In this work, near infrared (NIR) spectroscopy and multivariate data analysis were investigated as a fast and non-disruptive method to classify green coffee beans on continents and countries bases. ...FT-NIR spectra of 191 coffee samples, origin from 2 continents and 9 countries, were acquired by two different laboratories.
Laboratory-independent Partial Least Square-Discriminant Analysis and interval PLS-DA models were developed by following a hierarchical approach, i.e. considering at first the continent and then the country of origin as discrimination rule.
The best continent-based classification model was able to identify correctly more than 98% in prediction, whereas 100% of them were correctly predicted by the best country-based classification model. The inter-laboratory reliability of the proposed method was confirmed by McNemar test, since no significant differences (P > 0.05) were found. Furthermore, a validation was performed predicting the spectral test set of a laboratory using the model developed by the other one.
•NIR Spectroscopy investigated to identify the geographical origin of green coffee.•Partial Least Square-Discriminant Analysis applied to build classification models.•Coffee origin for both continent and country was tested as discrimination parameter.•Variable selection applied to NIR spectra allowed improving the model performance.•Inter-laboratory comparison of the classification results was made by McNemar test.
•Accurate NIR calibration models were obtained for main quality indexes of apples.•Total phenolics and antioxidant activity of apples could be predicted by NIR-based models.•Classification models ...were able to discriminate apple lots on the basis of storage time.
Apples can be stored for long time under controlled temperature and atmosphere conditions, and therefore, non-destructive and rapid tools are required to assess fruit quality and to monitor changes during the postharvest period. The aim of this study was to evaluate the feasibility of NIR spectroscopy to optimize postharvest apple management and to follow changes in fruit quality during storage. An FT-NIR system operating in diffuse reflectance in the range 12,500–3600cm−1 was used to evaluate the physico-chemical (dry matter, soluble solids, colour and firmness) and some nutraceutical characteristics (total phenolics, total flavonoids and antioxidant activity) of ‘Golden Delicious’ apples, which were stored for about six months at 1°C in controlled atmosphere, over two subsequent years. Spectral data were elaborated by PLS regression and LDA classification techniques. Good correlation models between spectral data and chemical and physical parameters were obtained for soluble solids, a* colour coordinate and firmness (0.81<R2<0.90 in calibration and 0.79<R2<0.89 in cross validation). Even higher correlation values (0.89<R2<0.95 in calibration and 0.86<R2<0.92 in cross validation) were obtained for indexes correlated to the antioxidant capacity of apples. The classification technique Linear Discriminant Analysis was applied to spectral data, in order to discriminate apples on the basis of storage time. Average correct classification was higher than 93% in validation and close to 100% in calibration, indicating high potential of NIR spectroscopy for the estimation of storage time of apple lots.
The aims were: (1) to follow the freshness decay of minced beef stored in high-oxygen modified atmosphere packaging at different temperatures (4.3, 8.1 and 15.5
°C) by applying traditional methods ...(microbiological counts, color evaluation, thiobarbituric acid assay TBA, headspace gas composition) and e-nose; (2) to model the decay kinetics to obtain information about the maximum shelf life as function of storage conditions. The minced beef, packaged in modified atmosphere was supplied by a manufacturer at the beginning of its commercial life. The study demonstrated the ability of the traditional methods to describe the kinetics of freshness decay. The modeling of the experimental data and the comparison with microbiological or chemical thresholds allowed the setting, for each index, of a stability time above which the meat was no longer acceptable. The quality decay of meat was also evaluated by the headspace fingerprint of the same set of samples by means of a commercial e-nose. A clear discrimination between “fresh” and “old” samples was obtained using PCA and CA, determining at each temperature a specific range of stability time. The mean value of the stability times calculated for each index was 9
days at 4.3
°C (recommended storage temperature), 3–4
days at 8.1
°C (usual temperature in household refrigerators) and 2
days at 15.5
°C (abuse temperature). Resolution of the stability times allowed calculation of mean Q
10 values, i.e. the increase in rate for a 10
°C increase in temperature.
The results show that the
Q
10 values from the traditional methods (3.6–4.0 range) overlapped with those estimated with e-nose and color indexes (3.4 and 3.9, respectively).
When dealing with datasets comprising high-dimensional points, it is usually advantageous to discover some data structure. A fundamental information needed to this aim is the minimum number of ...parameters required to describe the data while minimizing the information loss. This number, usually called intrinsic dimension, can be interpreted as the dimension of the manifold from which the input data are supposed to be drawn. Due to its usefulness in many theoretical and practical problems, in the last decades the concept of intrinsic dimension has gained considerable attention in the scientific community, motivating the large number of intrinsic dimensionality estimators proposed in the literature. However, the problem is still open since most techniques cannot efficiently deal with datasets drawn from manifolds of high intrinsic dimension and nonlinearly embedded in higher dimensional spaces. This paper surveys some of the most interesting, widespread used, and advanced state-of-the-art methodologies. Unfortunately, since no benchmark database exists in this research field, an objective comparison among different techniques is not possible. Consequently, we suggest a benchmark framework and apply it to comparatively evaluate relevant state-of-the-art estimators.
