OFFICE OF THE TEXAS STATE CHEMIST Texas Feed and Fertilizer Control Service Agriculture Analytical Service Application of Raman Spectroscopy for Detection of Aflatoxins and Fumonisins in Ground Maize Samples Office of the Texas State Chemist, Texas A&M AgriLife Research Kyung-Min Lee and Timothy J. Herrman January 07, 2013 Mycotoxin Working Group Meeting 2013 Texas A&M AgriLife Conference
22
Embed
Application of Raman Spectroscopy for Detection of ...otscweb.tamu.edu/OTSC-Present/2013/Application of...Raman spectroscopy measurements Calibration models would be more stable and
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
OFFICE OF THE TEXAS STATE CHEMIST Texas Feed and Fertilizer Control Service Agriculture Analytical Service
Application of Raman Spectroscopy
for Detection of Aflatoxins and Fumonisins
in Ground Maize Samples
Office of the Texas State Chemist, Texas A&M AgriLife Research
Kyung-Min Lee and Timothy J. Herrman
January 07, 2013
Mycotoxin Working Group Meeting
2013 Texas A&M AgriLife Conference
OFFICE OF THE TEXAS STATE CHEMIST
Mycotoxin detection methods
Diverse mycotoxin analytical methods available in laboratory and non-laboratory locations
a RMSEC: root-mean-square error of calibration b RMSEP: root-mean-square error of prediction c R2: correlation coefficient of determination
OFFICE OF THE TEXAS STATE CHEMIST
LC-MS/MS Ref. vs Predicted Values (fumonisins)
chemometrics preprocessing
method
paired differences (mg/kg) ra sig (2-tailed) RPDb
mean std error mean
MLR Normalization -4.06 1.75 0.9843 0.0384 4.324
1st derivative -2.47 2.44 0.9603 0.3321 3.579
2nd derivative -1.46 2.33 0.9670 0.5423 3.839
Deconvolution -1.51 1.65 0.9824 0.3782 5.316
PCR Normalization -4.73 2.20 0.9734 0.0529 3.511
1st derivative -0.52 2.53 0.9577 0.8423 3.583
2nd derivative -0.04 2.95 0.9512 0.9889 3.083
Deconvolution 0.43 2.79 0.9487 0.8790 3.258
PLSR Normalization -5.36 2.29 0.9710 0.0378 3.287
1st derivative -1.68 2.30 0.9649 0.4779 3.877
2nd derivative 0.59 2.77 0.9539 0.8355 3.277
Deconvolution -1.93 2.04 0.9726 0.3617 4.303
a Pearson correlation coefficient b RPD (residual prediction deviation): ratio of standard deviation of reference to root mean square error of
cross-validation
OFFICE OF THE TEXAS STATE CHEMIST
Conclusions Raman spectroscopic method: proved to be successfully applicable
as alternative rapid and non-destructive technique
Classification and quantification models showed a good predictive performance with high accuracy and low error rate
Ideal for real-time monitoring of critical performance attributes
Anticipating several difficulties and constraints in using this technique numerous opportunities to improve the accuracy and precision of Raman spectroscopy measurements
Calibration models would be more stable and practically applicable by continuing to analyze maize samples with diverse genetic and environmental backgrounds and mycotoxin levels
Raman spectroscopy: easy, rapid, and inexpensive screening system for mycotoxins a powerful tool for quality control of grains improve the safety of feed and food products supplied to consumers.
OFFICE OF THE TEXAS STATE CHEMIST
OFFICE OF THE TEXAS STATE CHEMIST 445 Agronomy Road College Station, TX 77840