Linear Discriminant Analysis Model Research Articles

Authenticating peanut butter and yoghurt in the Kumasi Metropolis of Ghana using near infrared spectroscopy

AbstractPeanut butter and yoghurt are targeted for adulteration intended at consumer deception. This study aimed to fingerprint and detect peanut butter and yoghurt adulteration with cassava flour and starch using Near Infrared Spectroscopy (NIRS) in a quasi-experimental approach. Ingredients for laboratory sample preparation were obtained from the Kumasi Metropolis. Peanut butter was adulterated at 1, 3, 5, 10, 15, 20% w/w and yoghurt at 0.25, 0.5, 1, 3, 5, 10, 15, 20, 25, 45, 50% w/w. Selected concentrations mimicked practices on the market. Marketed products were randomly sampled from six markets in the Kumasi Metropolis to validate the study models. Samples were scanned with a hand-held NIRS in triplicates. Chemometric (Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA) and Partial Least Square Regression (PLSR) models) statistical methods were employed to develop classification and prediction models. Peaks with spectral bands such as 1050 , 1200 and 1450 nm were observed for peanut butter and 990–1100 nm, 1100–1200 nm and 1300–1408 nm were observed for yoghurt in the NIR spectrum. Some yoghurt brands were suspected of containing cassava starch, while Peanut butter from the different markets differed based on classification models. Cassava flour and starch concentrations were quantitatively predicted by PLSR with an R2CV of 0.98 and an error of 0.9 g/100 g (low error).

Predictive model for epileptogenic tubers from all tubers in patients with tuberous sclerosis complex based on 18F-FDG PET: an 8-year single-centre study

BackgroundMore than half of patients with tuberous sclerosis complex (TSC) suffer from drug-resistant epilepsy (DRE), and resection surgery is the most effective way to control intractable epilepsy. Precise preoperative localization of epileptogenic tubers among all cortical tubers determines the surgical outcomes and patient prognosis. Models for preoperatively predicting epileptogenic tubers using 18F-FDG PET images are still lacking, however. We developed noninvasive predictive models for clinicians to predict the epileptogenic tubers and the outcome (seizure freedom or no seizure freedom) of cortical tubers based on 18F-FDG PET images.MethodsForty-three consecutive TSC patients with DRE were enrolled, and 235 cortical tubers were selected as the training set. Quantitative indices of cortical tubers on 18F-FDG PET were extracted, and logistic regression analysis was performed to select those with the most important predictive capacity. Machine learning models, including logistic regression (LR), linear discriminant analysis (LDA), and artificial neural network (ANN) models, were established based on the selected predictive indices to identify epileptogenic tubers from multiple cortical tubers. A discriminating nomogram was constructed and found to be clinically practical according to decision curve analysis (DCA) and clinical impact curve (CIC). Furthermore, testing sets were created based on new PET images of 32 tubers from 7 patients, and follow-up outcome data from the cortical tubers were collected 1, 3, and 5 years after the operation to verify the reliability of the predictive model. The predictive performance was determined by using receiver operating characteristic (ROC) analysis.ResultsPET quantitative indices including SUVmean, SUVmax, volume, total lesion glycolysis (TLG), third quartile, upper adjacent and standard added metabolism activity (SAM) were associated with the epileptogenic tubers. The SUVmean, SUVmax, volume and TLG values were different between epileptogenic and non-epileptogenic tubers and were associated with the clinical characteristics of epileptogenic tubers. The LR model achieved the better performance in predicting epileptogenic tubers (AUC = 0.7706; 95% CI 0.70–0.83) than the LDA (AUC = 0.7506; 95% CI 0.68–0.82) and ANN models (AUC = 0.7425; 95% CI 0.67–0.82) and also demonstrated good calibration (Hosmer‒Lemeshow goodness-of-fit p value = 0.7). In addition, DCA and CIC confirmed the clinical utility of the nomogram constructed to predict epileptogenic tubers based on quantitative indices. Intriguingly, the LR model exhibited good performance in predicting epileptogenic tubers in the testing set (AUC = 0.8502; 95% CI 0.71–0.99) and the long-term outcomes of cortical tubers (1-year outcomes: AUC = 0.7805, 95% CI 0.71–0.85; 3-year outcomes: AUC = 0.8066, 95% CI 0.74–0.87; 5-year outcomes: AUC = 0.8172, 95% CI 0.75–0.87).ConclusionsThe 18F-FDG PET image-based LR model can be used to noninvasively identify epileptogenic tubers and predict the long-term outcomes of cortical tubers in TSC patients.