Recently, a great deal of research work has been devoted to the development of algorithms to estimate the intrinsic dimensionality (id) of a given dataset, that is the minimum number of parameters ...needed to represent the data without information loss. id estimation is important for the following reasons: the capacity and the generalization capability of discriminant methods depend on it; id is a necessary information for any dimensionality reduction technique; in neural network design the number of hidden units in the encoding middle layer should be chosen according to the id of data; the id value is strongly related to the model order in a time series, that is crucial to obtain reliable time series predictions.
Although many estimation techniques have been proposed in the literature, most of them fail on noisy data, or compute underestimated values when the id is sufficiently high. In this paper, after reviewing some of the most important id estimators related to our work, we provide a theoretical motivation of the bias that causes the underestimation effect, and we present two id estimators based on the statistical properties of manifold neighborhoods, which have been developed in order to reduce this effect. We exhaustively evaluate the proposed techniques on synthetic and real datasets, by employing an objective evaluation measure to compare their performance with those achieved by state of the art algorithms; the results show that the proposed methods are promising, and produce reliable estimates also in the difficult case of datasets drawn from non-linearly embedded manifolds, characterized by high id.
Display omitted
•Non-destructive diagnostic tool was proposed for olive ripening monitoring.•Image analysis, vis/NIR and NIR spectroscopy, and chemometrics were used.•Replacing of traditional ...subjective method in a view of fully automated application.•Comparison of modelling performance of benchtop FT-NIR and portable vis/NIR devices.•Selection of 12 (1.5 nm wide) vis/NIR bands to improve classification models.
A diagnostic visible/near infrared tool calibrated by means of image analysis,was proposed to evaluate the maturation degree of oil olives in order to replace traditional subjective methods in a view of future fully automated applications.
Thirteen varieties of Olea europaea deriving from four regions of the south of Italy were analyzed. In order to objectify the ripening stage assessment, the RGB image was acquired. Spectroscopic analyses were performed using a benchtop FT-NIR and a portable vis/NIR instrument. The benchtop device was equipped with an optical fiber probe and the spectra were collected over the 800–2500 nm range, nominal resolution of 1.6 nm; the portable spectrophotometer cover the range of 400–1000 nm, nominal resolution 0.3 nm. The olive spectral data were modelled using Partial Least Squares - Discriminant Analysis (PLS-DA). The prediction capability reached by the global model (13 varieties were used) obtained from data acquired with both the devices results promising. The PLS-DA models calculated on the olives from Calabria, Sardinia and Abruzzo revealed high prediction capabilities, i.e. sensitivity, specificity and accuracy higher than 83%. The prediction capability of Apulia samples could be improved increasing the variability of the samples since for this region only 3 sampling times were considered. To compare the modelling performance between the benchtop FT-NIR and the portable vis/NIR device, a McNemar’s test was performed resulting no significant difference between the PLS-DA global models. Finally, considering the good performance of the vis/NIR model, a variable selection using the interval PLS (iPLS) algorithm was applied. To reduce the complexity keeping the performance of the model built using the whole vis/NIR spectra (1647 variables), 12 bands (1.5 nm wide) were selected. The new model showed an improvement in terms of model stability and complexity (Sensitivity 86%; Specificity 87%; Accuracy 87%) than the two global models built with the whole vis/NIR and NIR range.
The classification performance provided the groundwork for the development of (i) simplified systems for a direct olives ripening determination on the olive tree, and (ii) automated systems to be applied both in field and at the mill for olives sorting according to the ripening degree.
In the past decades, a great deal of research work has been devoted to the development of systems that could improve radiologists' accuracy in detecting lung nodules. Despite the great efforts, the ...problem is still open. In this paper, we present a fully automated system processing digital postero-anterior (PA) chest radiographs, that starts by producing an accurate segmentation of the lung field area. The segmented lung area includes even those parts of the lungs hidden behind the heart, the spine, and the diaphragm, which are usually excluded from the methods presented in the literature. This decision is motivated by the fact that lung nodules may be found also in these areas. The segmented area is processed with a simple multiscale method that enhances the visibility of the nodules, and an extraction scheme is then applied to select potential nodules. To reduce the high number of false positives extracted, cost-sensitive support vector machines (SVMs) are trained to recognize the true nodules. Different learning experiments were performed on two different data sets, created by means of feature selection, and employing Gaussian and polynomial SVMs trained with different parameters; the results are reported and compared. With the best SVM models, we obtain about 1.5 false positives per image (fp/image) when sensitivity is approximately equal to 0.71; this number increases to about 2.5 and 4 fp/image when sensitivity is ap0.78 and ap0.85, respectively. For the highest sensitivity (ap0.92 and 1.0), we get 7 or 8 fp/image
A successful pregnancy is an exceptional event on dialysis. Few data are available comparing pregnancy rates on dialysis, transplantation and the overall population. The aim of the study was to ...assess the incidence of live births from mothers on chronic dialysis compared with the overall population and with kidney transplant patients.
