CN112798529A - Novel coronavirus detection method and system based on enhanced Raman spectrum and neural network - Google Patents
Novel coronavirus detection method and system based on enhanced Raman spectrum and neural network Download PDFInfo
- Publication number
- CN112798529A CN112798529A CN202110006417.9A CN202110006417A CN112798529A CN 112798529 A CN112798529 A CN 112798529A CN 202110006417 A CN202110006417 A CN 202110006417A CN 112798529 A CN112798529 A CN 112798529A
- Authority
- CN
- China
- Prior art keywords
- data
- raman
- sample
- raman spectrum
- coronavirus
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
- G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
- G01N21/65—Raman scattering
- G01N21/658—Raman scattering enhancement Raman, e.g. surface plasmons
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Pathology (AREA)
- Chemical & Material Sciences (AREA)
- Biochemistry (AREA)
- Analytical Chemistry (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Immunology (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
The invention provides an integrated detection system comprising a virus spectrum database, a customized enhanced substrate, algorithm software and a mobile spectrometer. According to the invention, a nano porous gold reinforced substrate is adopted to carry out specific capture on new coronavirus molecules, so that the identification capability of virus Raman signals is improved; collecting a surface scanning Raman spectrum of a sample, and improving the signal-to-noise ratio through spectral clustering; randomly modulating the intensity of Raman spectrum data of the new coronavirus S protein in different states, and then superposing the negative Raman spectrum data to obtain a large amount of credible new coronavirus positive spectrum data with different concentrations; analyzing the spectral data of a sample to be detected by training a residual neural network model, realizing high-throughput processing of the data, intelligently judging whether the new coronavirus is negative or positive, and completing rapid and accurate detection of the sample; and a mobile detection and cloud platform scheme is also provided, so that the system can be deployed quickly and monitor epidemic situations in real time.
Description
Technical Field
The invention belongs to the field of virus detection, and particularly relates to novel rapid detection of coronavirus based on a Raman spectrum technology.
Background
The rapid, accurate and real-time detection of the new coronavirus (SARS-CoV-2) plays an important role in preventing and controlling the epidemic situation of the new coronavirus (COVID-19). At present, the detection of the new coronavirus (SARS-CoV-2) is mainly based on a real-time fluorescence polymerase chain reaction (RT-PCR) nucleic acid detection method, which is the 'gold standard' for the accurate diagnosis of the new coronavirus at present, but the detection steps are more, the whole detection process requires about 3-5 hours, and professional personnel and places are required, so that the portability and the rapidity are still obviously insufficient. In addition, as Chao Yan and the like report a new coronavirus nucleic acid detection method based on reverse transcription loop-mediated isothermal amplification (RT-LAMP), the detection time can be shortened to about 30 minutes, but the method is complex to operate, has higher biological safety risk and is difficult to realize large-scale deployment; qiuyuan Lin et al report that an immunological detection method based on enzyme-linked immunosorbent assay (ELISA) can detect new coronavirus IgG and IgM antibodies or antigens quickly, safely and in real time, but the technical sensitivity is low; and the antibody can be detected only after the antibody of an infected person appears, and the window period from the virus infection of the patient to the antibody appearance is from several days to several weeks, so the method for detecting the virus infection by the antibody has great risk of missed detection in the window period, and early warning of the infection is difficult to realize.
The enhanced raman scattering technique is widely used for biological detection and analysis due to its ultra-high sensitivity and spectral "fingerprint" characteristics. In the prior research, the raman spectroscopy technology is used for detecting a novel coronavirus (SARS-CoV-2), for example, patent CN11443072A discloses that a method for specifically detecting a novel coronavirus based on a novel corona receptor protein (ACE2) or an antibody modification enhancement layer can be used, but in an actual detection application scene, a pathological sample needs to be inactivated, the virus loses the attack capability on the receptor protein, namely, the enhancement layer cannot capture virus molecules to detect the virus actually; for another example, CN111650370A discloses a method and an apparatus for detecting SARS-CoV-2, which utilizes raman spectroscopy to quantitatively analyze the specific IgG antibody and IGM antibody of SARS-CoV-2 in a detection sample, and CN111665356A discloses a virus detection method based on SERS, which can be used for detecting and analyzing virus antibodies by using surface enhanced raman technique, but as mentioned above, there is detection omission in the window period of antibody detection, and the antibodies vary from person to person, and the false rate of antibody detection is high. In short, the existing research for detecting the novel coronavirus based on the raman spectrum focuses on improving a base for virus detection, the condition limitation of a real application scene is not considered, the detection experiment verification of a positive pathological sample of a real new coronavirus patient is lacked, the raman spectrum acquisition, processing, identification and judgment of the pathological sample of the novel coronavirus patient are not performed, and the practical research is particularly performed on the detection and identification under the condition of low virus concentration of an early infected person.
That is, although the enhanced raman spectroscopy has great potential to realize rapid and convenient detection of the novel coronavirus (SARS-GoV-2), it is still in the laboratory research stage, and there are still some key problems to be overcome due to its practical implementation:
firstly, an effective action area of a common Raman enhancement substrate is mainly concentrated in a narrow 'hot spot' area range 2-10 nm away from the surface, viruses belong to biological macromolecules, the size of the viruses is larger than 100nm, and groups entering the 'hot spot' area on the substrate have certain randomness, so that the enhanced Raman spectrum signal reproducibility is poor, and the negative/positive high-confidence-level judgment of a biological macromolecule sample is difficult to realize;
in the novel coronavirus detection, most of sample collection is throat swabs, nose swabs and the like, all are mixed object systems with complex components, the signal-to-noise ratio of the generated Raman spectrum is poor, accurate and rapid identification and judgment of trace virus molecular characteristic signals are realized, and the difficulty is great depending on the traditional spectrum collection peak searching and data analysis method;
thirdly, a large amount of real and credible Raman spectrum data of the novel coronavirus positive pathological sample is still lacked at present, and particularly the credible Raman spectrum data of the low-virus-concentration positive sample is lacked;
and fourthly, a rapid and accurate mobile field detection scheme aiming at the public places with densely-populated intersections such as frontier inspection ports, communities, schools, hotels, restaurants, bars and the like is lacked, and the full-process detection of sample collection, sample processing, spectrum collection and result analysis is realized on the field.
