CN112326622A - SIMCA-SVDD-based bacteria Raman spectrum identification and classification method - Google Patents
SIMCA-SVDD-based bacteria Raman spectrum identification and classification method Download PDFInfo
- Publication number
- CN112326622A CN112326622A CN202010480668.6A CN202010480668A CN112326622A CN 112326622 A CN112326622 A CN 112326622A CN 202010480668 A CN202010480668 A CN 202010480668A CN 112326622 A CN112326622 A CN 112326622A
- Authority
- CN
- China
- Prior art keywords
- simca
- svdd
- bacteria
- data
- raman spectrum
- 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.)
- Pending
Links
- 241000894006 Bacteria Species 0.000 title claims abstract description 47
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000001237 Raman spectrum Methods 0.000 title claims abstract description 44
- 230000001580 bacterial effect Effects 0.000 claims abstract description 28
- 238000001069 Raman spectroscopy Methods 0.000 claims abstract description 21
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 17
- 238000007781 pre-processing Methods 0.000 claims abstract description 6
- 238000012360 testing method Methods 0.000 claims abstract description 5
- 239000011159 matrix material Substances 0.000 claims description 15
- 238000012795 verification Methods 0.000 claims description 10
- 238000002835 absorbance Methods 0.000 claims description 8
- 230000003595 spectral effect Effects 0.000 claims description 8
- 239000013598 vector Substances 0.000 claims description 7
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000001228 spectrum Methods 0.000 claims description 5
- 238000012937 correction Methods 0.000 claims description 4
- 238000001134 F-test Methods 0.000 claims description 2
- 238000002790 cross-validation Methods 0.000 claims description 2
- 238000013499 data model Methods 0.000 claims description 2
- 238000005259 measurement Methods 0.000 claims description 2
- 238000006243 chemical reaction Methods 0.000 claims 1
- 238000003909 pattern recognition Methods 0.000 claims 1
- 241000193155 Clostridium botulinum Species 0.000 abstract description 15
- 238000004458 analytical method Methods 0.000 abstract description 6
- 241000193163 Clostridioides difficile Species 0.000 abstract description 2
- 241000607762 Shigella flexneri Species 0.000 abstract description 2
- 238000001514 detection method Methods 0.000 description 9
- 238000000513 principal component analysis Methods 0.000 description 5
- 239000000126 substance Substances 0.000 description 3
- 244000052616 bacterial pathogen Species 0.000 description 2
- 238000012569 chemometric method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 201000010099 disease Diseases 0.000 description 2
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 241000588724 Escherichia coli Species 0.000 description 1
- 240000001046 Lactobacillus acidophilus Species 0.000 description 1
- 235000013956 Lactobacillus acidophilus Nutrition 0.000 description 1
- 241000186673 Lactobacillus delbrueckii Species 0.000 description 1
- 238000001530 Raman microscopy Methods 0.000 description 1
- 241000194020 Streptococcus thermophilus Species 0.000 description 1
- 210000004666 bacterial spore Anatomy 0.000 description 1
- 238000003759 clinical diagnosis Methods 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 238000012136 culture method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000014670 detection of bacterium Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 235000013611 frozen food Nutrition 0.000 description 1
- 238000007901 in situ hybridization Methods 0.000 description 1
- 229940039695 lactobacillus acidophilus Drugs 0.000 description 1
- 238000004949 mass spectrometry Methods 0.000 description 1
- 244000005700 microbiome Species 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 231100000572 poisoning Toxicity 0.000 description 1
- 230000000607 poisoning effect Effects 0.000 description 1
- 238000003752 polymerase chain reaction Methods 0.000 description 1
- 235000013324 preserved food Nutrition 0.000 description 1
- 210000004215 spore Anatomy 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
- 235000013618 yogurt Nutrition 0.000 description 1
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
- G01N21/62—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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
- G06V20/695—Preprocessing, e.g. image segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/69—Microscopic objects, e.g. biological cells or cellular parts
- G06V20/698—Matching; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/03—Recognition of patterns in medical or anatomical images
Abstract
The invention discloses a bacteria Raman spectrum identification and classification method based on SIMCA-SVDD, and belongs to the field of bacteria Raman spectrum analysis. Firstly, preprocessing collected bacteria Raman data, inputting the preprocessed data into an SIMCA-SVDD algorithm for modeling, and then performing prediction classification through bacteria Raman test set data. In the SIMCA-SVDD algorithm, SVDD maps original data to a high-dimensional space, and an original nonlinear indivisible problem is converted into linear divisible in the high-dimensional space, so that a smaller division area range containing more target bacteria Raman spectrum sample information can be obtained, and a better classification result is obtained. And verifying the obtained bacterial Raman spectrum data of the four different types of bacteria. The classification accuracy of SIMCA and SIMCA-SVDD is 100% for colibacillus, Shigella flexneri and Clostridium difficile. However, for clostridium botulinum, the classification accuracy of SIMCA and SIMCA-SVDD is 86.7% and 93.3%, respectively, and it can be seen that SIMCA-SVDD has certain superiority for the identification of clostridium botulinum.
