CN101419165A - Mequindox content detecting method in veterinary drugs - Google Patents
Mequindox content detecting method in veterinary drugs Download PDFInfo
- Publication number
- CN101419165A CN101419165A CNA2008102361456A CN200810236145A CN101419165A CN 101419165 A CN101419165 A CN 101419165A CN A2008102361456 A CNA2008102361456 A CN A2008102361456A CN 200810236145 A CN200810236145 A CN 200810236145A CN 101419165 A CN101419165 A CN 101419165A
- Authority
- CN
- China
- Prior art keywords
- mequindox
- spectrum
- sample
- content
- regression
- 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
Images
Abstract
The invention discloses a method for detecting the content of mequindox in animal drugs, which comprises the following steps: step one, the establishment of a sample spectrum is performed, wherein mequindox powder samples are collected, a spectrophotometer method is used to detect chemical values of the contents of mequindox powders in the samples, near-infrared diffuse reflection spectrums of the samples are acquired, and an average value is taken as the spectrum of each sample; step two, the preprocessing of the spectrum is performed, wherein a spectrum signal obtained in step one is preprocessed; step three, model establishment is performed, wherein a regression model is established, and the square (R<2>) of the chemical values and a predicted value correlation coefficient as well as predicted root mean square error (PRMSE) are taken as evaluation objects to preferably select a preprocessing method and the regression model with maximum R<2> and minimum PRMSE as a best preprocessing method and a best model; and step four, a diffuse reflection spectrum of a sample to be detected is acquired, then the spectrum preprocessing is performed, and the best regression model is applied to calculate to obtain the content of the mequindox in the sample to be detected. The method can realize a large quantity of quick and nondestructive detection to drugs, and has simple operation.
Description
Technical field
The invention belongs to the drug test technical field, the lossless detection method of mequindox content in particularly a kind of veterinary drug.
Background technology
Mequindox is the extensive pedigree antibiotic of synthetic, and colibacillosis, salmonellosis, pasteurellosis etc. are all had therapeutic action preferably.At present existing mequindox mixes the solid powder formulation of forming with starch, commodity mequindox powder by name or dysentery bacterium etc.Domestic animals such as chicken, ox cure the disease and diseases prevention in be widely used.
In recent years, quality problems often appear in the mequindox powder.Excessive concentration (reach indicatrix 185.5%) or low (reaching indicatrix content 43.8%) were once found the time in Ministry of Agriculture's sampling observation with excessivelying, and this economic animal that may cause taking medicine poisons or infection is not healed, and causes damage.
All inspect by random samples before the mequindox powder dispatches from the factory, the method for use is the spectrophotometer method of regulation in " Chinese veterinary drug standard " (version in 1992).But the shortcoming that this method exists is time-consuming, require great effort and consume certain chemical reagent.Usually need more than 30 minute detection time; Need use chemical reagent in the detection, therefore need to detect cost; Also produce waste liquid simultaneously, environment is produced pollute; Need consume a certain amount of mequindox powder during detection, therefore bring certain loss to product.Because human cost is higher now, enterprise is difficult to arrange more reviewer that the product of producing is carried out the check of vast scale; Also there are these problems in medicine prison department to the sampling observation of market circulation product equally, is difficult to strengthen the sampling observation ratio.Therefore press for a kind of quick, harmless detection method.
Summary of the invention
Long in order to solve the detection time that prior art exists, the method complexity can not be inspected by random samples in a large number, the shortcoming that medicine after the detection is impaired, the invention provides a kind of detection method of content of mequindox in animal drugs, can realize, Non-Destructive Testing a large amount of, quick to medicine, detection method is simple.
Technical scheme of the present invention is:
A kind of detection method of content of mequindox in animal drugs comprises the steps:
(1) foundation of sample spectra
Collecting mequindox powder sample, use the content chemical score of the middle mequindox powder of the spectrophotometer method test sample in " Chinese veterinary drug standard " (version in 1992), is to detect the gained content of material with chemical method; This research method measured result compares with it, and is approaching more, illustrates that the effect of detection method is good more; At temperature 10-40 ℃, relative humidity 25-50%, use near infrared spectrometer, resolution 1cm
-1-16cm
-1, scanning times 4-64, wave-number range 10000-3500cm
-1, the near-infrared diffuse reflection spectrum of collected specimens, each sample collecting three times is averaged as the spectrum of each sample.