Disclosure of Invention
The invention provides a novel coronavirus full-flow detection method and system based on an enhanced Raman spectroscopy technology, which are rapid and high in accuracy. The specific scheme is as follows:
a novel rapid detection method for coronavirus based on enhanced Raman spectroscopy comprises the following steps:
s01, constructing a database:
negative raman spectral data: carrying out sample collection and sample preparation on people without infecting the novel coronavirus, carrying out surface scanning by using a Raman spectrometer, and then carrying out spectrum data preprocessing to obtain negative Raman spectrum data;
positive raman spectral data: respectively collecting surface scanning data of the purified novel coronavirus solid S protein and the purified novel coronavirus S protein solution with different concentrations in a concentration range of 10ppm-1000ppm by using a Raman spectrometer, carrying out random modulation on an intensity range [0.1, 1] after data preprocessing, and linearly superposing the data with the obtained negative Raman spectrum data to obtain positive Raman spectrum data;
s02, model training: randomly dividing Raman spectrum data in a database into a training set and a verification set; taking the training set as an input training neural network discrimination model, and carrying out verification by using a verification set;
s1, collecting, inactivating and preparing a sample to be detected: collecting and inactivating a sample of an object to be detected to prepare a sampling solution, and preparing the sample to be detected by using a Raman detection enhanced substrate;
s2, collecting Raman spectra: collecting at least 1 surface scanning enhanced Raman spectrum data of the sample to be detected by using a Raman spectrometer as Raman spectrum data to be processed of the sample to be detected;
s3, preprocessing Raman spectrum data and clustering spectrums: performing data preprocessing on the Raman spectrum data to be processed to obtain Raman spectrum data to be determined; then performing spectral clustering on the Raman spectrum to be judged to obtain class mean data to be judged;
s4, judging a sample to be detected: scoring the Raman spectrum data to be judged and the class mean data to be judged by using the neural network judgment model trained in the step S02, wherein the score is positive when the score is more than or equal to 0, and the score is negative when the score is less than 0; and when the data positive number is more than or equal to 3 in the Raman spectrum data to be judged and the class mean value data to be judged of a sample to be detected, judging that the novel coronavirus is positive, otherwise, judging that the novel coronavirus is negative.
Further, the surface scanning spatial resolution of the Raman spectrometer is less than 2 μm, and the surface scanning range is not less than 50 μm multiplied by 50 μm; laser wavelength of 785nm, power of 100mW or more, and spectral range of 50cm-1～4000cm-1Spectral resolution < 3cm-1The detector pixels are not lower than 1024 × 256.
Further, the sample collected in step S01 is at least one of sputum, throat swab, nose swab, and saliva; the sample collected in step S1 is one of sputum, throat swab, nasal swab, and saliva.
Further, the positive raman spectrum data in step S01 further includes: carrying out surface scanning enhanced Raman spectroscopy data on the novel coronavirus S protein solution with different concentrations in the concentration range of 1ppb-1000ppb and surface scanning Raman spectroscopy on the inactivated novel coronavirus S protein solution with different concentrations, carrying out random modulation on the intensity range [0.1, 1] after data preprocessing, and then linearly superposing the intensity range and the obtained negative Raman spectroscopy data to obtain Raman spectroscopy data.
Further, the step S01 is the same as the data preprocessing method in the step S3, including filtering smoothing, baseline deduction and normalization. The neural network model in the step S02 is a residual neural network model, and includes a one-dimensional convolutional layer, a pooling layer, a residual block, and a full-link layer; the residual block is a fourth order residual block.
Further, the raman detection enhancement substrate in step S1 employs nanoporous gold, the surface of the raman detection enhancement substrate is uniformly distributed with nanoporous gold clusters and nano channels between clusters, and the size of the nano channels is 60nm to 1200 nm.
Further, in step S2, 1250 pieces of spectral data in total are acquired for each sample to be detected by collecting 2 surface-scanning raman spectral data.
A novel rapid detection system for coronavirus based on enhanced Raman spectroscopy comprises
At least one test platform comprising
The Raman spectrometer has the surface scanning spatial resolution less than 2 μm and the surface scanning range not less than 50 μm multiplied by 50 μm; laser wavelength of 785nm, power of 100mW or more, and spectral range of 50cm-1～4000cm-1Spectral resolution < 3cm-1The pixel of the detector is not lower than 1024 multiplied by 256;
the Raman detection enhancement substrate is used for bearing a sample to be detected and enhancing Raman spectrum signals;
novel coronavirus Raman spectrum database, including
Negative raman spectral data: collecting samples of people without infecting the novel coronavirus, performing area scanning by using the Raman spectrometer, and performing data preprocessing to obtain negative Raman spectrum data;
positive raman spectral data: respectively acquiring the novel coronavirus purified solid S protein and the surface scanning data of the purified novel coronavirus S protein solution with different concentrations in the concentration range of 10ppm-1000ppm by using the Raman spectrometer, performing data preprocessing, performing random modulation on the data intensity range [0.1, 1], and linearly superposing the data with the obtained negative Raman spectrum data to obtain positive Raman spectrum data;
a data processing module comprising
The preprocessing module is used for preprocessing the Raman spectrum data of the sample to be detected, which is acquired by the Raman spectrometer, so as to generate the Raman spectrum data to be judged;
the spectral clustering module is used for performing spectral clustering on the Raman spectral data to be judged to obtain mean value data to be judged of each cluster;
the judgment module is used for scoring the Raman spectrum data to be judged and the mean value data to be judged by adopting a residual error neural network judgment model, wherein the score is positive when the score is more than or equal to 0, and the score is negative when the score is less than 0; finally, judging whether the novel coronavirus is positive or negative according to a judgment rule, and outputting a judgment result;
the judgment rule is as follows: and judging that the novel coronavirus is positive when the positive number of the data in the Raman spectrum data to be judged and the mean value data to be judged of one sample to be detected is more than or equal to 3, otherwise, judging that the novel coronavirus is negative.
Furthermore, the Raman detection enhancement substrate adopts nano-porous gold, nano-porous gold clusters and nano-channels among the clusters are uniformly distributed on the surface of the Raman detection enhancement substrate, and the size of the nano-channels is 60 nm-1200 nm.
