Fourier-Transform Infrared Spectroscopy as a Screening Tool for Osteosarcopenia in Community-Dwelling Older Women.

Osteosarcopenia is a complex geriatric syndrome characterized by the presence of both sarcopenia and osteopenia/osteoporosis. This condition increases rates of disability, falls, fractures, mortality, and mobility impairments in older adults. The purpose of this study was to analyze the Fourier-transform infrared (FTIR) spectroscopy diagnostic power for osteosarcopenia in community-dwelling older women (n = 64; 32 osteosarcopenic and 32 non-osteosarcopenia). FTIR is a fast and reproducible technique highly sensitive to biological tissues, and a mathematical model was created using multivariate classification techniques that denoted the graphic spectra of the molecular groups. Genetic algorithm and support vector machine regression (GA-SVM) was the most feasible model, achieving 80.0% of accuracy. GA-SVM identified 15 wave numbers responsible for class differentiation, in which several amino acids (responsible for the proper activation of the mammalian target of rapamycin) and hydroxyapatite (an inorganic bone component) were observed. Imaging tests and low availability of instruments that allow the observation of osteosarcopenia involve high health costs for patients and restrictive indications. Therefore, FTIR can be used to diagnose osteosarcopenia due to its efficiency and low cost and to enable early detection in geriatric services, contributing to advances in science and technology that are potential "conventional" methods in the future.

Stratification of Length of Stay Prediction following Surgical Cytoreduction in Advanced High-Grade Serous Ovarian Cancer Patients Using Artificial Intelligence; the Leeds L-AI-OS Score.

(1) Background: Length of stay (LOS) has been suggested as a marker of the effectiveness of short-term care. Artificial Intelligence (AI) technologies could help monitor hospital stays. We developed an AI-based novel predictive LOS score for advanced-stage high-grade serous ovarian cancer (HGSOC) patients following cytoreductive surgery and refined factors significantly affecting LOS. (2) Methods: Machine learning and deep learning methods using artificial neural networks (ANN) were used together with conventional logistic regression to predict continuous and binary LOS outcomes for HGSOC patients. The models were evaluated in a post-hoc internal validation set and a Graphical User Interface (GUI) was developed to demonstrate the clinical feasibility of sophisticated LOS predictions. (3) Results: For binary LOS predictions at differential time points, the accuracy ranged between 70-98%. Feature selection identified surgical complexity, pre-surgery albumin, blood loss, operative time, bowel resection with stoma formation, and severe postoperative complications (CD3-5) as independent LOS predictors. For the GUI numerical LOS score, the ANN model was a good estimator for the standard deviation of the LOS distribution by ± two days. (4) Conclusions: We demonstrated the development and application of both quantitative and qualitative AI models to predict LOS in advanced-stage EOC patients following their cytoreduction. Accurate identification of potentially modifiable factors delaying hospital discharge can further inform services performing root cause analysis of LOS.

Spectrochemical differentiation in gestational diabetes mellitus based on attenuated total reflection Fourier-transform infrared (ATR-FTIR) spectroscopy and multivariate analysis.

Gestational diabetes mellitus (GDM) is a hyperglycaemic imbalance first recognized during pregnancy, and affects up to 22% of pregnancies worldwide, bringing negative maternal-fetal consequences in the short- and long-term. In order to better characterize GDM in pregnant women, 100 blood plasma samples (50 GDM and 50 healthy pregnant control group) were submitted Attenuated Total Reflection Fourier-transform infrared (ATR-FTIR) spectroscopy, using chemometric approaches, including feature selection algorithms associated with discriminant analysis, such as Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA) and Support Vector Machines (SVM), analyzed in the biofingerprint region between 1800 and 900 cm-1 followed by Savitzky-Golay smoothing, baseline correction and normalization to Amide-I band (~ 1650 cm-1). An initial exploratory analysis of the data by Principal Component Analysis (PCA) showed a separation tendency between the two groups, which were then classified by supervised algorithms. Overall, the results obtained by Genetic Algorithm Linear Discriminant Analysis (GA-LDA) were the most satisfactory, with an accuracy, sensitivity and specificity of 100%. The spectral features responsible for group differentiation were attributed mainly to the lipid/protein regions (1462-1747 cm-1). These findings demonstrate, for the first time, the potential of ATR-FTIR spectroscopy combined with multivariate analysis as a screening tool for fast and low-cost GDM detection.