Identification of Milk Adulteration in Camel Milk Using FT-Mid-Infrared Spectroscopy and Machine Learning Models.

Camel milk, esteemed for its high nutritional value, has long been a subject of interest. However, the adulteration of camel milk with cow milk poses a significant threat to food quality and safety. Fourier-transform infrared spectroscopy (FT-MIR) has emerged as a rapid method for the detection and quantification of cow milk adulteration. Nevertheless, its effectiveness in conveniently detecting adulteration in camel milk remains to be determined. Camel milk samples were collected from Alxa League, Inner Mongolia, China, and were supplemented with varying concentrations of cow milk samples. Spectra were acquired using the FOSS FT6000 spectrometer, and a diverse set of machine learning models was employed to detect cow milk adulteration in camel milk. Our results demonstrate that the Linear Discriminant Analysis (LDA) model effectively distinguishes pure camel milk from adulterated samples, maintaining a 100% detection rate even at cow milk addition levels of 10 g/100 g. The neural network quantitative model for cow milk adulteration in camel milk exhibited a detection limit of 3.27 g/100 g and a quantification limit of 10.90 g/100 g. The quantitative model demonstrated excellent precision and accuracy within the range of 10-90 g/100 g of adulteration. This study highlights the potential of FT-MIR spectroscopy in conjunction with machine learning techniques for ensuring the authenticity and quality of camel milk, thus addressing concerns related to food integrity and consumer safety.

Mut (Mersin) Bölgesinde Hasat Edilen Natürel Sızma Zeytinyağlarının Senkron Floresans Spektroskopisi ve Kemometri ile Orijinalliğinin Doğrulanması: LDA ve HCA Model Yaklaşımları

In this work, synchronous fluorescence (SyF) spectroscopy data of extra virgin olive oils (VOOs) harvested from two different harvest terms (2019 and 2020 October) in Mut (Mersin) region and other edible oils were analyzed by linear discriminant analysis (LDA) and hierarchical cluster analysis (HCA) to verify their authenticity. The emission and excitation monochromators were synchronously employed and the data were obtained in the emission wavelength region of 200–800 nm. A robust LDA model was gained to authenticate the extra VOOs and other edible oils: 98.28% of the tested samples were correctly predicted, and HCA dendograms showed successful discriminations with high sensitivity and specificity. The findings show the capability of SyF spectroscopy coupled with chemometrics and offer a promising methodology for confirming the possible adulteration of extra VOOs with inexpensive edible oils. Furthermore, this simple and rapid methodology doesn’t need any chemical usage and laborious procedures through the analysis.

Effect of forchlorfenuron and thidiazuron on kiwifruits’ internal qualities, optical properties and their relationship during growth

Forchlorfenuron (1-(2-chloropyridin-4-yl)-3-phenylurea, CPPU) and thidiazuron (N-Phenyl-N′-1,2,3-thiadiazol-5-ylurea, TDZ) are two widely used plant growth regulators in kiwifruit cultivation. They can promote fruit size, but it is unclear whether they have same effect on internal qualities, optical properties and cell structure of kiwifruit, and whether the kiwifruits treated with CPPU and TDZ can be identified based on optical properties. To answer these questions, the kiwifruits treated with 20 mg/L CPPU and 2 mg/L TDZ solutions were used as samples, and the untreated kiwifruits were used as control to investigate the optical properties (absorption coefficient μa and reduced scattering coefficient μs′), internal qualities (soluble solids content (SSC), firmness and moisture content) and microstructure of pulp tissue during the growth. Moreover, the relationship between the optical properties and internal qualities were analyzed, and the potential for identifying the kiwifruits treated with CPPU and TDZ based on optical properties was evaluated. The results showed that CPPU and TDZ increased the SSC and reduced the firmness of kiwifruits, but had some different effects on the moisture content and cell size. CPPU and TDZ did not influence the change trend of μa and μs′ with wavelength, but affected their values and the relationship with internal qualities. In general, the mean μa of the kiwifruits treated with CPPU and with TDZ was the largest and the smallest at the absorption peaks (980 nm, 1190 nm and 1420 nm), respectively. The linear discriminant analysis modeling results showed that the spectra of μa with μs′ had greater potential in identifying the kiwifruits treated with CPPU/TDZ with accuracy of 75.76 %.