Further, the at least one detection platform comprises a vehicle-mounted mobile detection platform, and the raman spectrometer comprises a mobile raman spectrometer; the vehicle-mounted mobile detection platform further comprises a micro negative pressure system, a sampling port, a biological safety cabinet, power generation equipment, a refrigerator and sterilization equipment; the novel coronavirus Raman spectrum database and the data processing module are arranged on a remote platform, and the detection platform carries out data transmission with the remote platform in a remote communication mode.
The invention relates to a novel coronavirus rapid detection method and a novel coronavirus rapid detection system based on an enhanced Raman spectroscopy technology and a neural network, which have the following advantages:
1. the novel coronavirus negative Raman spectrum is linearly superposed after intensity random modulation is respectively carried out by utilizing Raman spectra of novel coronavirus solid S protein and S protein solutions with different concentrations to obtain positive Raman spectrum data, a large amount of real and credible novel coronavirus positive Raman spectrum data can be obtained, and particularly, the large amount of credible low virus concentration positive Raman spectrum data is used for model training and verification and is more beneficial to detection of early infectors under the condition of low virus concentration; the larger the training data volume is, the better the generalization performance of the model is;
2. the surface scanning spatial resolution is less than 2 μm, the surface scanning range is 50 μm × 50 μm, the step length is 2 positive, and the spectral resolution is less than 3cm by using a Raman spectrometer-1Detector pixels 1024 × 256; at least 1 high-quality surface scanning data is collected for one sample to be detected, namely at least 625 Raman spectrum data of each sample to be detected; the method has the advantages that the scanning points are multiple, the scanning area is large, and the method is more beneficial to detecting a sample with low virus concentration;
3. the Raman detection enhancement substrate adopts nano porous gold, and the nano porous gold is provided with a nano channel, the size of the nano channel is matched with the size of the new coronavirus particles, so that the new coronavirus particles can be specifically captured, and the identification capability of Raman signals of the new coronavirus is improved;
4. spectral clustering processing is carried out on the Raman spectral data of the sample to be detected, and the intra-class mean data of the spectral clustering is also incorporated into neural network model analysis, so that noise generated randomly can be offset, and the spectral signal-to-noise ratio is improved;
5. a residual error neural network model is adopted to realize high-throughput processing and intelligent judgment of Raman spectrum data;
6. the detection limit of the novel coronavirus S protein of the detection method can be as low as 0.01ppb (10 ppb)-11Magnitude); the accuracy of the negative sample is 92.5 percent and the accuracy of the positive sample is 94.4 percent when the test is carried out on the real sample obtained from the city hospital of science in Sichuan province and the disease control center in Sichuan province; the detection method takes about 20 minutes in the whole process from sample collection to acquisition of a detection result of the novel coronavirus; that is to say, the detection method can have rapidity and accuracy, so that the rapid detection of the novel coronary virus based on the enhanced Raman technology really has practical application prospect;
7. the detection system also designs a mobile detection scheme, can form a Raman detection mobile laboratory in a vehicle-mounted mode, and has the capabilities of rapid deployment and field detection; aiming at the requirements of epidemic situation prevention and control real-time monitoring, a cloud platform scheme is also designed, a database and a discrimination algorithm are arranged at the cloud end, cloud discrimination is realized, the condition of virus carriers at each detection point can be mastered in real time, and the data processing and discrimination algorithm can be updated in time according to the virus variation condition, so that the virus detection quick response capability is formed.
Drawings
FIG. 1 is a schematic flow chart of the method for rapidly detecting a novel coronavirus according to the present invention;
FIG. 2 is a schematic diagram of a system for rapid detection of a coronavirus according to an embodiment of the present invention;
FIG. 3 illustrates a nanoporous gold enhanced substrate with nanochannel 60nm, as used in one embodiment of the invention;
FIG. 4 illustrates a nanoporous gold enhanced substrate with a nanochannel 1200nm, as used in one embodiment of the invention;
FIG. 5 is a schematic diagram of a mobile testing platform according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a cloud platform of the rapid coronavirus detection system according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of a residual neural network model used in an embodiment of the present invention;
FIG. 8 is a diagram illustrating a sample discrimination according to an embodiment of the present invention;
FIG. 9 shows the statistics of the detection results for real samples according to an embodiment of the present invention.
Detailed Description
The invention is explained in further detail below with reference to the figures and the embodiments.
The novel rapid detection method for coronavirus based on enhanced Raman spectroscopy, disclosed by the invention, has the simple process as shown in the attached figure 1, and comprises the following steps:
s01 construction database
Negative raman spectral data: carrying out sample collection and sample preparation on the crowd without infecting the novel coronavirus, carrying out area scanning by using an enhanced Raman spectrometer, and then carrying out spectrum data preprocessing to obtain negative Raman spectrum data; the collected sample is at least one of sputum, throat swab, nose swab and saliva.
Positive raman spectral data: respectively acquiring surface scanning data of the novel coronavirus purified solid S protein and the novel coronavirus purified S protein solution with different concentrations in the concentration range of 10ppm-1000ppm by using a Raman spectrometer, carrying out random modulation on the intensity range [0.1, 1] after data preprocessing, and linearly superposing the data with the obtained negative Raman spectrum data to obtain positive Raman spectrum data;
the positive raman spectral data may further include: carrying out surface scanning enhanced Raman spectrum data on novel S protein solutions of the coronary viruses with different concentrations in the concentration range of 1ppb-1000ppb, carrying out random modulation on the intensity range [0.1, 1] after data preprocessing, and then linearly superposing the obtained data with the obtained negative Raman spectrum data to obtain Raman spectrum data; and the inactivated novel coronavirus S protein solution with different concentrations is subjected to surface scanning Raman spectroscopy, and is subjected to pretreatment and random modulation in the intensity range [0.1, 1] and then is subjected to linear superposition of negative Raman spectroscopy data to obtain the spectrum data of the inactivated novel coronavirus S protein. Here the inactivation is usually done with heat inactivation at 56 ℃ for 30 minutes or with 75% alcohol inactivation; alcohol inactivation is more rapid.
The reason for selecting the novel coronavirus S protein as the detection object is as follows: the novel coronavirus particles are substantially spherical or medium polygonal in shape and have a diameter in the range of 80-160 nm. The novel coronavirus particle has hundreds of corollary projections, namely spike protein (S protein), distributed on the surface, the width of the top of the S protein is about 6nm, and the length is about 23 nm. Given the structure and size of the novel coronavirus, the S protein is the most likely object to be detected without opening the envelope. Raman spectra of S protein with different concentrations indicate that bands with higher similarity are mainly amide vibration bands of protein and main chain C-C, C-N skeleton vibration bands, and the bands reflect fingerprint information of S protein.
