Identification of pesticide varieties by testing microalgae using Visible/Near Infrared Hyperspectral Imaging technology

Rapid smartphone-based assays for pesticides inspection in foods: current status, limitations, and future directions

Smartphone-based assays to inspect pesticides in foods have attracted much attention as such assays can transform tedious laboratory-based assays into real-time, on-site, or even home-based assay and hence overcoming the limitations of conventional assays. Although an array of smartphone-based assays is available, information on the use of these assays for pesticides inspection is scattered. The purposes of this review are therefore to compile, summarize and discuss state-of-the-art as well as advantages and limitations of the relevant technologies. Suggestions are provided for further development of smartphone-based assays for rapid inspection of pesticides in foods. Smartphone-based assays relying on enzyme inhibitions are noted to be nonselective qualitative, capable of reporting results in a quantitative manner only when a sample contains an individual pesticide. Smartphone-based assays relying on chemical reactions also target only individual pesticides. Smartphone-based visible spectroscopy can, on the other hand, inspect individual and multiple pesticides with the aid of appropriate colorimetry-, luminescence-, or fluorescence-based assay. Smartphone-based visible-near infrared and Raman spectroscopies are suitable for simultaneous multiple pesticides inspection. Raman spectroscopy is of particular interest as it can detect pesticides even at lower concentrations. This spectroscopic technique can also serve as a real-time assay with the aid of cloud network computations.

An Improvised Machine Learning Model Based on Mutual Information Feature Selection Approach for Microbes Classification

The accurate classification of microbes is critical in today's context for monitoring the ecological balance of a habitat. Hence, in this research work, a novel method to automate the process of identifying microorganisms has been implemented. To extract the bodies of microorganisms accurately, a generalized segmentation mechanism which consists of a combination of convolution filter (Kirsch) and a variance-based pixel clustering algorithm (Otsu) is proposed. With exhaustive corroboration, a set of twenty-five features were identified to map the characteristics and morphology for all kinds of microbes. Multiple techniques for feature selection were tested and it was found that mutual information (MI)-based models gave the best performance. Exhaustive hyperparameter tuning of multilayer layer perceptron (MLP), k-nearest neighbors (KNN), quadratic discriminant analysis (QDA), logistic regression (LR), and support vector machine (SVM) was done. It was found that SVM radial required further improvisation to attain a maximum possible level of accuracy. Comparative analysis between SVM and improvised SVM (ISVM) through a 10-fold cross validation method ultimately showed that ISVM resulted in a 2% higher performance in terms of accuracy (98.2%), precision (98.2%), recall (98.1%), and F1 score (98.1%).

Classification, identification, and growth stage estimation of microalgae based on transmission hyperspectral microscopic imaging and machine learning

A transmission hyperspectral microscopic imager (THMI) that utilizes machine learning algorithms for hyperspectral detection of microalgae is presented. The THMI system has excellent performance with spatial and spectral resolutions of 4 µm and 3 nm, respectively. We performed hyperspectral imaging (HSI) of three species of microalgae to verify their absorption characteristics. Transmission spectra were analyzed using principal component analysis (PCA) and peak ratio algorithms for dimensionality reduction and feature extraction, and a support vector machine (SVM) model was used for classification. The average accuracy, sensitivity and specificity to distinguish one species from the other two species were found to be 94.4%, 94.4% and 97.2%, respectively. A species identification experiment for a group of mixed microalgae in solution demonstrates the usability of the classification method. Using a random forest (RF) model, the growth stage in a phaeocystis growth cycle cultivated under laboratory conditions was predicted with an accuracy of 98.1%, indicating the feasibility to evaluate the growth state of microalgae through their transmission spectra. Experimental results show that the THMI system has the capability for classification, identification and growth stage estimation of microalgae, with strong potential for in-situ marine environmental monitoring and early warning detection applications.