Machine Learning Algorithms for Intelligent Decision Recognition and Quantification of Cr(III) in Chromium Speciation.

Cr(III) is a common oxidation state of chromium, and its presence in the environment can occur naturally or as a result of human activities, such as industrial processes, mining, and waste disposal. This article explores the application of machine learning algorithms for the intelligent decision recognition and quantification of Cr(III) in chromium speciation. Three different machine learning models, namely, the Decision Tree (DT) model, the PCA-SVM (Principal Component Analysis-Support Vector Machine) model, and the LDA (Linear Discriminant Analysis) model, were employed and evaluated for accurate and efficient classification of chromium concentrations based on their fluorescence responses. Furthermore, stepwise multiple linear regression analysis was utilized to achieve a more precise quantification of trivalent chromium concentrations through fluorescence visualization. The results demonstrate the potential of machine learning algorithms in accurately detecting and quantifying Cr(III) in chromium speciation with implications for environmental and industrial applications in chromium detection and quantification. The findings from this research pave the way for further exploration and implementation of these models in real-world scenarios, offering valuable insights into various environmental and industrial contexts.

Geographical discrimination of Asian red pepper powders using 1H NMR spectroscopy and deep learning-based convolution neural networks

This study investigated an innovative approach to discriminate the geographical origins of Asian red pepper powders by analyzing one-dimensional 1H NMR spectra through a deep learning-based convolution neural network (CNN). 1H NMR spectra were collected from 300 samples originating from China, Korea, and Vietnam and used as input data. Principal component analysis − linear discriminant analysis and support vector machine models were employed for comparison. Bayesian optimization was used for hyperparameter optimization, and cross-validation was performed to prevent overfitting. As a result, all three models discriminated the origins of the test samples with over 95 % accuracy. Specifically, the CNN models achieved a 100 % accuracy rate. Gradient-weighted class activation mapping analysis verified that the CNN models recognized the origins of the samples based on variations in metabolite distributions. This research demonstrated the potential of deep learning-based classification of 1H NMR spectra as an accurate and reliable approach for determining the geographical origins of various foods.

Predictive techniques for authenticating and quantifying crude palm oil adulterated with leaf extract and food color − An ultra-violet visible spectrophotometric approach

Analytical techniques such as GC-MS and HPLC are benchmarks for determining vegetable oil adulteration but novel methods of oil manipulation have presented new challenges. In addition to the complexities of sorghum leaf and food color adulteration in crude palm oil (CPO), the sophistication, cost and analytical time of these benchmark methods have elicited the search for alternative approaches. Spectroscopy is one of such approaches that have been lauded for non-invasive and rapid analysis but has never been applied to authenticate CPO containing sorghum leaf extract and food color. This study tested the feasibility of using conventional techniques and UV-Vis spectrophotometry and chemometrics to detect CPO containing different concentrations of food colour and sorghum leaf extract. Linear discriminant analysis models were able to classify different concentrations of sorghum leaf dye and food color in crude palm oil with good accuracy. Testing the developed models with samples from 5 different markets revealed a possible adulteration of some market samples after cross-validation. The purity of crude palm oil could be predicted with an R2CV of 0.937, RMSECV of 0.161 g/100 g and RPD of 3.362 after cross-validation using partial least squares regression and testing 18 different mathematical spectral pretreatment methods. The results suggest a potential for rapid detection of concentrations as low as 0.2 g/mL of food color in CPO and color extract from sorghum leaf size 10 cm by 1 cm.

Prediction of preoperative microvascular invasion by dynamic radiomic analysis based on contrast-enhanced computed tomography.