S02 model training
Randomly dividing the Raman spectrum data in the database constructed in the step S01 into a training set and a verification set; taking the training set as an input training neural network discrimination model, and carrying out verification by using a verification set;
the model is preferably a residual neural network model, and comprises a one-dimensional convolutional layer, a pooling layer, a residual block and a full-link layer. During training, data in the database are selected in a uniform random sampling mode for training, and the rest data are used as a verification set for verifying the training result of the neural network. The passed model is used for rapid detection of new coronavirus. And a residual error neural network model is adopted to realize high-throughput processing and intelligent judgment of Raman spectrum data.
S1, collecting, inactivating and preparing samples to be detected
Collecting and inactivating a sample of an object to be detected to prepare a sampling solution, and preparing the sample to be detected by using a Raman detection enhanced substrate;
collecting samples: the detection object samples applicable to the invention comprise samples which contain virus molecules and can be rapidly obtained, such as patient sputum, throat swabs, nasal swabs, saliva and the like. Taking a time-consuming throat swab sample as an example, the commercialized sampling kit comprises 1 test tube of sampling solution (the sampling solution contains various components such as bovine serum albumin, amino acid, gentamicin, cryoprotectant, fungal antibiotic, sodium chloride, magnesium sulfate, glucose, disodium hydrogen phosphate and the like, and is mainly used for storing virus nucleic acid) and 1 cotton swab, and the standard sampling process comprises 6 steps and can be completed within 2 minutes.
Inactivation treatment: the inactivation treatment is an essential step for virus detection, the detection object of the invention is virus surface S protein, and the chemical inactivation which takes short time can be adopted, namely, 75% ethanol is used for treating a sample for 5 minutes to destroy virus nucleic acid so as to inactivate the virus. Adding 75 μ L of 99.9% ethanol into the sample tube, dripping 25 μ L of throat swab solution, mixing, and standing for 5min to inactivate virus. In the practical application of the invention, a special sampling set is customized, and 75% ethanol is directly used as a sampling solution, so that the inactivation treatment only needs 5 minutes.
Preparing a sample to be detected: the Raman detection enhancement substrate used by the invention is nano-porous gold, and the enhancement substrate has a good signal enhancement effect on biomacromolecules. The surface of the Raman detection enhancement substrate is uniformly distributed with nano-porous gold clusters and nano-channels among the clusters, and the size of the nano-channels is 60 nm-1200 nm; a plurality of pores are distributed in the nano-porous gold cluster, and the size of each pore is 10-20 nm. Embedding the enhanced substrate into a quartz substrate groove, taking 2 mu L of sampling solution, dripping the sampling solution onto the substrate, waiting for about 5 minutes to completely volatilize the solution, and leaving the solute on the substrate to form a detection area. And finally, covering a quartz plate for sealing and packaging to ensure that the laser does not sputter the virus in the detection process, and marking corresponding information on the label to obtain a detection sample to be detected, wherein the time is about 6 minutes.
S2 Raman spectrum collection
Collecting at least 1 surface scanning Raman spectrum data of the sample to be detected by using a Raman spectrometer as Raman spectrum data to be processed of the sample to be detected;
the invention selects the mobile Raman spectrometer preferably, the scanning spatial resolution is less than 2 mu m, and the area scanning range is not limitedLess than 50 μm by 50 μm; laser wavelength of 785nm, power of 100mW or more, and spectral range of 50cm-1～4000cm-1Spectral resolution < 3cm-1The detector pixels are not lower than 1024 × 256.
And (3) placing a sample to be detected on a sample table of the Raman spectrometer, adjusting spectrometer parameters, and then performing Raman spectrum surface scanning data acquisition, wherein each surface scanning comprises 625 Raman spectra. A total of 1250 raman spectra per sample are preferably collected as raman spectral data for the sample to be tested for 2 planar scans, which takes about 5 minutes.
S3 Raman spectrum data preprocessing and spectrum clustering
Performing data preprocessing on the processed Raman spectrum data to obtain Raman spectrum data to be judged; the preprocessing method can be one or more of filtering smoothing, baseline deduction and normalization; preferably, the three algorithms are preprocessed sequentially.
Then performing spectral clustering on the Raman spectrum to be judged to obtain class mean data to be judged; spectral clustering can offset noise generated randomly and improve spectral signal-to-noise ratio.
This step is already automatically implemented by the algorithm software, taking about 1 minute.
S4, judging the sample to be detected
And (4) scoring the Raman spectrum data to be judged and the class mean data to be judged by using the neural network judgment model trained in the step S02, wherein the score interval of [ -1, +1], the closer to +1, the more obvious the spectrum positive characteristic is. Positive score when the score is greater than or equal to 0 and negative score when the score is less than 0; and when the data positive number is more than or equal to 3 in the Raman spectrum data to be determined and the class mean value data to be determined of a sample to be detected, determining that the novel coronavirus is positive, otherwise, determining that the novel coronavirus is negative. This step is already automatically implemented by the algorithm software, taking about 1 minute.
Therefore, the total time taken to test the entire procedure was about 20 minutes.
In a specific embodiment, the system for rapidly detecting a novel coronavirus based on enhanced raman spectroscopy of the present invention, as shown in fig. 2, includes at least one detection platform, a raman spectroscopy database of a novel coronavirus, and a data processing module.
The detection platform is provided with a Raman spectrometer and a Raman detection enhancement substrate.
The surface scanning spatial resolution of the Raman spectrometer is less than 2 μm, and the surface scanning range is not less than 50 μm multiplied by 50 μm; laser wavelength of 785nm, power of 100mW or more, and spectral range of 50cm-1～4000cm-1Spectral resolution < 3cm-1The pixel of the detector is not lower than 1024 multiplied by 256; the Raman spectrometer is used for collecting surface scanning Raman spectrum data of a sample to be detected. Usually, at least 1 surface scanning data (625 spectral data) is collected, and the total scanning points of each sample to be detected are many, the scanning area is large, and the detection of the sample with low virus concentration is facilitated. Preferably a mobile raman spectrometer.