Description
Technical Field
The invention discloses a rapid identification and classification method for a bacterial Raman spectrum, and relates to the field of bacterial Raman spectrum analysis.
Background
Clostridium botulinum (Clostridium botulinum) is an anaerobic bacterium that is extremely susceptible to survive in a closed anaerobic environment. As an important pathogenic bacterium causing bacterial poisoning in the global range, clostridium botulinum has been a key monitoring subject in the field of disease control. Clostridium botulinum is widely used in various vacuum-packed foods including canned foods, frozen foods, and the like. Also, if clostridium botulinum is present in a water source, it will cause more extensive damage.
The field of bacterial identification and detection has created many new solutions to problems over the past decades, including fluorescence in situ hybridization, mass spectrometry, polymerase chain reaction, and the like. These methods consume time and labor costs, and are difficult to practically popularize. Therefore, the traditional bacterial culture method and animal experiment method still dominate the bacterial detection field. The conventional method also has its drawbacks: if the detection result can not realize the final strain identification, the detection accuracy is not high, and the like. The Raman spectrum analysis method enters the public visual field due to the advantages of simplicity, accuracy, rapidness and the like, and is a bacterium detection means with great potential.
The raman spectrum can be used as a molecular spectrum for analyzing the structural information of the composition of the substance. The use of raman analysis for the detection of microorganisms has first appeared in the last 90 th century. Gaus et al used UV resonance Raman technology to detect Lactobacillus acidophilus, Lactobacillus delbrueckii and Streptococcus thermophilus in yogurt. Xie et al obtained raw spectra of viable single bacterial spores using Raman techniques and used chemometric methods to classify six bacteria. It can be seen that raman spectroscopy can detect the spectral patterns of a single bacterium or spore. Therefore, micro-raman spectroscopy remains a common and effective tool in the field of bacterial detection. A common classification discriminant analysis method includes: principal Component Analysis (PCA), cluster-based Soft Independent Mode Classification (SIMCA), and the like.
The SIMCA method is used as a qualitative classification chemometric method, and is mature to be applied to analysis of various spectrograms. For the acquired bacterial raman data, the most commonly used classification method is the SIMCA classification. The sample boundary problem in the classification process by SIMCA remains a considerable problem to be discussed. How to draw a smaller and more accurate classification boundary, obtain a better classification result, and reduce the number of misclassified samples, the problem still needs to be solved urgently. The original SIMCA method uses euclidean distance to classify samples, and for the non-linearity problem encountered, the circle does not describe the boundary of the samples very accurately. Because the substances contained in the bacteria are similar, the spectrograms of the bacterial spectrums are relatively similar, and sample data classification per se has certain difficulty. Therefore, there is a need for improving the boundaries of the SIMCA original algorithm. The invention adds the support vector field to describe the SVDD, improves the division of the sample boundary in the SIMCA process, provides an improved method of the SIMCA-SVDD to carry out the rapid classification of the Raman spectrum of the bacteria, and provides a new solution for the rapid classification of the bacteria.
Disclosure of Invention
Because the structures of the bacteria are similar, the structures of the substances in the bacteria are also similar. The Raman spectrum spectra obtained by each type of bacteria have greater similarity, and the original SIMCA method is difficult to distinguish various types of bacteria. The invention provides a novel method for identifying and classifying bacteria Raman spectrum data based on SIMCA-SVDD, which is characterized in that: after modeling is carried out by adopting a PCA method in the SIMCA method, the target area is divided by adopting an SVDD principle instead of an original circular arc, and the rule of area division is changed, so that a better Raman data classification result is obtained.