Microvascular invasion (MVI) is a common complication of hepatocellular carcinoma (HCC) surgery, which is an important predictor of reduced surgical prognosis. This study aimed to develop a fully automated diagnostic model to predict pre-surgical MVI based on four-phase dynamic CT images. A total of 140 patients with HCC from two centers were retrospectively included (training set, n = 98; testing set, n = 42). All CT phases were aligned to the portal venous phase, and were then used to train a deep-learning model for liver tumor segmentation. Radiomics features were extracted from the tumor areas of original CT phases and pairwise subtraction images, as well as peritumoral features. Lastly, linear discriminant analysis (LDA) models were trained based on clinical features, radiomics features, and hybrid features, respectively. Models were evaluated by area under curve (AUC), accuracy, sensitivity, specificity, positive and negative predictive values (PPV and NPV). Overall, 86 and 54 patients with MVI- (age, 55.92 ± 9.62 years; 68 men) and MVI+ (age, 53.59 ± 11.47 years; 43 men) were included. Average dice coefficients of liver tumor segmentation were 0.89 and 0.82 in training and testing sets, respectively. The model based on radiomics (AUC = 0.865, 95% CI: 0.725-0.951) showed slightly better performance than that based on clinical features (AUC = 0.841, 95% CI: 0.696-0.936). The classification model based on hybrid features achieved better performance in both training (AUC = 0.955, 95% CI: 0.893-0.987) and testing sets (AUC = 0.913, 95% CI: 0.785-0.978), compared with models based on clinical and radiomics features (p-value < 0.05). Moreover, the hybrid model also provided the best accuracy (0.857), sensitivity (0.875), and NPV (0.917). The classification model based on multimodal intra- and peri-tumoral radiomics features can well predict HCC patients with MVI.

Machine learning approaches for cardiovascular hypertension stage estimation using photoplethysmography and clinical features.

Cardiovascular diseases (CVDs) are a leading cause of death worldwide, with hypertension emerging as a significant risk factor. Early detection and treatment of hypertension can significantly reduce the risk of developing CVDs and related complications. This work proposes a novel approach employing features extracted from the acceleration photoplethysmography (APG) waveform, alongside clinical parameters, to estimate different stages of hypertension. The current study used a publicly available dataset and a novel feature extraction algorithm to extract APG waveform features. Three distinct supervised machine learning algorithms were employed in the classification task, namely: Decision Tree (DT), Linear Discriminant Analysis (LDA), and Linear Support Vector Machine (LSVM). Results indicate that the DT model achieved exceptional training accuracy of 100% during cross-validation and maintained a high accuracy of 96.87% on the test dataset. The LDA model demonstrated competitive performance, yielding 85.02% accuracy during cross-validation and 84.37% on the test dataset. Meanwhile, the LSVM model exhibited robust accuracy, achieving 88.77% during cross-validation and 93.75% on the test dataset. These findings underscore the potential of APG analysis as a valuable tool for clinicians in estimating hypertension stages, supporting the need for early detection and intervention. This investigation not only advances hypertension risk assessment but also advocates for enhanced cardiovascular healthcare outcomes.

Classification of Organic and Conventional Cocoa Beans Using Laser-Induced Fluorescence Spectroscopy Combined with Chemometric Techniques.

The craving for organic cocoa beans has resulted in fraudulent practices such as mislabeling, adulteration, all known as food fraud, prompting the international cocoa market to call for the authenticity of organic cocoa beans before export. In this study, we proposed robust models using laser-induced fluorescence (LIF) and chemometric techniques for rapid classification of cocoa beans as either organic or conventional. The LIF measurements were conducted on cocoa beans harvested from organic and conventional farms. From the results, conventional cocoa beans exhibited a higher fluorescence intensity compared to organic ones. In addition, a general peak wavelength shift was observed when the cocoa beans were excited using a 445nm laser source. These results highlight distinct characteristics that can be used to differentiate between organic and conventional cocoa beans. Identical compounds were found in the fluorescence spectra of both the organic and conventional ones. With preprocessed fluorescence spectra data and utilizing principal component analysis, classification models such as Linear Discriminant Analysis (LDA), Support Vector Machine (SVM), Neural Network (NN) and Random Forest (RF) models were employed. LDA and NN models yielded 100.0% classification accuracy for both training and validation sets, while 99.0% classification accuracy was achieved in the training and validation sets using SVM and RF models. The results demonstrate that employing a combination of LIF and either LDA or NN can be a reliable and efficient technique to classify authentic cocoa beans as either organic or conventional. This technique can play a vital role in maintaining integrity and preventing fraudulent practices in the cocoa bean supply chain.