(2) pre-service of spectrum
In the spectral signal except that sample composition information to be measured, also comprise various high frequency noises, baseline wander, parasitic light, sample background etc., so need pre-service, to weaken and to eliminate the influence of various non-target factors to spectral information, stable for setting up, reliable model lays the foundation; Among the present invention, adopt multiple preprocess method, comprise that methods such as wavelet transformation, level and smooth, differentiate, standard normal conversion, centralization or normalization carry out pre-service to spectrum, thereby reduce spectral noise, highlight useful information;
(3) set up model
Linearity in the Applied Chemometrics and nonlinear method multivariate calibration methods are set up regression model, the preferred best approach.Linear method comprises multiple regression, multiple stepwise regression, principal component regression, partial least squares regression; Non-linear regression comprises artificial neural network, support vector machine.Square (R2) and predicted root mean square error (PRMSE) with chemical score and predicted value related coefficient are evaluation object, optimize preprocess method with maximum R2 and minimum PRMSE and regression model as best preprocess method and optimum regression model;
(4) detect testing sample
At temperature 10-40 ℃, relative humidity 25-50%, usefulness near infrared spectrometer (resolution 1cm
-1-16cm
-1, scanning times 4-64, wave-number range 10000-3500cm
-1), gather testing sample spectrum, adopt then that best preprocess method carries out pre-service to testing sample spectrum in the step 2, the optimum regression Model Calculation obtains mequindox content in the sample to be checked in the applying step 3.
The present invention compared with prior art, the beneficial effect that has is:
(1) utilize near-infrared spectrum technique to measure mequindox powder content in the mequindox powder, its analysis speed is accelerated greatly, and the mensuration of spectrum and cubage can be finished in 1 minute;
(2) can analyze great amount of samples at short notice, save time, the sampling observation ratio for manufacturing enterprise can increase product helps improving the quality of products;
(3) this detection method is a kind of lossless detection method, and the sample that detected still can be sold use;
(4) need not use any chemical reagent, reduce the detection cost, environmentally friendly, can not produce pollutant in the testing process.
Description of drawings
The final spectrum of the sample that Fig. 1 is gathered for embodiment 1.
Fig. 2 is the pretreated sample spectra of embodiment 1.
Fig. 3 is the linear model that embodiment 1 usefulness partial least square method is set up.
Fig. 4 is the result that the partial least square method of embodiment 1 testing sample is set up.
The final spectrum of the sample that Fig. 5 is gathered for embodiment 2.
Fig. 6 is the pretreated sample spectra of embodiment 2.
Fig. 7 is the linear model that embodiment 2 usefulness partial least square method are set up.
Fig. 8 is the result that the partial least square method of embodiment 2 testing samples is set up.
Embodiment
Collect 38 parts in mequindox powder sample, be used for modeling.Detect mequindox content with spectrophotometer method (Chinese veterinary drug standard, version in 1992), content distribution is 8.5～22.75%wt.At 20 ℃, under the environment of relative humidity 36%, divide the diffuse reflection spectrum of the WQF-400N of instrument company type FFIR collected specimens, wave number sweep limit 10000～3500cm with Beijing Rayleigh
-1, spectral resolution 4cm
-1, light source 6W/6V halogen tungsten lamp, vulcanized lead (PbS) detector, scanning times 16.Gather spectrum respectively three times, be averaged final spectrum, as shown in Figure 1 as sample;
Using Simca p 11.5 software systems carries out pre-service and sets up model spectrum, through preferred, preprocess method is the standard normal conversion, (wavelet basis is bio2.6 to wavelet de-noising, three layers of decomposition, the filtering of self-adaptation soft-threshold) and 5: 1 rank differential, through pretreated spectrum as shown in Figure 2; Modeling method is a partial least square method, sets up linear model, and the result is better, as shown in Figure 3, and R
2=0.9963, PRMSE is 0.36%.