The technical scheme adopted by the invention is as follows:
a method for rapid detection of bacteria, said method being useful for identifying bacteria Raman spectral data. The spectrometer used in the invention is a micro confocal Raman spectrometer, and the specific steps are as follows:
the technical scheme adopted by the invention for solving the technical problems is as follows: an electron microscopic image-based clostridium botulinum identification method aims to establish an online identification and classification system and method based on clostridium botulinum bacterial electron microscopic images.
The method mainly comprises the following steps:
step 1: and acquiring Raman spectrum data of various bacteria.
Step 2: and (3) carrying out data preprocessing on the obtained bacterial Raman spectrum data to eliminate the noise problem in the data.
And step 3: and inputting the obtained preprocessed Raman data into a SIMCA-SVDD algorithm for modeling.
And 4, step 4: inputting the data needing predictive classification into a trained SIMCA-SVDD model for predictive classification.
Further, the method for acquiring bacterial raman spectrum data in step 1 of the present invention specifically comprises the steps of:
assuming that the spectral data of the bacteria obtained from each measurement is yi=f(xi) The abscissa represents wavelength information in cm-1And the ordinate represents absorbance. The obtained bacteria raman spectrum data matrix is:wherein n is the total number of bacteria Raman spectrum samples.
Further, the bacterial raman data preprocessing process of step 2 of the present invention specifically includes the steps of:
the obtained bacterial raman spectral data Y was subjected to Standard Normal Transformation (SNV). It may reduce the noise of the spectral data. The calculation process is as follows:
wherein y iskiShowing the magnitude of absorbance of the sample at the kth row and the ith column,represents the average value of the absorbance of the line, SkThe variance of the absorbance of the sample in this row is indicated. The bacterial Raman spectrum data of each line is processed by SNV transformation, and Y matrix is processed to obtain YsnvAs input for the next step.
Further, the SIMCA-SVDD method in step 3 of the present invention specifically comprises the steps of:
for bacteria Raman spectrum data matrix Y subjected to data preprocessingsnvFirst, theUsing PCA, each type of sample matrix is decomposed into:
whereinT and P respectively represent a mean matrix, a score matrix and a load matrix. The number of principal components a was determined using cross-validation:
where E is the residual matrix. The statistic Q can be expressed as:
Q=1-PRESS/SS (5)
where PRESS is the sum of the squares of the prediction errors and SS is the sum of the squares of the residuals E. According to the selected main component A, Hotelling T2The calculation is as follows:
whereinIs calculated from the score vector and is,is tiaThe variance of (c). Calculation of T by F test2The critical value is obtained:
wherein v is a correction factor, eikResidual values representing the ith score value and the kth load value in the correction set. K is the number of load vectors, FcritIs the critical value for the F-test. Each sample can be calculatedAndthe values are used to describe sample characteristics. In the SIMCA algorithm, the samples are classified using the conventional euclidean distance:
in the SIMCA-SVDD algorithm, samples are classified not by the original Euclidean distance but by SVDD, and feature sets { x ] of n samples1,x2,...,xnUsing the center of sphere a and radius R, one class is represented as:
s.t.||xi-a||2≤R2+ξi,ξi≥0 (10)
where C is a penalty factor, ξiIs a relaxation factor. From the above equation, the lagrange function can be defined as:
wherein alpha isi(αi≥0),γi(γi> 0) is a Lagrangian multiplier. The center and radius R of the sphere a can be determined by solving the formula MaxMinL (R, a, xi)i,αi,γi) Solving, according to the formula, the following formula can be obtained:
substituting equations (12), (13), and (14) into equation (11) can yield:
the inner product in equation (15) is replaced with a gaussian kernel function such that L is maximized:
from equation (16) and defined C, α can be solved for each feature sample. The radius R can be calculated as:
where p is the support vector. Therefore, the relative distance of each bacteria Raman data in the Y matrix can be calculated and defined as
Employed in the final SVDD is the minimum value D in equation (18)iAs an index, the sample features are classified.
And 4, step 4: and inputting the data needing prediction classification into the trained SIMCA-SVDD model for prediction classification, and calculating the accuracy.
The further step 4 of the predictive classification process of the SIMCA-SVDD algorithm is as follows:
the bacteria Raman spectrum data of the verification set used by the verification algorithm also needs to be preprocessed by the data of the step 2, and D of each type of bacteria Raman data is marked in the step 3iThe limit values of (a) are plotted as an irregularity curve. Only the PCA process and the SVDD process in the step 3 are needed to calculate the D of each verification data set in each type of bacteria Raman spectrum data modeliEigenvalues, selection DiAnd taking the type of the bacteria model with the minimum characteristic value as the type of bacteria of the Raman spectrum data of the verification set.