Identification and classification of chill‐damaged versus sound kiwifruit using Raman spectroscopy and chemometrics

AbstractThe early detection of fruit disorders is crucial to maintaining a consistent, high‐quality kiwifruit product. Chilling injury is a physiological disorder found in kiwifruit that can be challenging to identify until it reaches a severe stage or the fruit is cut and opened. Considering this, Raman spectroscopy combined with chemometrics was investigated for sound and chill‐damaged ‘Zesy002’ kiwifruit. We carried out spectral analysis on fruit harvested in 2018 and 2019. Damaged and sound fruit samples were separated based on spectral signatures from phenyl propanoids and sugars. Furthermore, the 2018 fruit sample set was used to construct, validate, and test models using support vector machine, principal component analysis–linear discriminant analysis, and partial least squares–discriminant analysis. Additionally, the robustness of the model was assessed using the 2019 fruit sample set considering test set accuracy, sensitivity, and specificity. Support vector machine models were developed and resulted in a 93% accuracy, 85% sensitivity, and 100% specificity to differentiate the test set fruit (2018 season). Principal component analysis–linear discriminant analysis models and partial least squares–discriminant analysis model built with the same data set gave &gt;83% and 93% test accuracy, respectively. Models showed robustness with samples from the 2019 season. This study provides insights into the potential of using Raman spectroscopy for identifying chilling injury in kiwifruit.

Elucidating the mechanism of action of mycotoxins through machine learning-driven QSAR models: Focus on lipid peroxidation

Understanding the mechanisms of mycotoxin toxicity is crucial for establishing effective guidelines and preventive strategies. In this study, machine learning models based on quantitative structure-activity relationship (QSAR) were employed to predict the lipid peroxidation activity of mycotoxins. Two different algorithms using Linear Discriminant Analysis (LDA) and Artificial Neural Networks (ANNs) have been trained using a dataset of 70 mycotoxins. The LDA model had an average correct classification rate of 91%, while the ANN model achieved a perfect 100% classification rate. Following an internal validation process, the models were utilized to predict mycotoxins with known lipid peroxidation activity. The machine learning models achieved an 88% correct classification rate for these mycotoxins. Finally, by utilizing classified algorithms, the study aimed to infer the mechanism of action related to lipid peroxidation for 91 unstudied mycotoxins. These models provide a fast, accurate, and cost-effective means to assess the potential toxicity and mechanism of action of mycotoxins. The findings of this study contribute to a comprehensive understanding of mycotoxin toxicology and assist researchers and toxicologists in evaluating health risks associated with mycotoxin exposure and developing appropriate preventive strategies and potential therapeutic interventions to mitigate the effects of mycotoxins.

Applications of Different Classification Machine Learning Techniques to Predict Formation Tops and Lithology While Drilling.

Accurate prediction of formation tops and lithology plays a critical role in optimizing drilling processes, cost reduction, and risk mitigation in hydrocarbon operations. Although several techniques like well logging, core sampling, cuttings analysis, seismic surveys, and mud logging are available for identifying formation tops, they have limitations such as high costs, lower accuracy, manpower-intensive processes, and time or depth lags that impede real-time estimation. Consequently, this study aims to leverage machine learning models based on easily accessible drilling parameters to predict formation tops and lithologies, overcoming the limitations associated with traditional methods. Data from two wells (A and B) in the Middle East, encompassing drilling mechanical parameters such as rate of penetration (ROP), drill string rotation (DSR), pumping rate (Q), standpipe pressure (SPP), weight on bit (WOB), and torque, were collected for real-field analysis. Machine learning models including Gaussian naive Bayes (GNB), logistic regression (LR), and linear discriminant analysis (LDA) were trained and tested on the data set from well A, while the data set from well B was utilized for model validation as unseen data. The formations of wells A and B consist of four lithologies, namely, sandstone, anhydrite, carbonate/shale, and carbonates, necessitating the development of multiclass classification models. The drilling parameters, specifically the WOB and ROP, exhibited a strong influence on lithology identification. Among the models, GNB demonstrated exceptional performance in predicting formation lithology from the drilling parameters, achieving accuracy and nearly perfect precision, recall, and F1 score for the different classes. LDA and LR models accurately predicted sandstone and carbonate lithologies, although some misclassifications occurred in approximately 5% of points for anhydrite and around 20% in carbonate/shale formations. During validation, the models demonstrated accuracies of around 0.96, 0.95, and 0.92 for the GNB, LR, and LDA, respectively. The study highlights the efficacy of the developed machine learning models in accurately predicting the formation lithology and tops in real time. This is achieved by utilizing readily available drilling parameters, making the approach highly accurate and cost effective by leveraging existing real-time drilling data.

Composition and Bioactivity of Chlorogenic Acids in Vegetable and Conventional Sweet Potato Vine Tips.

