Development of a Low-Cost Electronic Nose for Detection of Pathogenic Fungi and Applying It to Fusarium oxysporum and Rhizoctonia solani

The evaluation of crop health status and early disease detection are critical for implementing a fast response to a pathogen attack, managing crop infection, and minimizing the risk of disease spreading. Fusarium oxysporum f. sp. cepae, which causes fusarium basal rot disease, is considered one of the most harmful pathogens of onion and accounts for considerable crop losses annually. In this work, the capability of the PEN 3 electronic nose system to detect onion and shallot bulbs infected with F. oxysporum f. sp. cepae, to track the progression of fungal infection, and to discriminate between the varying proportions of infected onion bulbs was evaluated. To the best of our knowledge, this is a first report on successful application of an electronic nose to detect fungal infections in post-harvest onion and shallot bulbs. Sensor array responses combined with PCA provided a clear discrimination between non-infected and infected onion and shallot bulbs as well as differentiation between samples with varying proportions of infected bulbs. Classification models based on LDA, SVM, and k-NN algorithms successfully differentiate among various rates of infected bulbs in the samples with accuracy up to 96.9%. Therefore, the electronic nose was proved to be a potentially useful tool for rapid, non-destructive monitoring of the post-harvest crops.

Section: Introductionmentioning

confidence: 99%

Preliminary Studies on Detection of Fusarium Basal Rot Infection in Onions and Shallots Using Electronic Nose

Labanska

Amsterdam

Jenkins

et al. 2022

“…The relationships between readings from individual sensors and membership in the right class are characterized by high complexity. At the same time, in the family of classical machine learning models, such as multinomial logistic regression, linear discriminant analysis or single decision trees, the obtained solutions leave much to be desired [ 48 , 49 , 50 , 51 ]. Therefore, advanced machine learning models such as SVM, RF, or ANN are necessary to properly classify objects based on statistical models [ 52 , 53 , 54 , 55 , 56 ].…”

Section: Review Of Advances In Machine Learning Methods For Analysis ...mentioning

confidence: 99%

Application of Machine Learning Methods for an Analysis of E-Nose Multidimensional Signals in Wastewater Treatment

Piłat-Rożek

Łazuka

Majerek

et al. 2023

The work represents a successful attempt to combine a gas sensors array with instrumentation (hardware), and machine learning methods as the basis for creating numerical codes (software), together constituting an electronic nose, to correct the classification of the various stages of the wastewater treatment process. To evaluate the multidimensional measurement derived from the gas sensors array, dimensionality reduction was performed using the t-SNE method, which (unlike the commonly used PCA method) preserves the local structure of the data by minimizing the Kullback-Leibler divergence between the two distributions with respect to the location of points on the map. The k-median method was used to evaluate the discretization potential of the collected multidimensional data. It showed that observations from different stages of the wastewater treatment process have varying chemical fingerprints. In the final stage of data analysis, a supervised machine learning method, in the form of a random forest, was used to classify observations based on the measurements from the sensors array. The quality of the resulting model was assessed based on several measures commonly used in classification tasks. All the measures used confirmed that the classification model perfectly assigned classes to the observations from the test set, which also confirmed the absence of model overfitting.

“…The previously reported low-cost electronic noses built in our laboratory [ 32 , 33 , 34 ] were constructed as a round probe, fit to the size of a Petri dish or jar opening hole, and manually moved from the clean air conditions to the measured sample. The noses were used to classify samples of pathogenic oomycetes Phytophthora plurivora, Pythium intermedium and fungi Fusarium oxysporum, Rhizoctonia solani .…”

Section: Electronic Noses Sensor Chamber Designsmentioning

confidence: 99%

“…There are several types of electronic circuit topologies commonly used for the capture of the MOX sensors’ response [ 48 ]. In our case, we measure the voltage on a resistor serially connected to the sensor [ 32 , 33 , 34 ]. In Figure 4 , we present an example of a typical shape of measured response from one of the sensors.…”

Section: Electronic Nose Device Used In the Experimentsmentioning

confidence: 99%

Development of a Low-Cost Electronic Nose with an Open Sensor Chamber: Application to Detection of Ciboria batschiana

Borowik

Grzywacz

Tarakowski

et al. 2023

Self Cite

In the construction of electronic nose devices, two groups of measurement setups could be distinguished when we take into account the design of electronic nose chambers. The simpler one consists of placing the sensors directly in the environment of the measured gas, which has an important advantage, in that the composition of the gas is not changed as the gas is not diluted. However, that has an important drawback in that it is difficult to clean sensors between measurement cycles. The second, more advanced construction, contains a pneumatic system transporting the gas inside a specially designed sensor chamber. A new design of an electronic nose gas sensor chamber is proposed, which consists of a sensor chamber with a sliding chamber shutter, equipped with a simple pneumatic system for cleaning the air. The proposal combines the advantages of both approaches to the sensor chamber designs. The sensors can be effectively cleared by the flow of clean air, while the measurements are performed in the open state when the sensors are directly exposed to the measured gas. Airflow simulations were performed to confirm the efficiency of clean air transport used for sensors’ cleaning. The demonstrated electronic nose applies eight Figaro Co. MOS TGS series sensors, in which a transient response caused by a change of the exposition to measured gas, and change of heater voltage, was collected. The new electronic nose was tested as applied to the differentiation between the samples of Ciboria batschiana fungi, which is one of the most harmful pathogens of stored acorns. The samples with various coverage, thus various concentrations of the studied odor, were measured. The tested device demonstrated low noise and a good level of repetition of the measurements, with stable results during several hours of repetitive measurements during an experiment lasting five consecutive days. The obtained data allowed complete differentiation between healthy and infected samples.