The setting of the study is in Italy between 2000-12. Data on dialysis was aquired by phone inquiries that were carried out between June and September, 2013, involving all the public dialysis centres in Italy; the result was a 100% response rate. The date included was end-stage renal disease, type of dialysis, residual glomerular filtration rate, changes in dialysis and therapy, hospitalization; week of birth, birth weight, centile; and outcome of mother and child. Information on transplantation was acquired by inquiry by the kidney and pregnancy study group who were contacted by phone or e-mail; the result was a 60% response rate. Data concerning prevalence of women in childbearing age (20-45) were obtained from the Italian Dialysis and Transplant Registries (2010-11 update). Official site of the Italian Ministry of Health.
During the study period, 23 women on dialysis (three on peritoneal dialysis) delivered live-born babies and one woman delivered twins (24 babies). Three babies died in the first weeks-months of life (including one twin); 19 of 21 singletons with available data were pre-term (33.3% <34 weeks); the prevalence of children <10th gestational age-adjusted centile was 33.3%. Birth weight and gestational age were lower in children from on-dialysis mothers as compared with 110 pregnancies following kidney graft, (weight: 1200 versus 2500 g; gestational age: 30 versus 36 weeks; P < 0.001). Incidence of live-born babies was inferred as 0.7-1.1 per 1000 female dialysis patients aged 20-45 and 5.5-8.3 per 1000 grafted patients in the same age range (Italian live-birth rates: 72.5 per 1000 women aged 20-45 years).
Having a baby while on dialysis is rare but not impossible, though early mortality remains high. There is a 'scale of probability' estimating that women on dialysis have a 10-fold lower probability of delivering a live-born baby than those who have undergone renal transplantation, who in turn have a 10-fold lower probability of delivering a live-born baby as compared with the overall population.
Highlights • We evaluated the additional value of precontrast CT in polytrauma patient. • We evaluated the radiation dose CT in polytrauma patient. • Precontrast CT scan does not have an additional ...value in polytrauma patient. • Precontrast CT expose patient to an unjustified radiation dose.
Abstract
Funding Acknowledgements
Type of funding sources: Public grant(s) – National budget only. Main funding source(s): DIASyM cluster
Heart Failure is a heterogeneous clinical syndrome ...characterized by the inability to pump enough blood and thus provide enough oxygen to the human body. The current classification of heart failure is based on the left ventricular ejection faction (LVEF). Although the classification clearly defines heart failure subtypes and stages, and clinical guidelines are available for their management, it is preferable to prevent the disorder or take control over it from its onset. Evidence suggests strong molecular connections between human metabolism and heart failure. Exploiting the unique and diverse molecular characteristics that drive heart failure could be an effective strategy for patient stratification, to identify new therapeutic targets and to personalize treatments.
The current research attempts to identify clinically relevant metabolic subgroups of individuals with the help of a similarity network-based approach, based on targeted proteomics data of a heart failure cohort. With Olink targeted proteomics technology, six protein abundance panels (547 distinct proteins) from 3063 subjects (1103 females) were obtained. Using nonlinear combination methods, the created networks were integrated into a single similarity network and then used for clustering with unbiased network-based approaches.
The results have distinguished the subgroups with worse outcome prediction, supported by significantly different clinical profiles. Surprisingly, the significant clinical parameters have shown the dominance of not only classical metabolic signatures (diabetes) but also parameters related to kidney and liver function. The differential protein abundance profiles were then compared among clusters to derive novel insights into clinical subtypes, and have highlighted a significant contribution of inflammatory proteins to the unbiased subtyping rationale.
The study attempted to improve previous patient stratification efforts, by having a large sample size with a broader variety of molecular variables. These results will pave the way to precision medicine, and will allow indirectly to integrate and/or adjust the analysis by adding clinically relevant social parameters such as lifestyle, parents’ anamnesis, demographics that are often left out due to their lower contribution at the pre-processing steps in comparison to more objective laboratory parameters. Previous studies have already clustered the heart failure population, based on various clinical and biochemical parameters. However, the clustering was guided by well-established metabolic parameters and thus, the results can be highly biased. The current project allows an unbiased, data-driven approach that holds more potential to identify novel and clinically relevant metabolic subgroups and to define new molecular targets.