Collect 19 testing samples,, under the environment of relative humidity 30%, divide the diffuse reflection spectrum of the WQF-400N of instrument company type FFIR collected specimens, wave number sweep limit 10000～3500cm with Beijing Rayleigh at 25 ℃
-1, spectral resolution 4cm
-1, light source 6W/6V halogen tungsten lamp, vulcanized lead (PbS) detector, scanning times 16.Gather spectrum respectively three times, be averaged final spectrum as testing sample;
The pre-method standard normal transformation of optimization process, wavelet de-noising (wavelet basis is bio2.6, three layers of decomposition, the filtering of self-adaptation soft-threshold) and the 5: 1 rank differential determined before adopting are managed in advance to spectrum, adopt the best model partial least square model to calculate.The result who obtains then compares with the mequindox content that adopts spectrophotometer method (Chinese veterinary drug standard, version in 1992) detection to obtain, and effect is better, as shown in Figure 4, and R
2Be 0.9878, PRMSE is 0.68%, reaches required accuracy of detection requirement.
Implement 2
Collect 90 parts in mequindox powder sample, be used for modeling.Detect mequindox content with spectrophotometer method (Chinese veterinary drug standard, version in 1992), content distribution is 25.56～96.85%.At 20 ℃, under the environment of relative humidity 40%, divide the diffuse reflection spectrum of the WQF-400N of instrument company type FFIR collected specimens, wave number sweep limit 10000～3500cm with Beijing Rayleigh
-1, spectral resolution 8cm
-1, light source 6W/6V halogen tungsten lamp, vulcanized lead (PbS) detector, scanning times 64.Gather spectrum respectively three times, be averaged final spectrum, as shown in Figure 5 as sample;
On the Matlab7.0 software systems spectrum is carried out pre-service and sets up model, through preferred, preprocess method is that standard normal conversion, 5 SGF smoothly reach 5: 1 rank differential, and spectrum as shown in Figure 6 after pre-service; Modeling method is major component-support vector machine method, extracts preceding 7 major components, adopts support vector machine method to set up nonlinear model, and the result is better, R
2Be 0.9938, PRMSE is 0.87%, as shown in Figure 7.
Collect 48 testing samples,, under the environment of relative humidity 40%, divide the diffuse reflection spectrum of the WQF-400N of instrument company type FFIR collected specimens, wave number sweep limit 10000～3500cm with Beijing Rayleigh at 20 ℃
-1, spectral resolution 8cm
-1, light source 6W/6V halogen tungsten lamp, vulcanized lead (PbS) detector, scanning times 64.Gather spectrum respectively three times, be averaged final spectrum as testing sample;
As shown in Figure 8, adopt standard normal conversion, 5 SGF smoothly to reach 5: 1 rank differential spectrum is managed in advance, extract preceding 7 major components, adopt the supporting vector machine model of building to calculate, with adopt spectrophotometer method (Chinese veterinary drug standard before, version in 1992) detect the mequindox content that obtains and compare, effect is better, R
2Be 0.9881, PRMSE is 1.27%, reaches required accuracy of detection requirement.
Claims (3)
1. the detection method of a content of mequindox in animal drugs is characterized in that, step is:
The first step, the foundation of sample spectra: collect mequindox powder sample, use the content chemical score of the middle mequindox powder of spectrophotometer method test sample; Under temperature 10-40 ℃, the condition of relative humidity 25-50%, with the near-infrared diffuse reflection spectrum of near infrared spectrometer collected specimens, each sample collecting three times is averaged as the spectrum of each sample;
Second step, spectrum pre-service: adopt wavelet transformation, level and smooth, differentiate, standard normal conversion, centralization or method for normalizing that the spectral signal that the first step obtains is carried out pre-service;
In the 3rd step, set up model: linearity in the Applied Chemometrics and nonlinear method or multivariate calibration methods, set up regression model, with square (R of chemical score and predicted value related coefficient
2) and predicted root mean square error (PRMSE) be evaluation object, optimize and have maximum R
2And the preprocess method of minimum PRMSE and regression model are as best preprocess method and best model;
The 4th step, testing sample detects: under temperature 10-40 ℃, the condition of relative humidity 25-50%, gather testing sample spectrum with near infrared spectrometer, adopt the best preprocess method optimize that testing sample spectrum is carried out pre-service then, use the content that optimum regression Model Calculation in the 3rd step obtains mequindox in the sample to be checked.