And the Raman detection enhancement substrate is used for bearing the sample to be detected and enhancing Raman spectrum signals. The raman detection enhancing substrate is preferably nanoporous gold: a plurality of nano-porous gold clusters and nano-channels among the clusters are uniformly distributed on the surface of the enhanced substrate, and the size of each nano-channel is 60-1200 nm; a plurality of pores are distributed in the nano-porous gold cluster, and the size of each pore is 10-20 nm, as shown in the attached figure 3-4; the size of the nanometer channel is matched with the size of the virus particles, so that the new coronavirus particles can be captured specifically, and the identification capability of the new coronavirus Raman signals is improved.
A novel coronavirus raman spectroscopy database comprising negative raman spectroscopy data and positive raman spectroscopy data.
Negative raman spectral data: and collecting samples of people not infected with the novel coronavirus, performing area scanning by using the Raman spectrometer, and performing data preprocessing to obtain negative Raman spectrum data. The sample comprises at least one of sputum, throat swab, nasal swab, and saliva of the patient.
Positive raman spectral data: and respectively acquiring the surface scanning data of the purified novel coronavirus solid S protein and the novel coronavirus purified S protein solution with different concentrations in the concentration range of 10ppm-1000ppm by using the Raman spectrometer, performing data preprocessing, performing random modulation on the data intensity range [0.1, 1], and linearly superposing the data with the obtained negative Raman spectrum data to obtain positive Raman spectrum data.
The positive raman spectral data may further include: scanning enhanced Raman spectrum data of the novel coronavirus S protein solution with different concentrations in the concentration range of 1ppb to 1000ppb, carrying out random modulation on the intensity range [0.1, 1] after data preprocessing, and then linearly superposing the intensity range and the obtained negative Raman spectrum data to obtain Raman spectrum data; and the inactivated novel coronavirus S protein solution with different concentrations is subjected to surface scanning Raman spectroscopy, pretreated and randomly modulated in the intensity range of [0.1, 1], and then is linearly superposed with the negative Raman spectroscopy data to obtain the inactivated novel coronavirus S protein spectroscopy data. Here the inactivation is usually done with heat inactivation at 56 ℃ for 30 minutes or with 75% alcohol inactivation; alcohol inactivation is more rapid.
And the data processing module comprises a preprocessing module, a spectral clustering module and a judging module.
The preprocessing module is used for preprocessing the Raman spectrum data of the sample to be detected, which is acquired by the Raman spectrometer, so as to generate Raman spectrum data to be judged; the preprocessing can be one or more of filtering smoothing, baseline deduction and normalization; preferably, the three algorithms are preprocessed sequentially.
The spectral clustering module is used for performing spectral clustering on the Raman spectral data to be judged to obtain mean value data to be judged of each cluster; spectral clustering can offset noise generated randomly and improve spectral signal-to-noise ratio.
The judgment module is used for scoring the Raman spectrum data to be judged and the mean value data to be judged by adopting a residual error neural network judgment model, wherein the score is positive when the score is more than or equal to 0, and the score is negative when the score is less than 0; finally, judging whether the novel coronavirus is positive or negative according to a judgment rule, and outputting a judgment result; the judgment rule is as follows: and judging that the novel coronavirus is positive when the positive number of the data in the Raman spectrum data to be judged and the mean value data to be judged of one sample to be detected is more than or equal to 3, otherwise, judging that the novel coronavirus is negative.
Preferably, the at least one inspection platform comprises a vehicle-mounted mobile inspection platform, and the raman spectrometer comprises a mobile raman spectrometer; the vehicle-mounted mobile detection platform further comprises a micro negative pressure system, a sampling port, a biological safety cabinet, power generation equipment, a refrigerator and sterilization equipment, and is shown in the attached figure 5; the test platform may also include a data processing device, such as a computer or the like, for receiving, processing and transmitting test data associated with the platform. The novel coronavirus Raman spectrum database and the data processing module are arranged on a remote platform, and the detection platform performs data transmission with the remote platform in a remote communication mode, such as Ethernet, WiFi, 5G and the like, as shown in the attached drawing 6.
In a specific embodiment, 80 negative pharyngeal swab samples are obtained in the department of testing science city hospitals in Sichuan province, samples to be tested are manufactured, 2 surface scans are collected for each sample to be tested by using a Raman spectrometer, 160 surface scans (10 ten thousand Raman spectra) are obtained in total, and the samples are used as negative Raman spectrum training data after polynomial convolution smoothing, background deduction by a SNIP iterative method and normalization pretreatment.
Respectively obtaining Raman spectrum data and concentration range of purified novel coronavirus solid S protein by using Raman spectrometer
Raman spectrum data of 10ppm-1000ppm of different high-concentration novel coronavirus solid S protein solutions, enhanced Raman spectrum data of different low-concentration novel coronavirus solid S protein solutions with a concentration range of 1ppb-1000ppb, and Raman spectrum data of inactivated novel coronavirus solid S protein solutions with different concentrations; after the Raman spectrum data are respectively subjected to polynomial convolution smoothing, background deduction by an SNIP iterative method and normalization pretreatment, the negative Raman spectrum training data are linearly superposed to generate 10 ten thousand Raman spectra as positive training data.
The discriminant model selects a residual neural network model, as shown in fig. 7: using a convolution layer and a maximum pooling layer, then connecting with a four-order residual block, and finally connecting with a full-connection layer and outputting. Each order of residual block is composed of a convolution residual block and two identification residual blocks, each convolution residual block is composed of four convolution layers and a short connection from input to output, and each identification residual block is composed of three convolution layers and a short connection from input to output. And selecting 95% of data in the database as a training set and the rest as a verification set in a uniform random sampling mode. When the model is trained, a catam optimizer and a catam loss function are used, the batch processing size is 100, and the iteration number is as follows: 5 × 1805. As the number of iterations increases, the output error decreases and accuracy improves.
And finally, under the support of scientific city hospitals in Sichuan province and disease control centers in Sichuan province, 27 new coronavirus negative samples and 18 new coronavirus positive samples are detected (wherein the negative and positive samples are clinical diagnosis results of the hospitals and the disease control centers). And (3) carrying out alcohol inactivation on the collected samples, and then adopting a nano porous gold reinforced substrate to prepare 45 samples into samples to be detected. The method comprises the steps of placing a Raman detection sample on a sample table of a Raman spectrometer, focusing by using a 50 x telephoto lens, selecting a Mapping function, and collecting 1250 pieces of spectral data of 2 surface scanning data for each sample under the conditions of 785nm wavelength laser, 20mW laser power, 0.2s integration time, 600-1800 cm spectral wave number range, 50μm x 50μm surface scanning range and 2μm scanning step length.