Evaluation of quinclorac toxicity and alleviation by salicylic acid in rice seedlings using ground-based visible/near-infrared hyperspectral imaging

BACKGROUND

To investigate potential effects of herbicide phytotoxic on crops, a major challenge is a lack of non-destructive and rapid methods to detect plant growth that could allow characterization of herbicide-resistant plants. In such a case, hyperspectral imaging can quickly obtain the spectrum for each pixel in the image and monitor status of plants harmlessly.

METHOD

Hyperspectral imaging covering the spectral range of 380-1030 nm was investigated to determine the herbicide toxicity in rice cultivars. Two rice cultivars, Xiushui 134 and Zhejing 88, were respectively treated with quinclorac alone and plus salicylic acid (SA) pre-treatment. After ten days of treatments, we collected hyperspectral images and physiological parameters to analyze the differences. The score images obtained were used to explore the differences among samples under diverse treatments by conducting principal component analysis on hyperspectral images. To get useful information from original data, feature extraction was also conducted by principal component analysis. In order to classify samples under diverse treatments, full-spectra-based support vector classification (SVC) models and extracted-feature-based SVC models were established. The prediction maps of samples under different treatments were constructed by applying the SVC models using extracted features on hyperspectral images, which provided direct visual information of rice growth status under herbicide stress. The physiological analysis with the changes of stress-responsive enzymes confirmed the differences of samples under different treatments.

RESULTS

The physiological analysis showed that SA alleviated the quinclorac toxicity by stimulating enzymatic activity and reducing the levels of reactive oxygen species. The score images indicated there were spectral differences among the samples under different treatments. Full-spectra-based SVC models and extracted-feature-based SVC models obtained good results for the aboveground parts, with classification accuracy over 80% in training, validation and prediction set. The SVC models for Zhejing 88 presented better results than those for Xiushui 134, revealing the different herbicide tolerance between rice cultivars.

CONCLUSION

We develop a reliable and effective model using hyperspectral imaging technique which enables the evaluation and visualization of herbicide toxicity for rice. The reflectance spectra variations of rice could reveal the stress status of herbicide toxicity in rice along with the physiological parameters. The visualization of the herbicide toxicity in rice would help to provide the intuitive vision of herbicide toxicity in rice. A monitoring system for detecting herbicide toxicity and its alleviation by SA will benefit from the remarkable success of SVC models and distribution maps.

Identification of heavy metal by testing microalgae using confocal Raman microspectroscopy technology

Five copper concentrations (0, 0.5, 1, 2, and 4 mg/l) were used to stress C. pyrenoidosa continuously for five days. The biomass, chlorophyll, and carotenoids of microalgae were measured, and Raman mapping spectral data and Raman single-point spectral data of microalgae were acquired. Principal component-linear discriminant analysis, back propagation-artificial neural network (BP-ANN), and sensitive wavelengths-linear discriminant analysis were used to build models to identify different copper concentrations using the spectral data after pretreatment. The results showed that the BP-ANN model was optimal to identify copper concentrations with prediction accuracy of 92% on day 4.

NIR Hyperspectral Imaging Technology Combined with Multivariate Methods to Study the Residues of Different Concentrations of Omethoate on Wheat Grain Surface

In this study, a hyperspectral imaging system of 866.4-1701.0 nm was selected and combined with multivariate methods to identify wheat kernels with different concentrations of omethoate on the surface. In order to obtain the optimal model combination, three preprocessing methods (standard normal variate (SNV), Savitzky-Golay first derivative (SG1), and multivariate scatter correction (MSC)), three feature extraction algorithms (successive projections algorithm (SPA), random frog (RF), and neighborhood component analysis (NCA)), and three classifier models (decision tree (DT), k-nearest neighbor (KNN), and support vector machine (SVM)) were applied to make a comparison. Firstly, based on the full wavelengths modeling analysis, it was found that the spectral data after MSC processing performed best in the three classifier models. Secondly, three feature extraction algorithms were used to extract the feature wavelength of MSC processed data and based on feature wavelengths modeling analysis. As a result, the MSC-NCA-SVM model performed best and was selected as the best model. Finally, in order to verify the reliability of the selected model, the hyperspectral image was substituted into the MSC-NCA-SVM model and the object-wise method was used to visualize the image classification. The overall classification accuracy of the four types of wheat kernels reached 98.75%, which indicates that the selected model is reliable.

