high sensitivity, portability, and intelligence, providing a fast and accurate integrated solution for the safety supervision of traditional Chinese medicines. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 Schematic diagram of the detection system flow of the present invention. DETAILED DESCRIPTION

[0059] In order to make the technical means, creative features, objectives and effects achieved by the present invention easier to understand, the present invention is further described below in conjunction with specific implementation methods.

[0060] like Figure 1The process flow of the fiber optic Raman spectroscopy enhanced rapid detection system for aflatoxin in traditional Chinese medicine is shown, and is explained below in conjunction with a detailed embodiment.

[0061] 1. Sample preparation

[0062] Taking ginseng as an example, accurately weigh 10g of ginseng powder and place it in a microwave extraction container. Add 50mL of 50% ethanol solution. Place the container in a microwave extractor and set the parameters to 500W power, 80°C, and 30 seconds. Microwave-assisted extraction utilizes the thermal and non-thermal effects of microwaves to rapidly penetrate the solvent into the medicinal material, accelerating the dissolution of aflatoxins. After extraction, transfer the extract to a centrifuge tube and centrifuge at 5000 rpm for 10 minutes. The supernatant is then collected as the test solution.

[0063] 2. System Preparation

[0064] Optical excitation module: Turn on the 785nm semiconductor laser, set the output power to 300mW using the power adjustment knob, and set the acousto-optic tunable filter (AOTF) to a specific frequency to filter out a single wavelength of laser light. A power density closed-loop controller monitors the laser power in real time and automatically adjusts the power based on the color of the ginseng sample (ginseng powder is pale yellow) to prevent sample carbonization. A polarization-maintaining fiber connects the laser to the gradient refractive index lens (GRIN lens), ensuring stable laser transmission to the central excitation fiber.

[0065] Sampling probe module: Check the 7-channel helical integrated photonic crystal fiber array probe. The central excitation fiber has a core diameter of 50 μm, and the cladding has a periodic air hole array with a spacing of 2.1 μm and an aperture of 1.8 μm. Six collection fibers surround the central fiber at a 61.8° golden section helix angle, each with a diameter of 200 μm and a numerical aperture of 0.22. Connect the sampling channel, SERS substrate reaction chamber, and waste liquid reservoir of the PDMS microfluidic chip at the end of the probe in sequence, ensuring that all channels are unobstructed.

[0066] Signal enhancement module: Check the micro electroosmotic pump and set the flow rate to 5μL / min. Prepare a dynamic adaptive SERS substrate composed of a MXene-Au@SiO2 core-shell nanoflower structure. The construction process is as follows: Using Ti3C2TxMXene nanosheets as the substrate, chloroauric acid and seed growth reagents are added to the solution. The reaction is carried out under certain temperature and stirring conditions to generate Au nanoflower clusters on the MXene surface, and the spacing between nanopetals is controlled at 2-5nm. Next, a mesoporous SiO2 shell is coated on the surface of the Au nanoflower using a sol-gel method. The reaction conditions are controlled to have a pore size of 3-8nm and a shell thickness of 20-50nm. Then, a pH-responsive poly (N-isopropylacrylamide) (PNIPAM) hydrogel is fixed on the surface of the SiO2 shell, and the AFB1 / AFB2 / AFG1 / AFG2 aptamer probes are covalently linked. The temperature of the temperature-controlled stage was set to 30°C, and the pulsed laser activation unit was ready with a wavelength of 532 nm, a pulse width of 10 ms, and a single pulse energy of 5 mJ.

[0067] Spectral Acquisition Module: Turn on the échelle spectrometer and preheat for 15 minutes. Verify that the deep-cooled EMCCD detector is stable at -70°C and has a quantum efficiency of ≥90% in the 600-800 nm band. Connect the output ends of the six collection fibers to the fiber bundle-slit coupler at the spectrometer inlet, ensuring a spatial resolution of 8 spectra / mm.

[0068] Intelligent Analysis Module: The dual-channel deep residual network (DRN-SERS) algorithm is activated, and a pre-set Chinese medicinal material matrix spectral library (containing background spectral data for a variety of common medicinal materials) is loaded. The blockchain spectral database is initialized using the IPFS + Ethereum dual-chain structure, the 5G communication unit is connected to a stable network, and the embedded touch screen is powered on and calibrated.

[0069] 3. Testing process

[0070] The prepared test solution is slowly injected into the microfluidic chip's sampling channel. A microelectroosmotic pump drives the test solution through the microfluidic chamber at a flow rate of 5 μL / min, flowing through the SERS substrate reaction chamber. During this process, the aptamer probe specifically recognizes and captures the AFs molecules in the test solution. For example, if aflatoxin B1 (AFB1) is present in ginseng, the aptamer probe will specifically bind to AFB1.