Performing polynomial convolution smoothing, background subtraction by an SNIP iterative method and normalization pretreatment on the Raman spectrum data of the sample to be detected to obtain the Raman spectrum data to be determined; and then performing spectral clustering on the Raman spectrum to be determined of each sample to be detected, and obtaining various mean value data to be determined, wherein the Raman spectrum to be determined is clustered into 20 classes.
Scoring the Raman spectrum data to be judged and the class mean data to be judged by using a trained residual neural network judgment model, wherein each sample to be detected corresponds to 1250+20 and has 1270 scores, the score interval is between (-1 and + 1), the score is positive when the score is more than or equal to 0, and the score is negative when the score is less than 0; when the positive number is more than or equal to 3 in 1270 scores of a sample to be detected, judging the sample to be positive by the novel coronary virus, otherwise, judging the sample to be negative; as shown in FIG. 8, the number of positive in the sample to be tested is 6 > 3, and the sample is finally determined to be positive. The final test results of 45 samples are shown in FIG. 9. with the test method of the present invention, the accuracy of the negative sample is 92.5%, and the accuracy of the positive sample is 94.4%. The detection method takes about 20 minutes from sample collection to final result output in the whole process, has rapidness and accuracy, and has a very good practical application prospect. .
Claims (10)
1. A novel rapid detection method for coronavirus based on enhanced Raman spectroscopy comprises the following steps:
s01, constructing a database:
negative raman spectral data included: carrying out sample collection and sample preparation on the crowd without infecting the novel coronavirus, carrying out surface scanning by using a Raman spectrometer, and then carrying out spectrum data preprocessing to obtain negative Raman spectrum data;
positive raman spectral data included: respectively collecting surface scanning data of the purified novel coronavirus solid S protein and the purified novel coronavirus S protein solution with different concentrations in the concentration range of 10ppm-1000ppm by using a Raman spectrometer, carrying out random modulation on the intensity range [0.1, 1] after data preprocessing, and linearly superposing the data with the obtained negative Raman spectrum data to obtain positive Raman spectrum data;
s02, model training: randomly dividing Raman spectrum data in a database into a training set and a verification set; taking the training set as an input training neural network discrimination model, and carrying out verification by using a verification set;
s1, collecting, inactivating and preparing a sample to be detected: collecting and inactivating a sample of an object to be detected to prepare a sampling solution, and preparing the sample to be detected by using a Raman detection enhanced substrate;
s2, collecting Raman spectra: collecting at least 1 surface scanning enhanced Raman spectrum data of the sample to be detected by using a Raman spectrometer as Raman spectrum data to be processed of the sample to be detected;
s3, preprocessing Raman spectrum data and clustering spectrums: performing data preprocessing on the processed Raman spectrum data to obtain Raman spectrum data to be judged; then performing spectral clustering on the Raman spectrum to be judged to obtain class mean data to be judged;
s4, judging a sample to be detected: scoring the Raman spectrum data to be judged and the class mean data to be judged by using the neural network judgment model trained in the step S02, wherein the score is positive when the score is more than or equal to 0, and the score is negative when the score is less than 0; and when the data positive number is more than or equal to 3 in the Raman spectrum data to be determined and the class mean value data to be determined of a sample to be detected, determining that the novel coronavirus is positive, otherwise, determining that the novel coronavirus is negative.
2. The method for rapidly detecting a coronavirus according to claim 1, wherein: the surface scanning spatial resolution of the Raman spectrometer is less than 2 mu m, and the surface scanning range is not less than 50 mu m multiplied by 50 mu m; laser wavelength of 785nm, power of 100mW or more, and spectral range of 50cm-1～4000cm-1Spectral resolution < 3cm-1The detector pixels are not lower than 1024 × 256.
3. The method for rapidly detecting a coronavirus according to claim 1, wherein: the sample collected in the step S01 is at least one of sputum, throat swab, nose swab and saliva; the sample collected in step S1 is one of sputum, throat swab, nasal swab, and saliva.
4. The method for rapidly detecting a coronavirus according to claim 1, wherein: the positive raman spectrum data in step S01 further includes: the method comprises the steps of performing surface scanning enhanced Raman spectroscopy data on novel coronavirus S protein solutions with different concentrations and inactivated novel coronavirus S protein solutions with different concentrations within the concentration range of 1ppb-1000ppb, performing random modulation on the intensity range [0.1, 1] after data preprocessing, and performing linear superposition on the obtained negative Raman spectroscopy data to obtain Raman spectroscopy data.
5. The method for rapidly detecting a coronavirus according to claim 1, wherein: the step S01 is the same as the data preprocessing method in the step S3, including filtering smoothing, baseline subtraction and normalization; the neural network model in the step S02 is a residual neural network model, and includes a one-dimensional convolutional layer, a pooling layer, a residual block, and a full-link layer; the residual block is a fourth order residual block.
6. The method for rapidly detecting a coronavirus according to claim 1, wherein: the Raman detection enhancement substrate in the step S1 adopts nano-porous gold, nano-porous gold clusters and nano-channels among the clusters are uniformly distributed on the surface of the Raman detection enhancement substrate, and the size of the nano-channels is 60 nm-1200 nm.
7. The method for rapidly detecting a coronavirus according to claim 1, wherein: in step S2, 1250 pieces of spectral data in total are collected from 2 surface-scanning raman spectral data for each sample to be detected.
