Gas Sensor Array and Classifiers as a Means of Varroosis Detection

Abstract

The study focused on a method of detection for bee colony infestation with the Varroa destructor mite, based on the measurements of the chemical properties of beehive air. The efficient detection of varroosis was demonstrated. This method of detection is based on a semiconductor gas sensor array and classification module. The efficiency of detection was characterized by the true positive rate (TPR) and true negative rate (TNR). Several factors influencing the performance of the method were determined. They were: (1) the number and kind of sensors, (2) the classifier, (3) the group of bee colonies, and (4) the balance of the classification data set. Gas sensor array outperformed single sensors. It should include at least four sensors. Better results of detection were attained with a support vector machine (SVM) as compared with the k-nearest neighbors (k-NN) algorithm. The selection of bee colonies was important. TPR and TNR differed by several percent for the two examined groups of colonies. The balance of the classification data was crucial. The average classification results were, for the balanced data set: TPR = 0.93 and TNR = 0.95, and for the imbalanced data set: TP = 0.95 and FP = 0.53. The selection of bee colonies and the balance of classification data set have to be controlled in order to attain high performance of the proposed detection method.

The Effectiveness of Varroa destructor Infestation Classification Using an E-Nose Depending on the Time of Day

“…In our opinion, this idea allows for the detection of the infestation rate of the honey bee colony. The results of our previous studies of this issue support this assumption [ 36 , 37 , 38 , 39 ].…”

“…The responses of the all sensors, elements of the gas sensor array, were utilized. The employed approach to feature vector formation was based on the results of earlier works on gas sensor measurements using the dynamic mode of operation [ 37 , 38 , 39 , 55 ]. In particular, it was demonstrated by Szczurek et al [ 37 , 38 , 39 ] that the first several minutes of gas sensor exposure to beehive air were sufficient to attain the effective classification of V.d.…”

“…The monitoring based on a stationary E-noses is expensive, as one instrument is dedicated for only one beehive. In addition, the measurement equipment is under the influence of meteorological conditions [ 36 , 37 , 38 , 39 ]. Metrological problems result also from the long exposure of sensors to volatile compounds, which are components of the gas permanently flowing through the measurement chamber.…”

Varroa destructor: how does it harm Apis mellifera honey bees and what can be done about it?

“…New techniques, using technological developments such as gas sensors or computer vision systems, are being tested to estimate Varroa infestation [113,114]. They have yet to be transferred to the field and to the beekeeping community.…”

Diagnosis of Varroosis Based on Bee Brood Samples Testing with Use of Semiconductor Gas Sensors

“…Semiconductor metal oxide sensors (MOS—metal-oxide-semiconductor) are most commonly used in the electronic nose, which work through the reduction oxidation reactions of adsorbed gases on the sensory layer [ 4 ]. The odour classification is then based on the values of the sensor readings expressing the sensor resistance [ 5 , 6 ]. On the other hand, by creating three parametric sensorgrams from measurements containing data about: maximum standardized sensor response, reaction time, and cleaning time, it is also possible to accurately determine the substances included in the tested gas mixture, as well as to give their concentration [ 7 , 8 ].…”