[0071] By adding acidic buffer solution, the pH of the solution in the SERS substrate reaction chamber is adjusted to 5.0, and the pH-responsive poly (N-isopropylacrylamide) (PNIPAM) hydrogel shrinks with a shrinkage rate between 65% and 80%, forcing the captured AFs molecules into the plasma hotspot area between the gaps of the Au nanoflowers.

[0072] The pulse laser activation unit is triggered, and a 532nm pulse laser is irradiated on the SERS substrate to stimulate local plasma resonance and enhance the Raman signal. At this time, 6 collection optical fibers simultaneously collect Raman scattered light and transmit it to the echelle grating spectrometer through the fiber bundle-slit coupler. The spectrometer is in the range of 200-2000cm -1 Spectral data are collected within the spectral range with a resolution of 0.5 cm -1 .

[0073] The collected spectral data is transmitted to the intelligent analysis module, and the dual-channel deep residual network (DRN-SERS) algorithm starts working. Channel 1 uses the generative adversarial network (GAN) to simulate the background spectrum of the ginseng matrix based on parameters such as the type of ginseng, water content (assuming that the water content of ginseng is 12% before detection), and polysaccharide / alkaloid content (ginseng polysaccharide content is about 3%-5%, and alkaloid content is about 0.05%-0.1%). Channel 2 uses the attention mechanism residual block to focus on 550cm -1 (coumarin ring deformation vibration), 780cm -1 (cyclopentanone C=O stretching vibration), 1250cm -1 The system analyzes the molecular vibration modes of the target compound corresponding to characteristic peaks (such as difuran ring vibrations) to identify the characteristic spectra of AFs. Using a transfer learning framework, the system can rapidly update network weights within 5 minutes based on the current ginseng sample spectral data, improving detection accuracy.

[0074] 4. Result Analysis and Report

[0075] The intelligent analysis module calculated the concentrations of AFB1, AFB2, AFG1, and AFG2 in the ginseng sample. Assuming the calculated results are 0.8 ppb for AFB1, 0.2 ppb for AFB2, undetectable for AFG1, and undetectable for AFG2, with relative standard deviations (RSDs) ≤ 3%, a confidence report was also generated, indicating a high degree of reliability.

[0076] The embedded touchscreen displays a 3D image of AFs with a spatial resolution of 50 μm, visually demonstrating the distribution of AFs (if any) in ginseng samples. Quantitative results are presented graphically with a B1 / B2 separation of ≥1.5 for quick reading.

[0077] The test data is uploaded to the cloud in real time via a 5G communication unit and stored in a blockchain spectral database, generating a unique hash fingerprint. Regulatory authorities or relevant personnel can use blockchain browsers to query the test records, origin, and storage information of a particular batch of ginseng based on the hash fingerprint, achieving trusted traceability.

[0078] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. The fiber-optic Raman spectroscopy enhanced rapid detection system for aflatoxin in traditional Chinese medicines is characterized by: include: Optical excitation module, sampling probe module, signal enhancement module, spectrum acquisition module and intelligent analysis module; among which: The signal enhancement module includes a dynamic adaptive SERS substrate composed of a MXene-Au@SiO2 core-shell nanoflower structure. The surface of the MXene two-dimensional material is modified with aflatoxin aptamer probes, and hotspot dynamic regulation is achieved through pH-responsive hydrogel. The sampling probe module uses a 7-channel spiral integrated photonic crystal fiber array, including a central excitation fiber and six collection fibers arranged at golden section angles. The excitation fiber is a hollow photonic bandgap fiber, and the end of the collection fiber is integrated with a micro filter array. The intelligent analysis module has a built-in dual-channel deep residual network algorithm, which includes a background spectrum simulation channel based on a generative adversarial network and a feature peak analysis channel based on an attention mechanism.

2. The fiber-optic Raman spectroscopy-enhanced rapid detection system for aflatoxin in traditional Chinese medicine according to claim 1, characterized in that: The method for constructing the dynamic adaptive SERS substrate includes: Using Ti3C2Tx MXene nanosheets as substrate, Au nanoflower clusters were grown on their surface by a seed growth method, with a nanopetal spacing of 2-5nm; The surface of Au nanoflowers is coated with a mesoporous SiO2 shell with a pore diameter of 3-8nm and a shell thickness of 20-50nm; A pH-responsive poly (N-isopropylacrylamide) hydrogel was immobilized on the surface of the SiO2 shell, and the AFB1 / AFB2 / AFG1 / AFG2 aptamer probes were covalently linked to the inside of the hydrogel. When the pH drops to 5.0, the hydrogel shrinks by 65%-80%, forcing the target molecules into the plasmonic hotspot regions between the Au nanoflowers.

