Sergey V. Kucheryavskiy

Classification of maize kernels using NIR hyperspectral imaging.

NIR hyperspectral imaging was evaluated to classify maize kernels of three hardness categories: hard, medium and soft. Two approaches, pixel-wise and object-wise, were investigated to group kernels according to hardness. The pixel-wise classification assigned a class to every pixel from individual kernels and did not give acceptable results because of high misclassification. However by using a predefined threshold and classifying entire kernels based on the number of correctly predicted pixels, improved results were achieved (sensitivity and specificity of 0.75 and 0.97). Object-wise classification was performed using two methods for feature extraction - score histograms and mean spectra. The model based on score histograms performed better for hard kernel classification (sensitivity and specificity of 0.93 and 0.97), while that of mean spectra gave better results for medium kernels (sensitivity and specificity of 0.95 and 0.93). Both feature extraction methods can be recommended for classification of maize kernels on production scale.

Determination of fat and total protein content in milk using conventional digital imaging

The applicability of conventional digital imaging to quantitative determination of fat and total protein in cows milk, based on the phenomenon of light scatter, has been proved. A new algorithm for extracting features from digital images of milk samples has been developed. The algorithm takes into account spatial distribution of light, diffusely transmitted through a sample. The proposed method has been tested on two sample sets prepared from industrial raw milk standards, with variable fat and protein content. Partial Least-Squares (PLS) regression on the features calculated from images of monochromatically illuminated milk samples resulted in models with high prediction performance when analysed the sets separately (best models with cross-validated R(2)=0.974 for protein and R(2)=0.973 for fat content). However when analysed the sets jointly with the obtained results were significantly worse (best models with cross-validated R(2)=0.890 for fat content and R(2)=0.720 for protein content). The results have been compared with previously published Vis/SW-NIR spectroscopic study of similar samples.

Monitoring of whey quality with NIR spectroscopy: A feasibility study

The possibility of using near-infrared (NIR) spectroscopy for monitoring of liquid whey quality parameters during protein production process has been tested. The parameters included total solids, lactose, protein and fat content. The samples for the experiment were taken from real industrial processes and had a large variability for most of the parameters. Partial Least Squares (PLS) regression was used to make the prediction models based on NIR spectra taken at 30 and 40°C. Using proper wavelength range allowed to get models for prediction of fat, protein and amount of total solids with very high precision and accuracy. The lactose was found to be the most challenging parameter.

Assessment of PCBs and exposure risk to infants in breast milk of primiparae and multiparae mothers in an electronic waste hot spot and non-hot spot areas in Ghana

The aim of the study was to assess the levels of PCBs in the breast milk of some Ghanaian women at suspected hotspot and relatively non-hotspot areas and to find out if the levels of these PCBs pose any risk to the breastfed infants. A total of 128 individual human breast milk were sampled from both primiparae and multiparae mothers. The levels of PCBs in the milk samples were compared. Some of these mothers (105 individuals) work or reside in and around Agbogbloshie (hot-spot), the largest electric and electronic waste dump and recycling site in Accra, Ghana. Others (23 donor mothers) also reside in and around Kwabenya (non-hotspot) which is a mainly residential area without any industrial activities. Samples were analyzed using GC-MS/MS. The total mean levels and range of Σ7PCBs were 3.64ng/glipidwt and ˂LOD-29.20ng/glipidwt, respectively. Mean concentrations from Agbogbloshie (hot-spot area) and Kwabenya (non-hotspot areas) were 4.43ng/glipidwt and 0.03ng/glipidwt, respectively. PCB-28 contributed the highest of 29.5% of the total PCBs in the milk samples, and PCB-101 contributed the lowest of 1.74%. The estimated daily intake of PCBs and total PCBs concentrations in this work were found to be lower as compared to similar studies across the world. The estimated hazard quotient using Health Canadas guidelines threshold limit of 1μg/kgbw/day showed no potential health risk to babies. However, considering minimum tolerable value of 0.03μg/kgbw/day defined by the Agency for Toxic Substances and Disease Registry (ATSDR), the values of some mothers were found to be at the threshold limit. This may indicate a potential health risk to their babies. Mothers with values at the threshold levels of the minimum tolerable limits are those who work or reside in and around the Agbogbloshie e-waste site.

Metabotyping Patients’ Journeys Reveals Early Predisposition to Lung Injury after Cardiac Surgery

Cardiovascular disease is the leading cause of death worldwide and patients with severe symptoms undergo cardiac surgery. Even after uncomplicated surgeries, some patients experience postoperative complications such as lung injury. We hypothesized that the procedure elicits metabolic activity that can be related to the disease progression, which is commonly observed two-three days postoperatively. More than 700 blood samples were collected from 50 patients at nine time points pre-, intra-, and postoperatively. Dramatic metabolite shifts were observed during and immediately after the intervention. Prolonged surgical stress was linked to an augmented anaerobic environment. Time series analysis showed shifts in purine-, nicotinic acid-, tyrosine-, hyaluronic acid-, ketone-, fatty acid, and lipid metabolism. A characteristic ‘metabolic biosignature’ was identified correlating with the risk of developing postoperative complications two days before the first clinical signs of lung injury. Hence, this study demonstrates the link between intra- and postoperative time-dependent metabolite changes and later postoperative outcome. In addition, the results indicate that metabotyping patients’ journeys early, during or just after the end of surgery, may have potential impact in hospitals for the early diagnosis of postoperative lung injury, and for the monitoring of therapeutics targeting disease progression.

Spectral Unmixing Using the Concept of Pure Variables

Abstract The concepts of pure variables as well as spectral unmixing methods that utilize the concept are presented. The methods have a number of advantages, comparing to alternatives, including speed (one of the fastest curve resolution methods), sequential procedure for resolving of pure components (which means there is no need for preliminary determination of number of components), as well as a high degree of interactivity. The theoretical basics are supplemented with examples on how the method performs on spectroscopic data of different nature, including hyperspectral images, and how results can be improved interactively.

Blessing of randomness against the curse of dimensionality

Modern hyperspectral images, especially acquired in remote sensing and from on‐field measurements, can easily contain from hundreds of thousands to several millions of pixels. This often leads to a quite long computational time when, eg, the images are decomposed by Principal Component Analysis (PCA) or similar algorithms. In this paper, we are going to show how randomization can tackle this problem. The main idea is described in detail by Halko et al in 2011 and can be used for speeding up most of the low‐rank matrix decomposition methods. The paper explains this approach using visual interpretation of its main steps and shows how the use of randomness influences the speed and accuracy of PCA decomposition of hyperspectral images.

Application of Image Texture Analysis for Evaluation of X-Ray Images of Fungal-Infected Maize Kernels

The feasibility of image texture analysis to evaluate X-ray images of fungal-infected maize kernels was investigated. X-ray images of maize kernels infected with Fusarium verticillioides and control kernels were acquired using high-resolution X-ray micro-computed tomography. After image acquisition and pre-processing, several algorithms were developed to extract image textural features from selected two-dimensional (2D) images of the kernels. Four first-order statistics (mean, standard deviation, kurtosis and skewness) and four grey level co-occurrence matrix (GLCM) features (correlation, energy, homogeneity and contrast) were extracted from the side, front and top views of each kernel and used as inputs for principal component analysis (PCA). The first-order statistical image features gave a better separation of the control from infected kernels on day 8 post-inoculation. Classification models were developed using partial least squares discriminant analysis (PLS-DA), and accuracies of 67 and 73% were achieved using first-order statistical features and GLCM extracted features, respectively. This work provides information on the possible application of image texture as method for analysing X-ray images.