Prediction of meat quality traits in Nelore cattle by near-infrared reflectance spectroscopy.

The main definition for meat quality should include factors that affect consumer appreciation of the product. Physical laboratory analyses are necessary to identify factors that affect meat quality and specific equipment is used for this purpose, which is expensive and destructive, and the analyses are usually time consuming. An alternative method to performing several beef analyses is near-infrared reflectance spectroscopy (NIRS), which permits to reduce costs and to obtain faster, simpler, and nondestructive measurements. The objective of this study was to evaluate the feasibility of NIRS to predict shear force [Warner-Bratzler shear force (WBSF)], marbling, and color (*a = redness; b* = yellowness; and L* = lightness) in meat samples of uncastrated male Nelore cattle, that were approximately 2-yr-old. Samples of longissimus thoracis (n = 644) were collected and spectra were obtained prior to meat quality analysis. Multivariate calibration was performed by partial least squares regression. Several preprocessing techniques were evaluated alone and in combination: raw data, reduction of spectral range, multiplicative scatter correction, and 1st derivative. Accuracies of the calibration models were evaluated using the root mean square error of calibration (RMSEC), root mean square error of prediction (RMSEP), coefficient of determination in the calibration (R2C), and prediction (R2P) groups. Among the different preprocessing techniques, the reduction of spectral range provided the best prediction accuracy for all traits. The NIRS showed a better performance to predict WBSF (RMSEP = 1.42 kg, R2P = 0.40) and b* color (RMSEP = 1.21, R2P = 0.44), while its ability to accurately predict L* (RMSEP = 1.98, R2P = 0.16) and a* (RMSEP = 1.42, R2P = 0.17) was limited. NIRS was unsuitable to predict subjective meat quality traits such as marbling in Nelore cattle.

Area correlation constraint for the MCR-ALS quantification of cholesterol using EEM fluorescence data: A new approach.

This work demonstrates the use of a new additional constraint for the Multivariate Curve Resolution-Alternating Least Squares (MCR-ALS) algorithm called "area correlation constraint", introduced to build calibration models for Excitation Emission Matrix (EEM) data. We propose the application of area correlation constraint MCR-ALS for the quantification of cholesterol using a simulated data set and an experimental data system (cholesterol in a ternary mixture). This new constraint includes pseudo-univariate local regressions using the area of resolved profiles against reference values during the alternating least squares optimization, to provide directly accurate quantifications of a specific analyte in concentration units. In the two datasets investigated in this work, the new constraint retrieved correctly the analyte and interference spectral profiles and performed accurate estimations of cholesterol concentrations in test samples. This the first study using the proposed area constraint using EEM measurements. This new constraint approach emerges as a new possibility to be tested in general cases of second-order multivariate calibration data in the presence of unknown interferents or in more involved higher order calibration cases.

NIRS and iSPA-PLS for predicting total anthocyanin content in jaboticaba fruit.

The aim of this study was to evaluate the potential of the successive projection algorithm for interval selection in partial least squares (iSPA-PLS) together with near-infrared reflectance spectroscopy (NIRS) as a feasible method to determine the total anthocyanin content (TAC) of intact jaboticaba fruit [Myrciaria jaboticaba (Vell.) O. Berg]. A total of 579 jaboticaba fruit were collected in three different harvests in three separate years (2011 and 2013). The correlation coefficients between the predicted and measured TAC were between 0.65 and 0.89, the RMSEPs were 7.55 g kg(-1) and 9.35 g kg(-1) (good accuracy) for prediction set, respectively. The RPD ratios for TAC were in the range of 2.57-3.19 with iSPA-PLS, which showed better predictive performance (acceptable precision). These results suggest that the NIR spectroscopy and wavelength selection (iSPA-PLS) algorithm can be used to determine the TAC of intact jaboticaba fruit.

A colorimetric microwell method using a desktop scanner for biochemical assays.