Sweet potato vine tips are abundant in chlorogenic acid (CGA). In this study, CGA was extracted from vegetable and conventional sweet potato vine tips using ethanol, followed by subsequent purification of the extract through a series of sequential steps. Over 4 g of the purified product was obtained from 100 g of sweet potato vine tip powder, producing more than 85% of purified CGA. The LC-MS analysis of all samples indicated that 4-CQA was the predominant isomer in both sweet potato cultivars. Significant variations of p-coumaroyl quinic acids, feruloyl quinic acids, dicaffeoyl quinic acids, and tricaffeoyl quinic acid were identified, whereas the mono-caffeoyl quinic acids did not vary when the two sweet potato varieties were compared. Compared to conventional sweet potatoes, vegetable sweet potatoes exhibit a high negative correlation between 4-CQA and 5-pCoQA, while showing a high positive correlation between 3,5-CQA and 3-pCoQA. A series of principal component analyses (PCA) using CGA isomers enables a clear differentiation between vine tips derived from vegetable and conventional sweet potatoes. The model of linear discriminant analysis, based on the characteristic CGA, achieved a 100% accuracy rate in distinguishing between vegetable and conventional sweet potatoes. The high purity of sweet potato CGA (SCGA) exhibited potent anti-breast cancer activity. The results demonstrated that SCGA significantly suppressed the clonogenicity of MB231 and MCF7 cells, and impeded the migratory, invasive, and lung metastatic potential of MB231 cells.

Recovery Gestures Classification Using KNN and LDA Models.

Hand and joint mobility recovery involve performing a set of exercises. Gestures are often used in the hand mobility recovery process. This paper discusses the selection and the use of 2 models of neural networks for the classification of data that describe Leap Motion gestures. The gestures are: the hand opening and closing gesture and the palm rotation gesture. The purpose is the optimal selection of the neural network model to be used in the classification of the data describing the recovery gestures. The models chosen for the classification of the two gestures were: Linear Discriminant Analysis (LDA) and K-neighbors Classifier (KNN). The accuracies achieved in the classification of the gestures for each model are: 0.91 - LDA and 0.98 - KNN.

Abnormal spontaneous activity of regions related to mood regulation mediates the effect of childhood emotional neglect on major depressive disorder

This study investigated the mediating factors between childhood emotional neglect (EN) and major depressive disorder (MDD) and whether combining multi-indicator could help diagnose MDD. Regional homogeneity (ReHo) and clinical features were compared between 33 MDD patients and 36 healthy controls (HC). Mediation analysis was employed to explore whether social support or ReHo mediates the association between EN and MDD. The linear discriminant analysis model was constructed with EN, social support, and ReHo, and applied to distinguish MDD from HC in both primary and replication cohorts. We found that MDD patients experienced severer EN and poorer social support, and exhibited lower ReHo in the left middle occipital gyrus and bilateral postcentral gyrus, and higher ReHo in the right cerebellum crus1. EN could affect MDD directly and indirectly through ReHo in these discrepant brain regions and social support. Combining ReHo values of these four distinct brain regions, EN, and objective support could classify MDD patients from HC, and the 10-fold cross-validation accuracy within-study replication and in the independent cohort was 83.78 % ± 1.49 % and 82.72 % ± 2.22 %, respectively. These findings suggested that childhood EN, social support, and emotional-related regions’ ReHo were associated with risks of MDD, providing new insights into the pathological mechanisms underlying MDD.

Metabolomics Meets Clinics: A Multivariate Analysis of Plasma and Urine Metabolic Signatures in Pulmonary Arterial Hypertension.

Pulmonary arterial hypertension (PAH) is a severe, multifactorial, and frequently misdiagnosed disorder. The aim of this observational study was to compare the plasma and urine metabolomic profiles of PAH patients and healthy control subjects. Plasma and urine metabolomic profiles were analyzed using the GC-MS technique. Correlations between metabolite levels and clinical parameters among PAH patients, as well as the between-group differences, were evaluated. The linear discriminant analysis model, which allows for subject classification in terms of PAH with the highest possible precision, was developed and proposed. Plasma pyruvic acid, cholesterol, threonine, urine 3-(3-hydroxyphenyl)-3-hydroxypropanoic acid, butyric acid, 1,2-benzenediol, glucoheptonic acid, and 2-oxo-glutaric acid were found to build a relatively accurate classification model for PAH patients. The model reached an accuracy of 91% and significantly improved subject classification (OR = 119 [95% CI: 20.3-698.3], p < 0.0001). Five metabolites were detected in urine that provide easily available and noninvasive tests as compared to right heart catheterization. The selected panel of metabolites has potential for early recognition of patients with dyspnea and faster referral to a reference center.

