Skip to main content
ARS Home » Southeast Area » Athens, Georgia » U.S. National Poultry Research Center » Quality and Safety Assessment Research Unit » Research » Publications at this Location » Publication #346236

Research Project: Assessment and Improvement of Poultry Meat, Egg, and Feed Quality

Location: Quality and Safety Assessment Research Unit

Title: Evaluation of growth characteristics of Aspergillus parasiticus inoculated in different culture media by shortwave infrared (SWIR) hyperspectral imaging

item CHU, XUAN - China Agricultural University
item WANG, WEI - China Agricultural University
item NI, XINZHI - US Department Of Agriculture (USDA)
item ZHAO, XI - China Agricultural University
item Zhuang, Hong
item Lawrence, Kurt
item LI, CHUNYANG - China Agricultural University

Submitted to: Journal of Innovative Optical Health Sciences
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/13/2018
Publication Date: 9/12/2018
Citation: Chu, X., Wang, W., Ni, X., Zhao, X., Zhuang, H., Lawrence, K.C., Li, C. 2018. Evaluation of growth characteristics of Aspergillus parasiticus inoculated in different culture media by shortwave infrared (SWIR) hyperspectral imaging. Journal of Innovative Optical Health Sciences. 11(5). https:/10.1142/S1793545818500311.

Interpretive Summary: Fungi are a group of microorganisms with great environmental significance. However, their growth causes food spoilage and safety concern. Mold growth results in spoilage of plant-based food products. Mycotoxin aflatoxin and fumonisins taken up by live animal through crops are associated with fatal humans and livestock disease. One of methods that can be used to prevent fungi from contaminating food products is to control or eliminate fungal growth. However, the traditional microbiological methods for measuring fungal growth are generally time consuming and require specialized instrument and trained personnel. Recently, hyperspectral image (HSI) technique has been successfully developed to rapidly and nondestructively predict quality and microbial contaminations of food and agricultural products. The objective of this study was to assess HSI technique to predict fungal growth. Results show that HSI technique combined with multivariable statistical analyses can well predict growth of fungus A. parasiticus on different culture media.

Technical Abstract: The growth characteristics of Aspergillus parasiticus incubated on two 16 culture media were examined using shortwave infrared (SWIR) hyperspectral imaging (HSI) with wavelength range between 1000 and 2500 nm in this work. The HSI images of the A. parasiticus colonies growing on rose bengal medium (RBM) or maize agar medium (MAM) were recorded daily for 6 d. After principal component analysis (PCA), with the help of the score value of the first principal component (PC1) and its threshold value, noisy background were removed and A. parasiticus colonies were extracted. The growth phases of A. parasiticus were indicated through the pixel number and the characteristics of average spectra of colonies incubated on a specific medium for different durations. When plotting PC1 against PC2 which were calculated from the colonies extracted images, four concentric growth zones were identified in the direction of PC1 in almost equal segments, and the growth curve of each zone also helped to explain the rationality of this division. Seven wavelengths (1145, 1195, 1279, 1442, 1655, 1834 and 1929 nm) were selected as feature signatures of analysis, the fungal colony growth from PC1 loading plot, the average spectra of each colony, as well as each growth zone. Furthermore, support vector machine (SVM) classifier based on the selected characteristic wavelengths was built, and the classification accuracy for the four zones (from outer to inner zones) on the growing colonies were 99.79, 99.58, 99.37 and 99.29% of colonies on RBM, and 99.68, 99.41, 99.78 and 98.80% for colonies on MAM. In addition, a new PCA model was developed to differentiate the colonies incubated on RBM and MAM for 6 d. Score plot of PC2 and PC3 showed that there were two clusters corresponded to each colony. Then a new SVM model was developed based on spectra at wavelengths of 1067, 1369, 1459, 1694, 1834 and 1929 nm, which were selected from the loading of PC2 and PC3 could differentiate pixels of fungal colonies on RBM and MAM with accuracy of 99.99 and 100.00%, respectively. In conclusion, SWIR hyperspectral image is a powerful tool for evaluation of growth characteristics of A. parasiticus incubated in different culture media.