The method can be applied to the clinical rapid pre-detection of the clostridium botulinum.
Has the advantages that:
1. the invention can reduce a large amount of time cost consumed by the traditional method for detecting bacteria, can meet the requirements of rapidness, convenience and accuracy of clinical pre-detection, and can effectively help doctors to diagnose.
2. The bacterial Raman spectrum test data used by the invention can be continuously increased according to clinical application, the accuracy can be effectively improved, and the establishment of a standard bacterial Raman spectrum database is promoted.
Drawings
FIG. 1 shows the procedure of step 1 for obtaining Raman spectrum data of bacteria.
FIG. 2 is a SIMCA-SVDD modeling process of the preprocessed bacterial Raman spectral data of step 3.
FIG. 3 is the SIMCA-SVDD predictive classification process of the test set bacterial Raman spectral data of step 4.
FIGS. 4 to 5 are a region division diagram of Clostridium botulinum of the original SIMCA algorithm and a region division diagram of Clostridium botulinum of the SIMCA-SVDD algorithm, respectively.
Table 1 shows the classification results of the SIMCA algorithm and the SIMCA-SVDD.
Detailed Description
The invention is explained in detail with reference to the content and embodiments of the drawings.
As shown in FIGS. 1 to 5, the present invention provides a method for identifying and classifying bacteria by Raman spectroscopy based on SIMCA-SVDD, and the present invention will be described in further detail below in order to make the objects, technical solutions and effects of the present invention clearer and clearer. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
As can be seen from fig. 4-5: the original SIMCA and SIMCA-SVDD methods achieve 100% accuracy in the identification and classification of Escherichia coli, Shigella flexneri and Clostridium difficile. However, when identifying and classifying Clostridium botulinum, the accuracy of SIMCA was 86.7% and the accuracy of SIMCA-SVDD was 93.3%. Therefore, the SIMCA-SVDD algorithm has certain advantages in identification and classification of the bacterial Raman spectrum data, and can better identify the Raman spectrum data of clostridium botulinum and provide help for clinical diagnosis.
It should be understood that the application of the present invention is not limited to the above-mentioned identification of the raman spectrum data of clostridium botulinum, and that other pathogenic bacteria identification is also true in the present invention for the field of disease control, and the design concept and concept of the system should fall within the scope of the appended claims.
TABLE 1 bacterial Raman Spectroscopy sample distribution
Claims (6)
1. A bacterial Raman spectrum identification and classification method based on SIMCA-SVDD can be used for classifying Raman spectrum data of bacteria with similar components, is based on SIMCA-SVDD, models each type of bacterial Raman spectrum data through PCA in SIMCA after the bacterial Raman spectrum data are obtained, performs region division by using SVDD rules, calculates the relative distance from verification data to each type of model, and classifies the bacterial Raman spectrum data used for verification.
2. The method for identifying and classifying Raman spectra of bacteria based on SIMCA-SVDD according to claim 1 wherein:
(1) and acquiring Raman spectrum data of various bacteria.
(2) And (3) carrying out data preprocessing on the obtained bacterial Raman spectrum data to eliminate the noise problem in the data.
(3) And inputting the obtained preprocessed Raman data into a SIMCA-SVDD algorithm for modeling.
(4) Inputting the data needing predictive classification into a trained SIMCA-SVDD model for predictive classification.
3. The process of acquiring data of raman spectra of bacteria according to claim 2, wherein:
assuming that the spectral data of the bacteria obtained from each measurement is yi=f(xi) The abscissa represents wavelength information in cm-1And the ordinate represents absorbance. The obtained bacteria raman spectrum data matrix is:wherein n is the total number of bacteria Raman spectrum samples.