Utilização de espectroscopia de infravermelho médio e modelos quimiométricos para discriminação de variedades de cacau seco

ABSTRACT Generally, the taxonomic classification of cocoa beans is based on the theobromine/caffeine ratio determined using high-performance liquid chromatography (HPLC). However, this technique involves laborious and time-consuming calculations. Attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectroscopy is a valuable, effective, and rapid tool for analyzing the chemical composition of food products. The objective of this study was to examine the potential of ATR-FTIR combined with chemometric tools such as principal component analysis (PCA), linear discriminant analysis (LDA), and partial least squares regression-discriminant analysis (PLS-DA) to discriminate between the Trinitario and Forastero dry bean cocoa varieties defined by theobromine and caffeine measurements via HPLC. The cocoa varieties were evaluated using HPLC analysis of 36 dry cocoa bean samples to determine the theobromine/caffeine ratio. Moreover, ATR-FTIR spectra were analyzed in the mid-infrared (MIR) region, and signals associated with theobromine and caffeine were identified and analyzed using the LDA and PLS-DA models. The LDA and PLS-DA models allowed the satisfactory differentiation between cocoa varieties, providing overall prediction capacity values of 98.2 ± 1.8% and 96.1 ± 2.4%, respectively. The results show the potential of ATR-FTIR spectroscopy for the reliable, fast, and easy differentiation of dried cocoa beans.

Pattern Analysis of Daily Lesion Volume Trajectories for Early Prediction of Glioblastoma Progression During MR-Linac Radiotherapy

Distinguishing between true progression (TP) and pseudoprogression (PP) post-radiotherapy (RT) is of paramount importance for treatment management of patients with glioblastoma (GBM). MR-Linac systems allow for daily monitoring of tumor changes throughout the course of RT. We hypothesized that the patterns of tumor volume change during RT may enable early prediction of treatment response. Using an IRB-approved prospective cohort of GBM patients undergoing 30 fractions of chemoRT to 60 Gy on a 0.35T MR-Linac, tumor/edema (tumor lesion) regions of interest (ROI) were contoured on daily T2-weighted treatment set-up scans. The obtained tumor lesion (TL) volume changes during treatment were smoothed with a moving average Gaussian window over time. Non-negative Matrix Factorization (NMF) was applied to the data matrix D (N x F), containing the trajectories in its rows for each patient, where N is the number of patients analyzed and F is the number of fractions. NMF represents D as a linear combination of three temporal (hidden) patterns and their weights in each individual trajectory. The same analysis was performed for ΔD, containing the changes in volumes with reference to the first fraction. The calculated weights were scaled in [0, 1], expressed as probabilities (by ℓ1-normalization) and used as features in Linear Discriminant Analysis (LDA). The LDA model was trained to differentiate between no progression (NP), PP and TP, and assessed by leave-one-subject-out cross-validation. Thirty-six patients were screened for inclusion: 9 were excluded due to no T2 lesion (resection cavity only). Of the remaining 27 GBM patients analyzed, 10 had no tumor growth on first post-RT diagnostic MRI, 6 were determined to have PP based on regression or long-term stability of findings, and 11 had TP due to continued progression of disease past 6 months, rapid patient death from disease, or tissue sampling showing active malignancy. With the use of only 2 features, LDA achieved an overall accuracy of 70.4% classifying correctly: 6 (60%), 4 (67%), and 9 (82%) patients with NP, PP, and TP, respectively. The temporal NMF patterns (monotonous decrease, rapid increase during the third part of the treatment, etc.) indicate that there is enough signal to classify patients' response based on the pattern tumor volume changes during RT. We identified tumor dynamics' patterns during RT, indicative of differential behavior of tumor growth between TP and PP. Although with a limited number of patients, these initial results suggest that tumor volume changes during treatment may provide early markers of treatment response. This could allow physicians to adapt/intensify treatment in real time for poorly responding patients. Next steps include automating the process of real-time tumor volume monitoring by incorporating a deep learning solution for automatic volume delineation on daily treatment set-up scans.