8. A novel rapid detection system for coronavirus based on enhanced Raman spectroscopy comprises
At least one test platform comprising
The Raman spectrometer has a surface scanning spatial resolution less than 2 μm, a surface scanning range not less than 50 μm × 50 μm, a laser wavelength 785nm, a power not less than 100mW, and a spectral range 50cm-1～4000cm-1Spectral resolution < 3cm-1The pixel of the detector is not lower than 1024 multiplied by 256;
the Raman detection enhancement substrate is used for bearing a sample to be detected and enhancing Raman spectrum signals;
novel coronavirus Raman spectrum database, including
Negative raman spectral data included: carrying out sample collection and sample preparation on people without infecting the novel coronavirus, carrying out surface scanning by using the Raman spectrometer, and then carrying out data preprocessing to obtain negative Raman spectrum data;
positive raman spectral data included: respectively acquiring the surface scanning data of the purified novel coronavirus solid S protein and the purified novel coronavirus S protein solution with different concentrations in the concentration range of 10ppm-1000ppm by using the Raman spectrometer, performing data preprocessing, performing random modulation on the data intensity range [0.1, 1], and linearly superposing the data with the obtained negative Raman spectrum data to obtain positive Raman spectrum data;
a data processing module comprising
The preprocessing module is used for preprocessing the Raman spectrum data of the sample to be detected, which is acquired by the Raman spectrometer, so as to generate the Raman spectrum data to be judged;
the spectral clustering module is used for performing spectral clustering on the Raman spectral data to be judged to obtain mean value data to be judged of each cluster;
the judgment module is used for scoring the Raman spectrum data to be judged and the mean value data to be judged by adopting a residual error neural network judgment model, wherein the score is positive when the score is more than or equal to 0, and the score is negative when the score is less than 0; finally, judging whether the novel coronavirus is positive or negative according to a judgment rule, and outputting a judgment result;
the judgment rule is as follows: and judging that the novel coronavirus is positive when the positive number of the data in the Raman spectrum data to be judged and the mean value data to be judged of one sample to be detected is more than or equal to 3, otherwise, judging that the novel coronavirus is negative.
9. The rapid coronavirus detection system according to claim 8, wherein: the Raman detection enhancement substrate adopts nano-porous gold, nano-porous gold clusters and nano-channels among the clusters are uniformly distributed on the surface of the Raman detection enhancement substrate, and the size of the nano-channels is 60 nm-1200 nm.
10. The rapid detection system for coronavirus according to claim 8 or 9, wherein: the at least one detection platform comprises a vehicle-mounted mobile detection platform, and the Raman spectrometer comprises a mobile Raman spectrometer; the vehicle-mounted mobile detection platform further comprises a micro negative pressure system, a sampling port, a biological safety cabinet, power generation equipment, a refrigerator and sterilization equipment; the novel coronavirus Raman spectrum database and the data processing module are arranged on a remote platform, and the detection platform performs data transmission with the remote platform in a remote communication mode.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110006417.9A CN112798529B (en) | 2021-01-04 | 2021-01-04 | Novel coronavirus detection method and system based on enhanced Raman spectrum and neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110006417.9A CN112798529B (en) | 2021-01-04 | 2021-01-04 | Novel coronavirus detection method and system based on enhanced Raman spectrum and neural network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112798529A true CN112798529A (en) | 2021-05-14 |
| CN112798529B CN112798529B (en) | 2022-05-10 |
Family
ID=75808064
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110006417.9A Active CN112798529B (en) | 2021-01-04 | 2021-01-04 | Novel coronavirus detection method and system based on enhanced Raman spectrum and neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112798529B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113252640A (en) * | 2021-06-03 | 2021-08-13 | 季华实验室 | Rapid virus screening and detecting method |
| CN114371160A (en) * | 2021-12-31 | 2022-04-19 | 中国工程物理研究院激光聚变研究中心 | Device for Raman detection of novel coronavirus |
| CN115753738A (en) * | 2022-12-05 | 2023-03-07 | 中国工程物理研究院激光聚变研究中心 | Pathological diagnosis method based on biomarker-enhanced Raman spectrum database |
| WO2023118947A1 (en) * | 2021-12-25 | 2023-06-29 | Ai Innobio Limited | System and method for classifying fluid substance with absorption spectroscopy |
| EP4382890A1 (en) | 2022-12-09 | 2024-06-12 | Instytut Chemii Fizycznej PAN | Method of sars-cov-2 virus detection using surface enhanced raman spectroscopy (sers) |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5553616A (en) * | 1993-11-30 | 1996-09-10 | Florida Institute Of Technology | Determination of concentrations of biological substances using raman spectroscopy and artificial neural network discriminator |
| WO2004099763A1 (en) * | 2003-05-12 | 2004-11-18 | Erasmus University Medical Center Rotterdam | Automated characterization and classification of microoganisms |
| WO2004109289A1 (en) * | 2003-06-10 | 2004-12-16 | Agency For Science, Technology And Research | Method of diagnosing sars corona virus infection |
| CA2668259A1 (en) * | 2008-06-16 | 2009-12-16 | The University Of Georgia Research Foundation, Inc. | Surface enhanced raman spectroscopy (sers) systems for the detection of bacteria and methods of use thereof |
| CN103645170A (en) * | 2013-12-03 | 2014-03-19 | 北京航空航天大学 | Device for detecting tumor characters by Raman spectrum |
| WO2018066587A1 (en) * | 2016-10-04 | 2018-04-12 | Atonarp Inc. | System and method for accurately quantifying composition of a target sample |
| CN108088834A (en) * | 2017-09-13 | 2018-05-29 | 新疆大学 | Echinococcosis serum Raman spectrum diagnostic equipment based on optimization reverse transmittance nerve network |
| CN109001180A (en) * | 2018-08-10 | 2018-12-14 | 青岛启明生物科技有限公司 | A kind of Raman spectrum combination artificial intelligence high throughput single cell analysis identification method |
| WO2019071076A1 (en) * | 2017-10-06 | 2019-04-11 | Lonza Ltd. | Automated control of cell culture using raman spectroscopy |
| WO2020087040A1 (en) * | 2018-10-26 | 2020-04-30 | Pataigin, Llc | Method for analyzing samples |