4. The data pre-processing process of claim 2, wherein:
the acquired bacterial Raman spectrum data is subjected to SNV conversion, and noise of the spectrum data can be reduced. The calculation process is as follows:
5. The SIMCA-SVDD modeling process of claim 2, wherein:
SIMCA as a supervised pattern recognition algorithm, firstly PCA is adopted to decompose each type of sample matrix into:
whereinT and P respectively represent a mean matrix, a score matrix and a load matrix. The number of principal components a was determined using cross-validation:
where E is the residual matrix. The statistic Q can be expressed as:
Q=1-PRESS/SS (5)
where PRESS is the sum of the squares of the prediction errors and SS is the sum of the squares of the residuals E. According to the selected main component A, Hotelling T2The calculation is as follows:
whereinIs calculated from the score vector and is,is tiaThe variance of (c). Calculation of T by F test2The critical value is obtained:
wherein v is a correction factor, eikResidual values representing the ith score value and the kth load value in the correction set. K is the number of load vectors, FcritIs the critical value for the F-test. Each sample can be calculatedAndthe values are used to describe sample characteristics. In the SIMCA algorithm, the samples are classified using the conventional euclidean distance:
in the SIMCA-SVDD algorithm, samples are classified not by the original Euclidean distance but by SVDD, and feature sets { x ] of n samples1,x2,...,xnUsing the center of sphere a and radius R, one class is represented as:
s.t.||xi-a||2≤R2+ξi,ξi≥0 (10)
where C is a penalty factor, ξiIs a relaxation factor. From the above equation, the Lagrangian function can be defined as:
Wherein alpha isi(αi≥0),γi(γi> 0) is a Lagrangian multiplier. The center and radius R of the sphere a can be determined by solving the formula MaxMinL (R, a, xi)i,αi,γi) Solving, according to the formula, the following formula can be obtained:
substituting equations (12), (13), and (14) into equation (11) can yield:
the inner product in equation (15) is replaced with a gaussian kernel function such that L is maximized:
from equation (16) and defined C, α can be solved for each feature sample. The radius R can be calculated as:
where p is the support vector. In the multi-classification problem, the relative distance is defined as
Employed in the final SVDD is the minimum value D in equation (18)iAs an index, the sample features are classified.
6. The SIMCA-SVDD predictive classification process of claim 2, wherein:
the bacteria Raman spectrum data of the verification set used by the verification algorithm also needs to be preprocessed by the data of the step 2, and D of each type of bacteria Raman data is marked in the step 3iThe limit values of (a) are plotted as an irregularity curve. Only the PCA process and the SVDD process in the step 3 are needed to calculate the D of each verification data set in each type of bacteria Raman spectrum data modeliEigenvalues, selection DiAnd taking the type of the bacteria model with the minimum characteristic value as the type of bacteria of the Raman spectrum data of the verification set.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010480668.6A CN112326622A (en) | 2020-05-30 | 2020-05-30 | SIMCA-SVDD-based bacteria Raman spectrum identification and classification method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010480668.6A CN112326622A (en) | 2020-05-30 | 2020-05-30 | SIMCA-SVDD-based bacteria Raman spectrum identification and classification method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112326622A true CN112326622A (en) | 2021-02-05 |
Family
ID=74302908
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010480668.6A Pending CN112326622A (en) | 2020-05-30 | 2020-05-30 | SIMCA-SVDD-based bacteria Raman spectrum identification and classification method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112326622A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1820362A (en) * | 2003-05-16 | 2006-08-16 | 东京毅力科创株式会社 | Process system health index and method of using the same |
US20110242306A1 (en) * | 2008-12-19 | 2011-10-06 | The Johns Hopkins University | System and method for automated detection of age related macular degeneration and other retinal abnormalities |
US20120065899A1 (en) * | 2010-02-25 | 2012-03-15 | Olympus Corporation | Analysis method of vibrational spectra |
CN106127227A (en) * | 2016-06-14 | 2016-11-16 | 广西科技大学 | Use Raman spectrum that material carries out cluster analysis and know method for distinguishing |
CN107586823A (en) * | 2017-10-10 | 2018-01-16 | 徐州医科大学 | The rapid identification method of Carbapenems drug susceptibility based on Raman spectroscopy |
CN110243805A (en) * | 2019-07-30 | 2019-09-17 | 江南大学 | Fishbone detection method based on Raman high light spectrum image-forming technology |