A method for rapid, inexpensive and sensitive simultaneous analysis of glucose, creatinine, triglycerides, total cholesterol and total protein is needed to analyze blood. The proposed method is based on the production of a specific color after reaction. The method was adapted to a 64-microwell plate format, and it uses the transparency scanner feature of a commercially available desktop scanner. Each microwell plate had an 8×8 array of flat-bottomed 250µL microwells, and these microwells were used to simultaneously house the solutions for clinical assay. The scanned image was saved in TIFF format in a portable computer and then processed using a Graphic User Interface (GUI) designed in our laboratory to obtain analytical curves and to automate the mathematics and statistics calculations. This automation improved the analytical frequency of the method. The results showed that it is possible to measure a few microliters of solution with exactitude and precision better than 5.30%. The measured concentration ranges of glucose, triglycerides, creatinine, total cholesterol and total protein were 0.781 to 100, 1.56 to 200, 0.031 to 4.0, 1.56 to 200mg dL(-1) and 0.031 to 4.0g dL(-1), respectively. The limits of detection were 16.2, 51.7, 0.12, 41.5mg dL(-1) and 0.62g dL(-1) for glucose, triglycerides, creatinine, total cholesterol and total protein, respectively. The recoveries were from 98.7% to 101.3% for total cholesterol, 98.7% to 124.9% for triglycerides, 54.2% to 98.3% for total protein, 89.6% to 101% for glucose and 65.7% to 115.4% for creatinine. The results provided by the scanner were compared with those obtained with a commercial photometer and did not show significant differences at a confidence level of 95%. Good results were obtained for the correlation coefficient and Root Mean Square Error of Prediction (RMSEP) values for the five parameters, especially the total cholesterol and creatinine. The RMSEP values for glucose, creatinine, triglyceride, total cholesterol and total protein were 8.05, 0.28, 7.69, 1.41mg dL(-1) and 2.2g dL(-1), respectively.

Application of correlation constrained multivariate curve resolution alternating least-squares methods for determination of compounds of interest in biodiesel blends using NIR and UV-visible spectroscopic data.

This study describes two applications of a variant of the multivariate curve resolution alternating least squares (MCR-ALS) method with a correlation constraint. The first application describes the use of MCR-ALS for the determination of biodiesel concentrations in biodiesel blends using near infrared (NIR) spectroscopic data. In the second application, the proposed method allowed the determination of the synthetic antioxidant N,N'-Di-sec-butyl-p-phenylenediamine (PDA) present in biodiesel mixtures from different vegetable sources using UV-visible spectroscopy. Well established multivariate regression algorithm, partial least squares (PLS), were calculated for comparison of the quantification performance in the models developed in both applications. The correlation constraint has been adapted to handle the presence of batch-to-batch matrix effects due to ageing effects, which might occur when different groups of samples were used to build a calibration model in the first application. Different data set configurations and diverse modes of application of the correlation constraint are explored and guidelines are given to cope with different type of analytical problems, such as the correction of matrix effects among biodiesel samples, where MCR-ALS outperformed PLS reducing the relative error of prediction RE (%) from 9.82% to 4.85% in the first application, or the determination of minor compound with overlapped weak spectroscopic signals, where MCR-ALS gave higher (RE (%)=3.16%) for prediction of PDA compared to PLS (RE (%)=1.99%), but with the advantage of recovering the related pure spectral profile of analytes and interferences. The obtained results show the potential of the MCR-ALS method with correlation constraint to be adapted to diverse data set configurations and analytical problems related to the determination of biodiesel mixtures and added compounds therein.

Predicting soluble solid content in intact jaboticaba [Myrciaria jaboticaba (Vell.) O. Berg] fruit using near-infrared spectroscopy and chemometrics.

The aim of this study was to evaluate the potential of near-infrared reflectance spectroscopy (NIR) as a rapid and non-destructive method to determine soluble solid content (SSC) in intact jaboticaba [Myrciaria jaboticaba (Vell.) O. Berg] fruit. Multivariate calibration techniques were compared with pre-processed data and variable selection algorithms, such as partial least squares (PLS), interval partial least squares (iPLS), a genetic algorithm (GA), a successive projections algorithm (SPA) and nonlinear techniques (BP-ANN, back propagation of artificial neural networks; LS-SVM, least squares support vector machine) were applied to building the calibration models. The PLS model produced prediction accuracy (R(2)=0.71, RMSEP=1.33 °Brix, and RPD=1.65) while the BP-ANN model (R(2)=0.68, RMSEM=1.20 °Brix, and RPD=1.83) and LS-SVM models achieved lower performance metrics (R(2)=0.44, RMSEP=1.89 °Brix, and RPD=1.16). This study was the first attempt to use NIR spectroscopy as a non-destructive method to determine SSC jaboticaba fruit.

The use of EEM fluorescence data and OPLS/UPLS-DA algorithm to discriminate between normal and cancer cell lines: a feasibility study.