Feasibility of Laser-Induced Breakdown Spectroscopy and Hyperspectral Imaging for Rapid Detection of Thiophanate-Methyl Residue on Mulberry Fruit

An effective and rapid way to detect thiophanate-methyl residue on mulberry fruit is important for providing consumers with quality and safe of mulberry fruit. Chemical methods are complex, time-consuming, and costly, and can result in sample contamination. Rapid detection of thiophanate-methyl residue on mulberry fruit was studied using laser-induced breakdown spectroscopy (LIBS) and hyperspectral imaging (HSI) techniques. Principal component analysis (PCA) and partial least square regression (PLSR) were used to qualitatively and quantitatively analyze the data obtained by using LIBS and HSI on mulberry fruit samples with different thiophanate-methyl residues. The competitive adaptive reweighted sampling algorithm was used to select optimal variables. The results of model calibration were compared. The best result was given by the PLSR model that used the optimal preprocessed LIBS–HSI variables, with a correlation coefficient of 0.921 for the prediction set. The results of this research confirmed the feasibility of using LIBS and HSI for the rapid detection of thiophanate-methyl residue on mulberry fruit.

Effect of Dibutyl Phthalate on the Tolerance and Lipid Accumulation in the Green Microalgae Chlorella vulgaris

In the present study, Chlorella vulgaris were cultured in the presence of the common plasticizer dibutyl phthalate (DBP) with different concentrations for 10 days. The cell density, DBP concentrations, neutral lipid concentrations, and lipid morphology in C. vulgaris were studied using optical microscopy, gas chromatography (GC), fluorescence spectrophotometry, and laser scanning confocal microscopy (LSCM). We observed that the neutral lipid contents and cell density of C. vulgaris were negatively influenced by DBP of high concentrations (50 and 100 mg/L), but significantly stimulated by DBP of low concentrations (5, 10, and 20 mg/L). Lipid bodies were destroyed into pieces by DBP of high concentrations (50 and 100 mg/L), but were slightly suppressed by DBP at low concentrations (5, 10, and 20 mg/L). Chlorella vulgaris treated with DBP (50 mg/L) for 2 days showed the highest removal efficiency (31.69%). The results suggested that C. vulgaris could be used in practice to remove DBP and has the potential of being oleaginous microalgae in DBP contaminated water.

Detection and imaging of lipids of Scenedesmus obliquus based on confocal Raman microspectroscopy

In this study, confocal Raman microspectroscopy was used to detect lipids in microalgae rapidly and non-destructively. Microalgae cells were cultured under nitrogen deficiency. The accumulation of lipids in Scenedesmus obliquus was observed by Nile red staining, and the total amount of lipids accumulated in the cells was measured by gravimetric method. The signals from different microalgae cells were collected by confocal Raman microspectroscopy to establish a prediction model of intracellular lipid content, and surface scanning signals for drawing pseudo color images of lipids distribution. The images can show the location of pyrenoid and lipid accumulation in cells. Analyze Raman spectrum data and build PCA-LDA model using four different bands (full bands, pigments, lipids, and mixed features). Models of full bands or pigment characteristic bands were capable of identifying S. obliquus cells under different nitrogen stress culture time. The prediction accuracy of model of lipid characteristic bands is relatively low. The correlation between the fatty acid content measured by the gravimetric method and the integral Raman intensity of the oil characteristic peak (1445 cm-1) measured by Raman spectroscopy was analyzed. There was significant correlation (R2 = 0.83), which means that Raman spectroscopy is applicable to semi-quantitative detection of microalgal lipid content.