TY - JOUR
T1 - Comparison of the performance of metal oxide and conducting polymer electronic noses for detection of aflatoxin using artificially contaminated maize
AU - Machungo, Catherine
AU - Berna, Amalia Z.
AU - McNevin, Dennis
AU - Wang, Rosalind
AU - Trowell, Stephen
PY - 2022
Y1 - 2022
N2 - The electronic nose offers potential as a rapid and cost effective field portable diagnostic device that would allow for quick screening of produce for aflatoxin contamination at the market entry level. This study aimed to compare the performance of three electronic nose sensor technologies: metal oxide semiconductor sensors (Fox 3000), conducting polymer sensors (Cyranose 320) and doped metal oxide semiconductor sensors with thermocycling (DiagNose), for the detection of volatiles associated with maize contaminated with aflatoxins. Australian maize (variety DK703w) samples were artificially inoculated with aflatoxigenic and non-aflatoxigenic Aspergillus flavus isolates and 2 % v/v Tween 20 as a control. Mutual information was used to select features from the electronic nose sensor signals for classification of the samples. The effectiveness, of selected features to discriminate between the different classes of samples was evaluated by support vector machines and k-nearest neighbour with leave-one-out cross-validation. Cross-validated classification accuracy for the different sample classes ranged from 81 % to 94 % for DiagNose, 76 to 79 % for Fox 3000 and 68 to 75 % for Cyranose. The results suggest that an electronic nose equipped with doped metal oxide semiconductor sensors and thermocycling is more effective for detection of aflatoxin contamination of maize.
AB - The electronic nose offers potential as a rapid and cost effective field portable diagnostic device that would allow for quick screening of produce for aflatoxin contamination at the market entry level. This study aimed to compare the performance of three electronic nose sensor technologies: metal oxide semiconductor sensors (Fox 3000), conducting polymer sensors (Cyranose 320) and doped metal oxide semiconductor sensors with thermocycling (DiagNose), for the detection of volatiles associated with maize contaminated with aflatoxins. Australian maize (variety DK703w) samples were artificially inoculated with aflatoxigenic and non-aflatoxigenic Aspergillus flavus isolates and 2 % v/v Tween 20 as a control. Mutual information was used to select features from the electronic nose sensor signals for classification of the samples. The effectiveness, of selected features to discriminate between the different classes of samples was evaluated by support vector machines and k-nearest neighbour with leave-one-out cross-validation. Cross-validated classification accuracy for the different sample classes ranged from 81 % to 94 % for DiagNose, 76 to 79 % for Fox 3000 and 68 to 75 % for Cyranose. The results suggest that an electronic nose equipped with doped metal oxide semiconductor sensors and thermocycling is more effective for detection of aflatoxin contamination of maize.
UR - https://hdl.handle.net/1959.7/uws:67372
U2 - 10.1016/j.snb.2022.131681
DO - 10.1016/j.snb.2022.131681
M3 - Article
SN - 0925-4005
VL - 360
JO - Sensors and Actuators B: Chemical
JF - Sensors and Actuators B: Chemical
M1 - 131681
ER -