Excitation emission matrix (EEM) fluorescence spectroscopy combined with the OPLS method has been investigated as a promising tool to discriminate between normal and cancer cell lines in two datasets: (i) using several types of normal and cancer cells (including 3T3, ARPE, HEK, HepG2, HeLa, HT-29 and 786-0 cells); (ii) considering the expression of matrix metalloproteinase-2 and -9 (MMP-2 and MMP-9) in suspensions of HEK and 786-0 cell lines. Partial Least Squares-Discriminant Analysis (PLS-DA) using the score matrix from PARAFAC (Parallel Factor Analysis), UPLS-DA (Unfolded Partial Least Squares with Discriminant Analysis) and orthogonal projection to latent structures (OPLS) were used as the bases for the discrimination models. UPLS-DA presented relevant performance for cancer cells in both datasets, with 100% and 66.7% correct prediction for first and second cases, respectively, and poor discrimination relative to normal cells in the first dataset (25%). By using the OPLS, we achieved 75% correct prediction for normal cells and maintained 100% concordance for cancer objects. On applying OPLS to the second dataset, we obtained 100% correct prediction in both classes (normal and cancer) for calibration and prediction sets. These results suggest that EEM fluorescence spectroscopy combined with chemometrics could be used as a clinical tool for cancer cell detection based on intrinsic biomolecular signatures.

Feature selection strategies for identification of Staphylococcus aureus recovered in blood cultures using FT-IR spectroscopy successive projections algorithm for variable selection: a case study.

Staphylococcus aureus is one of the leading causes of bacteremia, with high levels of accompanying morbidity and mortality. Current gold standard for the detection of S. aureus is very time-consuming, typically taking 24h or longer. We set out to determine whether Fourier-transform infrared spectroscopy (FT-IR) combined with variable selection techniques, such as, genetic algorithm-linear discriminant analysis (GA-LDA) and successive projection algorithm-linear discriminant analysis (SPA-LDA) could be applied to detect this pathogen of bloodstream infection in samples based on the unique spectral "fingerprints" of their biochemical composition. Thirty real blood samples from healthy volunteers were contaminated with five different concentrations (10(7) until 10(3) CFU/mL) of microorganism and it analyzed by IR spectroscopy. The resulting GA-LDA model successfully classified all test samples with respect to their concentration in contaminated blood using only 18 wavenumbers. Discriminant functions revealed that GA-LDA clearly segregated different microorganism concentrations and the variable selected confirmed the chemical entities associated with the microorganism. The current study indicates that IR spectroscopy with feature selection techniques have the potential to provide one rapid approach for whole-organism fingerprint diagnostic microbial directly in blood culture.

The use of near infrared spectroscopy and multivariate techniques to differentiate Escherichia coli and Salmonella Enteritidis inoculated into pulp juice.

This study shows the application and usefulness of near infrared (NIR) transflectance spectra measurements in the identification and classification of Escherichia coli and Salmonella Enteritidis from commercial fruit pulp (pineapple). Principal component analysis (PCA), soft independent modeling of class analogy (SIMCA) analysis and partial least-squares discriminant analysis (PLS-DA) were used in the analysis. It was not possible to obtain total separation between the samples using PCA and SIMCA. PLS-DA presented good performance achieving prediction ability of 87.5% for E. coli and 88.3% for S. Enteritidis, respectively. For the best models, the sensitivity and specificity was 0.87 and 0.83 for PLS-DA with second derivative spectra. These results suggest that NIR spectroscopy and PLS-DA can be used to discriminate and detect bacteria in fruit pulp for modeling linear class boundaries.

Total anthocyanin content determination in intact açaí (Euterpe oleracea Mart.) and palmitero-juçara (Euterpe edulis Mart.) fruit using near infrared spectroscopy (NIR) and multivariate calibration.

The aim of this study was to evaluate near-infrared reflectance spectroscopy (NIR), and multivariate calibration potential as a rapid method to determinate anthocyanin content in intact fruit (açaí and palmitero-juçara). Several multivariate calibration techniques, including partial least squares (PLS), interval partial least squares, genetic algorithm, successive projections algorithm, and net analyte signal were compared and validated by establishing figures of merit. Suitable results were obtained with the PLS model (four latent variables and 5-point smoothing) with a detection limit of 6.2 g kg(-1), limit of quantification of 20.7 g kg(-1), accuracy estimated as root mean square error of prediction of 4.8 g kg(-1), mean selectivity of 0.79 g kg(-1), sensitivity of 5.04×10(-3) g kg(-1), precision of 27.8 g kg(-1), and signal-to-noise ratio of 1.04×10(-3) g kg(-1). These results suggest NIR spectroscopy and multivariate calibration can be effectively used to determine anthocyanin content in intact açaí and palmitero-juçara fruit.

