Identification of Modified Starches Using Infrared Spectroscopy and Artificial Neural Network Processing

The authentication of food is a very important issue both for the consumers and for the food industry with respect to all levels of the food chain from raw materials to finished products. Corn starch can be used in a wide variety of food preparation as bakery cream fillings, sauce, or dry mixes. There are many modifications of the corn starch in connection with its use in the agrofood industry. This paper describes a novel approach to the classification of modified starches and the recognition of their modifications by artificial neural network (ANN) processing of attenuated total reflection Fourier transform spectroscopy (ATR/FT-IR) spectra. Using the self-organizing artificial neural network of the Kohonen type, we can obtain natural groupings of similarly modified samples on a two-dimensional plane. Such mapping provides the expert with the possibility of analyzing the distribution of samples and predicting modifications of unknown samples by using their relative position with respect to existing clusters. On the basis of the available information in the infrared spectra, a feedforward artificial neural network, trained with the intensities of the derivative infrared spectra as input and the starch modifications as output, allows the user to identify modified starches presented as prediction samples.

[1]  Martin A. Riedmiller,et al.  A direct adaptive method for faster backpropagation learning: the RPROP algorithm , 1993, IEEE International Conference on Neural Networks.

[2]  Nathalie Dupuy,et al.  Classification of Green Coffees by FT-IR Analysis of Dry Extract , 1995 .

[3]  Teuvo Kohonen,et al.  Self-Organizing Maps , 2010 .

[4]  Daniel Cabrol-Bass,et al.  Neural network approach to structural feature recognition from infrared spectra , 1993, J. Chem. Inf. Comput. Sci..

[5]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[6]  Johann Gasteiger,et al.  Neural Networks for Chemists: An Introduction , 1993 .

[7]  Yoshua Bengio,et al.  Pattern Recognition and Neural Networks , 1995 .

[8]  D. D. Wit,et al.  Short-range structure in (partially) crystalline potato starch determined with attenuated total reflectance Fourier-transform IR spectroscopy , 1995 .

[9]  J. Huvenne,et al.  Multivariate Determination of Sugar Powders by Attenuated Total Reflectance Infrared Spectroscopy , 1993 .

[10]  Paul J. Werbos,et al.  The roots of backpropagation , 1994 .

[11]  Sohail Sheikh,et al.  Classification and Clustering: Using Neural Networks , 1994, Journal of chemical information and computer sciences.

[12]  B. Forrest.  Identification and quantitation of hydroxypropylation of starch by FTIR , 1992 .

[13]  Brian D. Ripley,et al.  Neural networks and flexible regression and discrimination , 1994 .

[14]  Jean-Pierre Doucet,et al.  Neural networks and carbon-13 NMR shift prediction , 1993, J. Chem. Inf. Comput. Sci..

[15]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[16]  Brian D. Ripley,et al.  Statistical aspects of neural networks , 1993 .

[17]  Sven P. Jacobsson.  Feature extraction of polysaccharides by low-dimensional internal representation neural networks and infrared spectroscopy , 1994 .

[18]  Martin A. Riedmiller,et al.  Advanced supervised learning in multi-layer perceptrons — From backpropagation to adaptive learning algorithms , 1994 .