-
2020
- 2020-05-30 CN CN202010480668.6A patent/CN112326622A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1820362A (en) * | 2003-05-16 | 2006-08-16 | 东京毅力科创株式会社 | Process system health index and method of using the same |
US20110242306A1 (en) * | 2008-12-19 | 2011-10-06 | The Johns Hopkins University | System and method for automated detection of age related macular degeneration and other retinal abnormalities |
US20120065899A1 (en) * | 2010-02-25 | 2012-03-15 | Olympus Corporation | Analysis method of vibrational spectra |
CN106127227A (en) * | 2016-06-14 | 2016-11-16 | 广西科技大学 | Use Raman spectrum that material carries out cluster analysis and know method for distinguishing |
CN107586823A (en) * | 2017-10-10 | 2018-01-16 | 徐州医科大学 | The rapid identification method of Carbapenems drug susceptibility based on Raman spectroscopy |
CN110243805A (en) * | 2019-07-30 | 2019-09-17 | 江南大学 | Fishbone detection method based on Raman high light spectrum image-forming technology |
Non-Patent Citations (6)
Title |
---|
FORINA M: "Multivariate range modeling, a new technique for multivariate class modeling:the uncertainty of the estimates of sensitivity and specificity", ANALYTICA CHIMICA ACTA * |
吴妍娴: "基于红外光谱SIMCA方法的研究及应用", 《中国优秀硕士学位论文全文数据库 (工程科技Ⅰ辑)》, pages 014 - 769 * |
王晓飞;王霄衣;史翔宇;阎秋静;陈向南;: "基于空间上下文单类分类器的目标检测算法", 红外与激光工程, no. 1, 25 December 2015 (2015-12-25) * |
赵杰文: "《茶叶质量与安全检测技术及分析方法》", 31 March 2001, 北京:中国轻工业出版社, pages: 102 - 103 * |
陈亮;谭红;谢峰;: "抹茶加工及其近红外检测中应用的研究进展", 食品工业, no. 05 * |
陈艳燕: "《单分类支持张量机模型及算法研究》", 30 September 2017, 中国农业大学出版社, pages: 20 - 21 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Yan et al. | Raman spectroscopy combined with machine learning for rapid detection of food-borne pathogens at the single-cell level | |
Dziuba et al. | Identification of lactic acid bacteria using FTIR spectroscopy and cluster analysis | |
Rothaermel et al. | The nanotech versus the biotech revolution: Sources of productivity in incumbent firm research | |
US8688384B2 (en) | Automated characterization and classification of microorganisms | |
Lecellier et al. | Differentiation and identification of filamentous fungi by high-throughput FTIR spectroscopic analysis of mycelia | |
Kang et al. | Single-cell classification of foodborne pathogens using hyperspectral microscope imaging coupled with deep learning frameworks | |
García-Timermans et al. | Label-free Raman characterization of bacteria calls for standardized procedures | |
Yu et al. | Classification of pathogens by Raman spectroscopy combined with generative adversarial networks | |
CN112712108B (en) | Raman spectrum multivariate data analysis method | |
Ren et al. | Using Raman spectroscopy and chemometrics to identify the growth phase of Lactobacillus casei Zhang during batch culture at the single-cell level | |
Kang et al. | Classification of foodborne bacteria using hyperspectral microscope imaging technology coupled with convolutional neural networks | |
Urtubia et al. | Using data mining techniques to predict industrial wine problem fermentations | |
Silge et al. | The application of UV resonance Raman spectroscopy for the differentiation of clinically relevant Candida species | |
Amiali et al. | Rapid identification of coagulase-negative staphylococci by Fourier transform infrared spectroscopy | |
Zhang et al. | Amplitude‐frequency images‐based ConvNet: Applications of fault detection and diagnosis in chemical processes | |
Gentner et al. | DBAM: Making virtual metrology/soft sensing with time series data scalable through deep learning | |
Li et al. | Accurate classification of Listeria species by MALDI-TOF mass spectrometry incorporating denoising autoencoder and machine learning | |
Zhang et al. | Classification and prediction of Klebsiella pneumoniae strains with different MLST allelic profiles via SERS spectral analysis | |
Wang et al. | Multi-point scanning confocal Raman spectroscopy for accurate identification of microorganisms at the single-cell level | |
CN112326622A (en) | SIMCA-SVDD-based bacteria Raman spectrum identification and classification method | |
Jamil et al. | Comparative Analysis on Machine Learning and One-Dimensional Convolutional Neural Network to Predict Surface Enhanced Raman Spectroscopy | |
Thompson et al. | ToF-SIMS studies of Bacillus using multivariate analysis with possible identification and taxonomic applications | |
Jarvis et al. | Rapid analysis of microbiological systems using SERS | |
Cai et al. | Deep metric learning framework combined with Gramian angular difference field image generation for Raman spectra classification based on a handheld Raman spectrometer | |
Zahn et al. | Discriminating cell line specific features of antibiotic‐resistant strains of Escherichia coli from Raman spectra via machine learning analysis |
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 |