Quantitative Determination of Fusarium proliferatum Concentration in Intact Garlic Cloves Using Near-Infrared Spectroscopy.

Sensors (Basel) 2016 Jul 15;16(7). Epub 2016 Jul 15.

Department of Life Science and Biotechnology, University of Ferrara, Via L. Borsari, 46, 44121 Ferrara, Italy.

Fusarium proliferatum is considered to be a pathogen of many economically important plants, including garlic. The objective of this research was to apply near-infrared spectroscopy (NIRS) to rapidly determine fungal concentration in intact garlic cloves, avoiding the laborious and time-consuming procedures of traditional assays. Preventive detection of infection before seeding is of great interest for farmers, because it could avoid serious losses of yield during harvesting and storage. Spectra were collected on 95 garlic cloves, divided in five classes of infection (from 1-healthy to 5-very highly infected) in the range of fungal concentration 0.34-7231.15 ppb. Calibration and cross validation models were developed with partial least squares regression (PLSR) on pretreated spectra (standard normal variate, SNV, and derivatives), providing good accuracy in prediction, with a coefficient of determination (R²) of 0.829 and 0.774, respectively, a standard error of calibration (SEC) of 615.17 ppb, and a standard error of cross validation (SECV) of 717.41 ppb. The calibration model was then used to predict fungal concentration in unknown samples, peeled and unpeeled. The results showed that NIRS could be used as a reliable tool to directly detect and quantify F. proliferatum infection in peeled intact garlic cloves, but the presence of the external peel strongly affected the prediction reliability.

Download full-text PDF

Source
http://dx.doi.org/10.3390/s16071099DOI Listing
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4970144PMC
July 2016

Publication Analysis

Top Keywords

garlic cloves
16
fungal concentration
12
intact garlic
12
standard error
8
near-infrared spectroscopy
8
cross validation
8
concentration intact
8
ppb calibration
8
fusarium proliferatum
8
garlic
5
concentration unknown
4
5-very highly
4
predict fungal
4
1-healthy 5-very
4
model predict
4
infection 1-healthy
4
infected range
4
calibration model
4
concentration 034-723115
4
range fungal
4

Similar Publications