Near infrared spectroscopy and multivariate calibration for simultaneous determination of glucose, triglycerides and high-density lipoprotein in animal plasma.

The quantitative analysis of glucose, triglycerides and high-density lipoprotein (HDL) in rat plasma without sample pre-treatment using direct near-infrared spectroscopy was studied. Comparison was made of several multivariate calibration techniques and algorithms for data pre-processing and variable selection, including partial least squares (PLS), interval partial least squares (iPLS), genetic algorithm (GA) and successive projections algorithm (SPA). Variable selection yielded good results for the correlation coefficient and Root Mean Square Error of Prediction (RMSEP) values for the three parameters, especially triglycerides. The RMSEP values for glucose, triglycerides and HDL produced by the PLS model were 6.08, 16.07 and 2.03 mg dl(-1), respectively. F tests and t-tests were performed to compare the results of the models with each other and with a reference method. These results suggests that the PLS method can be used to simultaneously determine the concentrations of glucose, triglycerides and HDL in complicated biological fluids with NIR spectroscopy, offering an alternative analysis in animals.

Non-destructive detection of adulterated tablets of glibenclamide using NIR and solid-phase fluorescence spectroscopy and chemometric methods.

This study describes a method for non-destructive detection of adulterated glibenclamide tablets. This method uses near infrared spectroscopy (NIRS) and fluorescence spectroscopy along with chemometric tools such as Soft Independent Modeling of Class Analogy (SIMCA), Partial Least Squares-Discriminant Analysis (PLS-DA) and Unfolded Partial Least Squares with Discriminant Analysis (UPLS-DA). Both brand name (Daonil) and generic glibenclamide tablets were used for analysis. The levels of glibenclamide in each type of tablet were evaluated by derivative spectrophotometry in the ultraviolet region. The results obtained from the NIR and fluorescence spectroscopy along with those obtained from multivariate data classification show that this combined technique is an effective way to detect adulteration in drugs for the treatment of diabetes. In the future, this method may be extended to detect different types of counterfeit medications.

Dissolution testing of isoniazid, rifampicin, pyrazinamide and ethambutol tablets using near-infrared spectroscopy (NIRS) and multivariate calibration.

This work utilized the near-infrared spectroscopy (NIRS) and multivariate calibration to measure the percentage drug dissolution of four active pharmaceutical ingredients (APIs) (isoniazid, rifampicin, pyrazinamide and ethambutol) in finished pharmaceutical products produced in the Federal University of Rio Grande do Norte (Brazil). The conventional analytical method employed in quality control tests of the dissolution by the pharmaceutical industry is high-performance liquid chromatography (HPLC). The NIRS is a reliable method that offers important advantages for the large-scale production of tablets and for non-destructive analysis. NIR spectra of 38 samples (in triplicate) were measured using a Bomen FT-NIR 160 MB in the range 1100-2500nm. Each spectrum was the average of 50 scans obtained in the diffuse reflectance mode. The dissolution test, which was initially carried out in 900mL of 0.1N hydrochloric acid at 37±0.5°C, was used to determine the percentage a drug that dissolved from each tablet measured at the same time interval (45min) at pH 6.8. The measurement of the four API was performed by HPLC (Shimadzu, Japan) in the gradiente mode. The influence of various spectral pretreatments (Savitzky-Golay smoothing, Multiplicative Scatter Correction (MSC), and Savitzky-Golay derivatives) and multivariate analysis using the partial least squares (PLS) regression algorithm was calculated by the Unscrambler 9.8 (Camo) software. The correlation coefficient (R(2)) for the HPLC determination versus predicted values (NIRS) ranged from 0.88 to 0.98. The root-mean-square error of prediction (RMSEP) obtained from PLS models were 9.99%, 8.63%, 8.57% and 9.97% for isoniazid, rifampicin, ethambutol and pyrazinamide, respectively, indicating that the NIR method is an effective and non-destructive tool for measurement of drug dissolution from tablets.