3. The fiber-optic Raman spectroscopy-enhanced rapid detection system for aflatoxin in traditional Chinese medicine according to claim 1, characterized in that: The specific structure of the 7-channel spiral integrated photonic crystal fiber array is: The central excitation fiber is a hollow-core photonic bandgap fiber with a core diameter of 50 μm and a cladding containing a periodic array of air holes with a hole spacing of Λ = 2.1 μm and an aperture of d = 1.8 μm. Six collecting optical fibers surround the central optical fiber with a golden section spiral angle of 61.8°. The diameter of each optical fiber is 200 μm and the numerical aperture NA is 0.

22. The probe end is integrated with a PDMS microfluidic chip, which includes a 500μm wide sampling channel, a 10μL SERS substrate reaction chamber, and a waste liquid pool. The chip surface is equipped with a serpentine mixing microstructure. The overall diameter of the probe is ≤3mm, the bending radius is ≥5mm, and the working distance is adjustable from 2-10mm.

4. The fiber-optic Raman spectroscopy-enhanced rapid detection system for aflatoxin in traditional Chinese medicine according to claim 1, characterized in that: The implementation of the dual-channel deep residual network algorithm includes: The background simulation part of channel 1 uses a generative adversarial network to construct a Chinese medicinal material matrix spectrum library. The generator consists of a 5-layer fully connected network. The input is the medicinal material type, water content, and polysaccharide / alkaloid content parameters, and the output is the corresponding background spectrum. The feature parsing part of channel 2 uses the attention mechanism residual block, focusing on 550cm -1 , 780cm -1 , 1250cm -1 The characteristic peak corresponds to the target molecular vibration mode, and the residual block cascade depth is 15 layers; The network weights are dynamically updated through the transfer learning framework, supporting rapid adaptation after users upload ≤100 sets of new medicinal material spectral data, with a training time of ≤5 minutes.

5. The fiber-optic Raman spectroscopy-enhanced rapid detection system for aflatoxin in traditional Chinese medicine according to claim 1, characterized in that: The optical excitation module comprises: 785nm semiconductor laser, equipped with an acousto-optic tunable filter and a power density closed-loop controller; The laser transmission optical path is connected by a polarization-maintaining fiber, and the end is coupled to the central excitation fiber through a gradient refractive index lens.

6. The fiber-optic Raman spectroscopy-enhanced rapid detection system for aflatoxin in traditional Chinese medicine according to claim 1, characterized in that: The spectrum acquisition module includes: Echelle grating spectrometer: spectral range 200-2000cm -1 , resolution 0.5cm -1 , equipped with deep-cooled EMCCD detector: cooling temperature -70℃, quantum efficiency ≥90%@600-800nm; The output ends of the six collecting fibers are connected to the spectrometer entrance through a fiber bundle-slit coupler, with a spatial resolution of 8 spectra / mm.

7. The fiber-optic Raman spectroscopy-enhanced rapid detection system for aflatoxin in traditional Chinese medicine according to claim 2, characterized in that: The signal enhancement module further comprises: Micro electroosmotic pump: Flow rate is adjustable from 0.1 to 10 μL / min, driving the sample liquid to circulate in the microfluidic chamber; Pulsed laser activation unit: wavelength 532nm, pulse width 10ms, single pulse energy 5mJ, used for in situ excitation of SERS substrate hot spots; Temperature controlled stage: temperature range 4-40℃, accuracy ±0.5℃, reaction temperature is adjusted by Peltier element.

8. The fiber-optic Raman spectroscopy-enhanced rapid detection system for aflatoxin in traditional Chinese medicine according to claim 1, characterized in that: The intelligent analysis module also integrates: Blockchain spectral database, using IPFS + Ethereum dual-chain structure to store spectral fingerprints and test results; 5G communication unit, supports real-time upload of detection data to the cloud and generates blockchain evidence; Embedded touch screen displays AFs three-dimensional imaging and quantitative results.

9. The fiber-optic Raman spectroscopy-enhanced rapid detection system for aflatoxin in traditional Chinese medicine according to any one of claims 1 to 8, characterized in that: The detection method of the system includes the following steps: S1. Extract the medicinal material powder by microwave-assisted extraction to obtain the test solution; S2: The microfluidic chip automatically executes the "injection-enrichment-detection-cleaning" cycle. The electroosmotic pump drives the test solution to flow through the SERS substrate, and the aptamer probe specifically captures the AFs molecules. S3, adjusting the pH to 5.0 to shrink the hydrogel, triggering local plasmon resonance with a pulsed laser, and synchronously collecting multi-channel Raman signals; S4. The DRN-SERS algorithm removes background interference and outputs the concentration values ​​and confidence levels of the four AFs.

Images