2. the detection method of content of mequindox in animal drugs as claimed in claim 1 is characterized in that the resolution of the near infrared spectrometer described in the first step and the 4th step is 1cm
-1-16cm
-1, scanning times is 4-64, wave-number range is 10000-3500cm
-1
3. the detection method of content of mequindox in animal drugs as claimed in claim 1 is characterized in that the linear method described in the 3rd step comprises multiple regression, multiple stepwise regression, principal component regression, partial least squares regression; Nonlinear method comprises artificial neural network, support vector machine method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008102361456A CN101419165A (en) | 2008-11-17 | 2008-11-17 | Mequindox content detecting method in veterinary drugs |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2008102361456A CN101419165A (en) | 2008-11-17 | 2008-11-17 | Mequindox content detecting method in veterinary drugs |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101419165A true CN101419165A (en) | 2009-04-29 |
Family
ID=40630062
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2008102361456A Pending CN101419165A (en) | 2008-11-17 | 2008-11-17 | Mequindox content detecting method in veterinary drugs |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101419165A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103808688A (en) * | 2014-01-22 | 2014-05-21 | 重庆医科大学 | Rapid non-destructive detection on quality consistency of finished medicine product by using near-infrared spectroscopy |
| CN117368146A (en) * | 2023-12-08 | 2024-01-09 | 苏陀科技(北京)有限公司 | Rapid detection method for mycelium protein content |
-
2008
- 2008-11-17 CN CNA2008102361456A patent/CN101419165A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103808688A (en) * | 2014-01-22 | 2014-05-21 | 重庆医科大学 | Rapid non-destructive detection on quality consistency of finished medicine product by using near-infrared spectroscopy |
| CN117368146A (en) * | 2023-12-08 | 2024-01-09 | 苏陀科技(北京)有限公司 | Rapid detection method for mycelium protein content |
| CN117368146B (en) * | 2023-12-08 | 2024-03-12 | 苏陀科技(北京)有限公司 | Rapid detection method for mycelium protein content |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102879353B (en) | The method of content of protein components near infrared detection peanut | |
| Jin et al. | Near-infrared analysis of the chemical composition of rice straw | |
| Burgula et al. | Review of Mid‐infrared fourier transform‐infrared spectroscopy applications for bacterial detection | |
| Chen et al. | Rapid measurement of total acid content (TAC) in vinegar using near infrared spectroscopy based on efficient variables selection algorithm and nonlinear regression tools | |
| CN101393114B (en) | Rapid measuring method for physical property of flavouring essences for tobacco | |
| CN102590129B (en) | Method for detecting content of amino acid in peanuts by near infrared method | |
| CN103278473B (en) | The mensuration of pipering and moisture and method for evaluating quality in white pepper | |
| CN102636450A (en) | Method for detecting wolfberry polyose content in Chinese wolfberry in a nondestructive way based on near infrared spectrum technology | |
| CN101210875A (en) | Damage-free measurement method for soil nutrient content based on near infrared spectra technology | |
| CN101504363A (en) | Edible fatty acid value detection method based on near-infrared spectrum analysis | |
| Chia et al. | Neural network and principal component regression in non-destructive soluble solids content assessment: a comparison | |
| CN1831516A (en) | Method for nondistructive discriminating variety and true and false of cigarette using visible light and near-infrared spectrum technology | |
| CN104749132A (en) | Method for measuring content of azodicarbonamide in flour | |
| CN102788794A (en) | Device and method for detecting pesticide residues on leaves of leaf vegetables on basis of multi-sensed information fusion | |
| CN103411906A (en) | Near infrared spectrum qualitative identification method of pearl powder and shell powder | |
| CN103645155A (en) | Quick nondestructive testing method for tenderness of fresh mutton | |
| CN105092513A (en) | Method for determining content of drug components in mixture drug | |
| CN105784672A (en) | Drug detector standardization method based on dual-tree complex wavelet algorithm | |
| CN102128807A (en) | Method for quickly detecting concentration of droplet on crop leaf | |
| CN104596979A (en) | Method for measuring cellulose of reconstituted tobacco by virtue of near infrared reflectance spectroscopy technique | |
| Chen et al. | Rapid identification of three varieties of Chrysanthemum with near infrared spectroscopy | |
| CN104596976A (en) | Method for determining protein of paper-making reconstituted tobacco through ear infrared reflectance spectroscopy technique | |
| CN104596980A (en) | Method for measuring hot water solvends of reconstituted tobacco by paper-making process by virtue of near infrared reflectance spectroscopy technique | |
| CN101419165A (en) | Mequindox content detecting method in veterinary drugs | |
| CN102928356A (en) | Method for measuring essence solvent content rapidly |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20090429 |