| CN111443072A (en) * | 2020-04-30 | 2020-07-24 | 苏州翊清环境科技有限公司 | Raman chip for virus detection, preparation method and virus rapid detection method |
| CN111665356A (en) * | 2020-05-29 | 2020-09-15 | 深圳网联光仪科技有限公司 | SERS-based virus detection method and device |
| CN111832477A (en) * | 2020-07-13 | 2020-10-27 | 华中科技大学 | Novel coronavirus detection method and system |
| CN112082957A (en) * | 2020-07-22 | 2020-12-15 | 北京理工大学 | Method for detecting monoclonal antibody and application thereof |
-
2021
- 2021-01-04 CN CN202110006417.9A patent/CN112798529B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5553616A (en) * | 1993-11-30 | 1996-09-10 | Florida Institute Of Technology | Determination of concentrations of biological substances using raman spectroscopy and artificial neural network discriminator |
| WO2004099763A1 (en) * | 2003-05-12 | 2004-11-18 | Erasmus University Medical Center Rotterdam | Automated characterization and classification of microoganisms |
| WO2004109289A1 (en) * | 2003-06-10 | 2004-12-16 | Agency For Science, Technology And Research | Method of diagnosing sars corona virus infection |
| CA2668259A1 (en) * | 2008-06-16 | 2009-12-16 | The University Of Georgia Research Foundation, Inc. | Surface enhanced raman spectroscopy (sers) systems for the detection of bacteria and methods of use thereof |
| CN103645170A (en) * | 2013-12-03 | 2014-03-19 | 北京航空航天大学 | Device for detecting tumor characters by Raman spectrum |
| WO2018066587A1 (en) * | 2016-10-04 | 2018-04-12 | Atonarp Inc. | System and method for accurately quantifying composition of a target sample |
| CN108088834A (en) * | 2017-09-13 | 2018-05-29 | 新疆大学 | Echinococcosis serum Raman spectrum diagnostic equipment based on optimization reverse transmittance nerve network |
| WO2019071076A1 (en) * | 2017-10-06 | 2019-04-11 | Lonza Ltd. | Automated control of cell culture using raman spectroscopy |
| CN109001180A (en) * | 2018-08-10 | 2018-12-14 | 青岛启明生物科技有限公司 | A kind of Raman spectrum combination artificial intelligence high throughput single cell analysis identification method |
| WO2020087040A1 (en) * | 2018-10-26 | 2020-04-30 | Pataigin, Llc | Method for analyzing samples |
| CN111443072A (en) * | 2020-04-30 | 2020-07-24 | 苏州翊清环境科技有限公司 | Raman chip for virus detection, preparation method and virus rapid detection method |
| CN111665356A (en) * | 2020-05-29 | 2020-09-15 | 深圳网联光仪科技有限公司 | SERS-based virus detection method and device |
| CN111832477A (en) * | 2020-07-13 | 2020-10-27 | 华中科技大学 | Novel coronavirus detection method and system |
| CN112082957A (en) * | 2020-07-22 | 2020-12-15 | 北京理工大学 | Method for detecting monoclonal antibody and application thereof |
Non-Patent Citations (3)
| Title |
|---|
| ALYSSON RONCALLY S. CARVALHO等: "COVID-19 chest computed tomography to stratify severity and disease extension by artificialneural network computer-aided diagnosis", 《FRONT MED(LAUSANNE)》 * |
| 杨凯 等: "基于拉曼光谱的快速细菌诊断", 《光散射学报》 * |
| 郭振东 等: "生物光谱技术在病原微生物检测中的应用进展", 《军事医学》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113252640A (en) * | 2021-06-03 | 2021-08-13 | 季华实验室 | Rapid virus screening and detecting method |
| WO2023118947A1 (en) * | 2021-12-25 | 2023-06-29 | Ai Innobio Limited | System and method for classifying fluid substance with absorption spectroscopy |
| CN114371160A (en) * | 2021-12-31 | 2022-04-19 | 中国工程物理研究院激光聚变研究中心 | Device for Raman detection of novel coronavirus |
| CN115753738A (en) * | 2022-12-05 | 2023-03-07 | 中国工程物理研究院激光聚变研究中心 | Pathological diagnosis method based on biomarker-enhanced Raman spectrum database |
| EP4382890A1 (en) | 2022-12-09 | 2024-06-12 | Instytut Chemii Fizycznej PAN | Method of sars-cov-2 virus detection using surface enhanced raman spectroscopy (sers) |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112798529B (en) | 2022-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112798529B (en) | Novel coronavirus detection method and system based on enhanced Raman spectrum and neural network | |
| CN109863239A (en) | The spectroscopy system and method for identification and quantification for pathogen | |
| US9170192B2 (en) | Systems and methods for identifying materials utilizing multivariate analysis techniques | |
| JP5597248B2 (en) | Biological and chemical collection and detection | |
| CN108088834B (en) | Echinococcosis serum Raman spectrum diagnostic apparatus based on optimized back propagation neural network | |
| EP1623212A1 (en) | Automated characterization and classification of microorganisms | |
| CN101356282A (en) | Method and apparatus for enhanced detection of toxic agents | |
| CN109266717A (en) | A kind of method and apparatus by single cell analysis detection bacterium drug resistance | |
| CN113252640B (en) | Rapid virus screening and detecting method | |
| CN113899702A (en) | Quantum Fourier transform-based vaccine multispectral rapid detection method | |
| CN111832477A (en) | Novel coronavirus detection method and system | |
| CN114324549B (en) | Pulmonary tuberculosis risk assessment method and system based on exhaled breath mass spectrometry detection | |
| CN114218985A (en) | Model system for classifying types of rejection resistance reactions in kidney transplantation and construction method thereof | |
| Mo et al. | Classification of coronavirus spike proteins by deep-learning-based Raman spectroscopy and its interpretative analysis | |
| CN112798678A (en) | Novel rapid detection method for coronavirus infection based on serum | |
| CN111766383A (en) | Method for detecting pathogenic virus and related protein thereof and predicting novel unknown virus | |
| CN113930526B (en) | Method and composition for identifying methamphetamine-related people and application of method and composition | |
| EP3933380A1 (en) | Method and system for characterizing particles using an angular detection in a flow cytometer | |
| Hussain et al. | Rapid identification of pathogens based on MIE light scattering and machine learning approach | |
| Yan et al. | Preliminary study on the quick detection of acquired immure deficiency syndrome by saliva analysis using surface enhanced Raman spectroscopic technique | |
| CN1664564A (en) | Method for quick-speed in-situ authentication of peripheral nerve tracts | |
| Sur et al. | Spectroscopy: A versatile sensing tool for cost-effective and rapid detection of novel coronavirus (COVID-19) | |
| JP2005055180A (en) | Bacterium identification device and method | |
| CN110874548A (en) | Lung cancer cell and normal cell recognition method based on combination of Raman spectrum and SVM | |
| Hadjigeorgiou et al. | Point-of-care diagnosis of urinary tract infection (UTI) using surface enhanced Raman spectroscopy (SERS) |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |