Search

In Europe, two waves of infections with SARS-CoV-2 (COVID-19) have been observed to date. Here, we have investigated whether common patterns of cytokines could be detected in individuals with mild and severe forms of COVID-19 in the two pandemic waves, and whether machine learning approach could be useful to identify the best predictors. An increasing trend of multiple cytokines was observed in patients with mild or severe/critical symptoms of COVID-19, compared with healthy volunteers. Linear Discriminant Analysis (LDA) clearly recognized the three groups based on cytokine patterns. Classification and Regression Tree (CART) further indicated that IL-6 discriminated controls and COVID-19 patients, whilst IL-8 defined disease severity. During the second wave of pandemics, a less intense cytokine storm was observed. CART analysis revealed that IL-6 was the most robust predictor of infection and discriminated moderate COVID-19 patients from healthy controls, regardless of epidemic peak curve. Thus, serum cytokine patterns provide non-invasive biomarkers useful for COVID-19 diagnosis and prognosis. Further definition of individual cytokines may allow to envision novel therapeutic options and pave the way to set up innovative diagnostic tools.

Multielement analysis and chemometric methods were proposed to discriminate the Sorrento lemon (PGI) juices according to geographical origin. In 2018 and 2019, 169 fruits from three farms in PGI area and two in not-PGI area were collected and analysed for essential and not-essential elements by ICP-MS. The PCA of multielement fingerprinting grouped lemon juices from PGI farms revealing a strong differentiation at small geographical scale. The S-LDA discriminated lemon juices for Mo, Ba, Rb, Mg, Co, Ca, Fe, Sr on the two production years, giving 97.7% correct classification, 98.5% accuracy and 93.8% external validation. The good correlation lemon juice vs cultivation soil and the soil discrimination by not-essential elements suggested the use of these elements as reliable indicators of lemon juice provenances. Despite lowering the number of variables, constituted by not-essential elements Ba, Rb, Ti, Co, the use of S-QDA discriminated the lemons juices with 87.5% accuracy and 83.9% validation.

In recent months, Coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has spread throughout the world. COVID-19 patients show mild, moderate or severe symptoms with the latter ones requiring access to specialized intensive care. SARS-CoV-2 infections, pathogenesis and progression have not been clearly elucidated yet, thus forcing the development of many complementary approaches to identify candidate cellular pathways involved in disease progression. Host lipids play a critical role in the virus life, being the double-membrane vesicles a key factor in coronavirus replication. Moreover, lipid biogenesis pathways affect receptor-mediated virus entry at the endosomal cell surface and modulate virus propagation. In this study, targeted lipidomic analysis coupled with proinflammatory cytokines and alarmins measurement were carried out in serum of COVID-19 patients characterized by different severity degree. Serum IL-26, a cytokine involved in IL-17 pathway, TSLP and adiponectin were measured and correlated to lipid COVID-19 patient profiles. These results could be important for the classification of the COVID-19 disease and the identification of therapeutic targets.

Publisher's version (útgefin grein), Coronary artery disease is one of the most prevalent chronic pathologies in the modern world, leading to the deaths of thousands of people, both in the United States and in Europe. This article reports the use of data mining techniques to analyse a population of 10,265 people who were evaluated by the Department of Advanced Biomedical Sciences for myocardial ischaemia. Overall, 22 features are extracted, and linear discriminant analysis is implemented twice through both the Knime analytics platform and R statistical programming language to classify patients as either normal or pathological. The former of these analyses includes only classification, while the latter method includes principal component analysis before classification to create new features. The classification accuracies obtained for these methods were 84.5 and 86.0 per cent, respectively, with a specificity over 97 per cent and a sensitivity between 62 and 66 per cent. This article presents a practical implementation of traditional data mining techniques that can be used to help clinicians in decision-making; moreover, principal component analysis is used as an algorithm for feature reduction., The authors wish to thank Alec Shawn for his contribute as regards "grammar and spell check". This work has been realized thanks to the collaboration of the Department of Advanced Biomedical Sciences of the University Hospital "Federico II" of Naples. The authors wish to thank Sabrina De Vita, Francesca D'Agostino, Giuseppina Toscano and Tania Di Monda for their valuable contribution to the implementation of data mining algorithm during their MS thesis. The work has been partially carried out under TablHealth [CUP B49J17000720008] project and AK12 s.r.l.", "Peer Reviewed"

Alkylresorcinols, fatty acid content, and antioxidant activity were analysed in Triticum monococcum, T. dicoccum, T. timopheevi, T. palaeo-colchicum and T. macha accessions from Turkey, Georgia, Armenia, Bulgaria and Italy, also including modern durum wheat (T. durum) and bread wheat (T. aestivum) varieties as controls, grown in a common environment. Total alkylresorcinol content ranged between 478.3 and 1153.9 mg kg−1 dm, with the highest contents detected in T. aestivum, T. macha and some T. monococcum accessions. All alkylresorcinol homologues effectively contributed at genotype discrimination. Hexaploid wheats showed higher A19 (23.3–27.5%) and lower A23 (14.0–15.9%) and A25 (4.3–8.1%) relative contents, with respect to other species. C18:2 (50.4–62.3%), C18:1 (14.5–25.8%) and C16 (13.3–16.7%) were the most abundant fatty acid compounds in all accessions. PUFA/MUFA, C18:1/C16 and C18:1/C18:2 ratios significantly differed among genotypes, with special respect to T. monococcum, T. aestivum and T. macha. Significant differences were also observed for protein and total phenolic content, and antioxidant activity. The data indicate that some genotypes could represent interesting sources of specific compounds, with a potential use either as promising material for the selection of improved varieties or raw materials to manufacture foods with naturally enhanced content of health promoting compounds.

Introduction The intricate linkage between Freezing of Gait (FoG) and postural control in Parkinson's disease (PD) is unclear. We analyzed the impact of FoG on dynamic postural control. Methods 24 PD patients, 12 with (PD + FoG), 12 without FoG (PD-FoG), and 12 healthy controls, were assessed in ON state. Mobility and postural control were measured with clinical scales (UPDRS III, BBS, MPAS) and with kinematic and kinetic analysis during three tasks, characterized by levels of increasing difficulty to plan sequential movement of postural control: walk (W), gait initiation (GI) and sit-to-walk (STW). Results The groups were balanced by age, disease duration, disease severity, mobility and balance. During STW, the spatial distribution of COP trajectories in PD + FoG patients are spread over medial-lateral space more than in the PD-FoG (p

Beef is a common staple food in many countries, and there is a growing concern over misinformation of beef products, such as false claims of origin, species and production methods. In this study, we used a mass spectrometry-based metabolomics approach to study the metabolite profiles of beef samples purchased from local retailers in Hong Kong. Using multivariate analysis, beef samples from different a) geographical origins, namely the United States (US), Japan and Australia, and b) feeding regimes could be differentiated. We identified twenty-four metabolites to distinguish beef samples from different countries, ten metabolites to identify Angus beef samples from others and seven metabolites to discriminate Australian beef produced by the organic farming from that produced using other farming modes. Based on results of this study, it is concluded that metabolomics provides an efficient strategy for tracing and authenticating beef products to ensure their quality and to protect consumer rights.

At present, the geographical origin of extra virgin olive oils can be ensured by documented traceability, although chemical analysis may add information that is useful for possible confirmation. This preliminary study investigated the effectiveness of flash gas chromatography electronic nose and multivariate data analysis to perform rapid screening of commercial extra virgin olive oils characterized by a different geographical origin declared in the label. A comparison with solid phase micro extraction coupled to gas chromatography mass spectrometry was also performed. The new method is suitable to verify the geographic origin of extra virgin olive oils based on principal components analysis and discriminant analysis applied to the volatile profile of the headspace as a fingerprint. The selected variables were suitable in discriminating between “100% Italian” and “non-100% Italian” oils. Partial least squares discriminant analysis also allowed prediction of the degree of membership of unknown samples to the classes examined.

In the context of cancer delineation using positron emission tomography datasets, we present an innovative approach which purpose is to tackle the real-time, three-dimensional segmentation task in a full, or at least nearly full automatized way. The approach comprises a preliminary initialization phase where the user highlights a region of interest around the cancer on just one slice of the tomographic dataset. The algorithm takes care of identifying an optimal and user-independent region of interest around the anomalous tissue and located on the slice containing the highest standardized uptake value so to start the successive segmentation task. The three-dimensional volume is then reconstructed using a slice-by-slice marching approach until a suitable automatic stop condition is met. On each slice, the segmentation is performed using an enhanced local active contour based on the minimization of a novel energy functional which combines the information provided by a machine learning component, the discriminant analysis in the present study. As a result, the whole algorithm is almost completely automatic and the output segmentation is independent from the input provided by the user. Phantom experiments comprising spheres and zeolites, and clinical cases comprising various body districts (lung, brain, and head and neck), and two different radio-tracers (18 F-fluoro-2-deoxy- d -glucose, and 11C-labeled Methionine) were used to assess the algorithm performances. Phantom experiments with spheres and with zeolites showed a dice similarity coefficient above 90% and 80%, respectively. Clinical cases showed high agreement with the gold standard (R2 = 0.98). These results indicate that the proposed method can be efficiently applied in the clinical routine with potential benefit for the treatment response assessment, and targeting in radiotherapy.

The study investigates the sunlight photodegradation process of carminic acid, a natural red colourant used in beverages. For this purpose, both carminic acid aqueous standard solutions and sixteen different commercial beverages, ten containing carminic acid and six containing E120 dye, were subjected to photoirradiation. The results show different patterns of degradation, not only between the standard solutions and the beverages, but also from beverage to beverage. Due to the different beverage recipes, unpredictable reactions take place between the dye and the other ingredients. To identify the dye degradation products in a very complex scenario, a methodology was used, based on the combined use of principal component analysis with discriminant analysis and ultra-high-performance liquid chromatography coupled with tandem high resolution mass spectrometry. The methodology is unaffected by beverage composition and allows the degradation products of carminic acid dye to be identified for each beverage. © 2014 Elsevier Ltd. All rights reserved.

In this paper, a bivariate-heuristic model (modified Stevenson's method) and two multivariate statistical procedures (discriminant analysis and logistic regression) were used in order to assess and map landslide susceptibility in the north-western side of Daunia region (Apulia, Southern Italy). The whole Daunia region is characterized by complex and composite landslides, which are located on clayey slopes, near urban centers, affecting structures and infrastructures. The high predisposition to landsliding of the Daunia hillslopes is related to the very poor strength properties of clayey formations. The comparative analysis of landslide susceptibility using different methods, on the same test site and with the same inventory map allowed understanding the dependence of the results from the dataset and the capability of models under different levels of use, from expert to simple operator. By comparing the performance of the three models through the success rate curves, it emerges that the simple modified Stevenson's method produces reliable outcomes, comparable with those deriving from more complex multivariate statistical models. This result is related to the characteristics of clayey slopes, in which the landslide occurrence is so much controlled by the poor strength properties of the clayey formations that the multivariate analysis of a large set of morphometric, geological and land-use variables results to be somehow superfluous. This suggests that, for clayey slopes, a simple, easy-to-manage bivariate-heuristic model based on expert opinion can be used with reliable results. © 2013 Springer-Verlag Berlin Heidelberg

The aim of this study was to evaluate the potential use of an e-nose in combination with lateral flow immunoassays for rapid aflatoxin and fumonisin occurrence/co-occurrence detection in maize samples. For this purpose, 161 samples of corn have been used. Below the regulatory limits, single-contaminated, and co-contaminated samples were classified according to the detection ranges established for commercial lateral flow immunoassays (LFIAs) for mycotoxin determination. Correspondence between methods was evaluated by discriminant function analysis (DFA) procedures using IBM SPSS Statistics 22. Stepwise variable selection was done to select the e-nose sensors for classifying samples by DFA. The overall leave-out-one cross-validated percentage of samples correctly classified by the eight-variate DFA model for aflatoxin was 81%. The overall leave-out-one cross-validated percentage of samples correctly classified by the seven-variate DFA model for fumonisin was 85%. The overall leave-out-one cross-validated percentage of samples correctly classified by the nine-variate DFA model for the three classes of contamination (below the regulatory limits, single-contaminated, co-contaminated) was 65%. Therefore, even though an exhaustive evaluation will require a larger dataset to perform a validation procedure, an electronic nose (e-nose) seems to be a promising rapid/screening method to detect contamination by aflatoxin, fumonisin, or both in maize kernel stocks.

Principal component analysis (PCA) provides a method to explore functional brain connectivity. The aim of this study was to identify regional cerebral blood flow (rCBF) distribution differences between Alzheimer's disease (AD) patients and controls (CTR) by means of volume of interest (VOI) analysis and PCA. Thirty-seven CTR, 30 mild AD (mildAD) and 27 moderate AD (modAD) subjects were investigated using single photon emission computed tomography with (99m)Tc-hexamethylpropylene amine oxime. Analysis of covariance (ANCOVA), PCA, and discriminant analysis (DA) were performed on 54 VOIs. VOI analysis identified in both mildAD and modAD subjects a decreased rCBF in six regions. PCA in mildAD subjects identified four principal components (PCs) in which the correlated VOIs showed a decreased level of rCBF, including regions that are typically affected early in the disease. In five PCs, including parietal-temporal-limbic cortex, and hippocampus, a significantly lower rCBF in correlated VOIs was found in modAD subjects. DA significantly discriminated the groups. The percentage of subjects correctly classified was 95, 70, and 81 for CTR, mildAD and modAD groups, respectively. PCA highlighted, in mildAD and modAD, relationships not evident when brain regions are considered as independent of each other, and it was effective in discriminating groups. These findings may allow neurophysiological inferences to be drawn regarding brain functional connectivity in AD that might not be possible with univariate analysis.

Catechins and methylxanthines were determined in 92 green tea (GT) samples originating from Japan and China by using micellar electrokinetic chromatography with the addition of (2-hydroxypropyl)-β-cyclodextrin. GT samples showed high concentrations of (-)-epigallocatechin gallate and caffeine, with (-)-epigallocatechin, (-)-epicatechin gallate and (-)-epicatechin in relevant content and (+)-catechin, (-)-catechin and theobromine in much lower amounts. The amount of all the considered compounds was higher for Chinese GTs, with the exception of (-)-epicatechin gallate. Pattern recognition methods were applied to discriminate GTs according to geographical origin, which is an important factor to determine quality and reputation of a commercial tea product. Data analysis was performed by principal component analysis and hierarchical cluster analysis as exploratory techniques. Linear discriminant analysis and quadratic discriminant analysis were utilized as discrimination techniques, obtaining a very good rate of correct classification and prediction.

La segmentazione con criterio di split a due stadi si propone di definire procedure di classificazione/regressione ad albero in cui i predittori giocano un ruolo globale e locale nella spiegazione della variabile dipendente. Ciò avviene attraverso l’impiego di indici statistici, modelli parametrici e tecniche fattoriali. Il presente lavoro mostra un software interattivo per la costruzione di alberi esplorativi (Exploratory Trees – E.T. Software) con criterio di split in due stadi integrando diversi contributi metodologici offerti nell’ultimo decennio e al contempo mostrando nuovi sviluppi metodologici e computazionali.

A geometric discriminant criterion is a rule that allows to assign a new observation to preexisting groups. In this paper we propose a new method that works for data sets in two and three dimensions. It is totally data driven, without any model or density function assumptions, unlike usual the parametric approaches. In order to set up the procedure we exploit the geometrical properties of the Voronoi tessellation. The proposed discriminant analysis induces a space partition, that allows to deal efficiently with non linearly separable or non convex population structures. We analize the computational cost of the proposed procedure and the topological conditions concerning the group-conditional density functions that optimize the procedure performance. Because of its geometric properties, the method can be also usefully applied in statistical pattern recognition.

We assessed the performance of 4 methods of discriminant analysis using as independent variables the age and 16 serum tests, for correctly identifying patients with liver cirrhosis among hospitalized patients affected by chronic liver disease without signs of liver failure; 290 patients entered this study: on the basis of laparoscopy with or without liver biopsy, 152 patients had a diagnosis of liver cirrhosis and 138 were classified as chronic hepatitic patients. Due to the non-multinormal distribution of the variables used and to the unequality of the variance-covariance matrices, we compared the following 4 methods: linear discriminant function, quadratic discriminant function, non-parametric discriminant function and logistic regression. The Receiver Operating Characteristic (ROC) analysis was used to compare diagnostic ability of the assessed methods: the quadratic discriminant function was the best performing method. The predictive ability of this function was compared to that reported for percutaneous liver biopsy, showing that this simple statistical method using age and biochemical tests can efficiently identify liver cirrhosis in the setting of chronic liver disease, reducing the need for invasive diagnostic procedures.

When intense working conditions (e.g. production on continuous shifts) do not permit plant inactivity, conjugating both the continuous request of productivity and the constant demand of high performance, in terms of qualitative standards, becomes one of the most important factor in determining the profit of a business. The aim of this paper is to propose a method based both on wavelet and discriminant analysis for the identification of simulated anomalies superimposed to real signals carried out on mechanical equipment.