Search

Abstract In this study, we implemented a combination of data augmentation and artificial intelligence (AI) model—Convolutional Neural Network (CNN)—to help physicians classify colonic polyps into traditional adenoma (TA), sessile serrated adenoma (SSA), and hyperplastic polyp (HP). We collected ordinary endoscopy images under both white and NBI lights. Under white light, we collected 257 images of HP, 423 images of SSA, and 60 images of TA. Under NBI light, were collected 238 images of HP, 284 images of SSA, and 71 images of TA. We implemented the CNN-based artificial intelligence model, Inception V4, to build a classification model for the types of colon polyps. Our final AI classification model with data augmentation process is constructed only with white light images. Our classification prediction accuracy of colon polyp type is 94%, and the discriminability of the model (area under the curve) was 98%. Thus, we can conclude that our model can help physicians distinguish between TA, SSA, and HPs and correctly identify precancerous lesions such as TA and SSA.

Sweet oranges have long been an integral part of global health and culinary practices, offering a wealth of nutrients and bioactive compounds. Ensuring the authenticity of these citrus fruits is essential for maintaining consumer confidence, promoting transparency in sourcing, and protecting producers' reputations in the marketplace. In this study, we explored the feasibility of using multi-element profiling combined with pattern recognition algorithms to trace the origin of sweet orange samples. To achieve this, we employed an optimized microwave plasma atomic emission spectroscopy (MP-AES) method to analyze the elemental composition (Al, Ba, Ca, Cd, Co, Cr, Cu, Fe, K, Li, Mg, Mn, Mo, Na, Ni, Pb, Sr, and Zn) of 183 orange samples from four production regions in northeastern Argentina. Support vector machine (SVM), random forest (RF), and gradient boosting tree (GBT) models were then built using the collected data to identify elemental tracer's indicative of origin. Based on a comprehensive evaluation of overall accuracy, receiver operating characteristic (ROC) curves, and area under the curve (AUC), the GBT model demonstrated the best classification performance, achieving a 96.5 % correct prediction rate on test samples, as confirmed by the ROC curve (AUC = 0.973). Consequently, this approach provides compelling evidence for the potential utility of MP-AES combined with supervised modeling to determine the geographic origin of sweet oranges produced in Argentina, thereby contributing to consumer protection against fraud.

In this study, we implemented a combination of data augmentation and artificial intelligence (AI) model—Convolutional Neural Network (CNN)—to help physicians classify colonic polyps into traditional adenoma (TA), sessile serrated adenoma (SSA), and hyperplastic polyp (HP). We collected ordinary endoscopy images under both white and NBI lights. Under white light, we collected 257 images of HP, 423 images of SSA, and 60 images of TA. Under NBI light, were collected 238 images of HP, 284 images of SSA, and 71 images of TA. We implemented the CNN-based artificial intelligence model, Inception V4, to build a classification model for the types of colon polyps. Our final AI classification model with data augmentation process is constructed only with white light images. Our classification prediction accuracy of colon polyp type is 94%, and the discriminability of the model (area under the curve) was 98%. Thus, we can conclude that our model can help physicians distinguish between TA, SSA, and HPs and correctly identify precancerous lesions such as TA and SSA. [ABSTRACT FROM AUTHOR]

Background: Epilepsy stands as an intricate disorder of the central nervous system, subject to the influence of diverse risk factors and a significant genetic predisposition. Within the pathogenesis of temporal lobe epilepsy (TLE), the apoptosis of neurons and glial cells in the brain assumes pivotal importance. The identification of differentially expressed apoptosis-related genes (DEARGs) emerges as a critical imperative, providing essential guidance for informed treatment decisions. Methods: We obtained datasets related to epilepsy, specifically GSE168375 and GSE186334. Utilizing differential expression analysis, we identified a set of 249 genes exhibiting significant variations. Subsequently, through an intersection with apoptosis-related genes, we pinpointed 16 genes designated as differentially expressed apoptosis-related genes (DEARGs). These DEARGs underwent a comprehensive array of analyses, including enrichment analyses, biomarker selection, disease classification modeling, immune infiltration analysis, prediction of miRNA and transcription factors, and molecular docking analysis. Results: In the epilepsy datasets examined, we successfully identified 16 differentially expressed apoptosis-related genes (DEARGs). Subsequent validation in the external dataset GSE140393 revealed the diagnostic potential of five biomarkers (CD38, FAIM2, IL1B, PAWR, S100A8) with remarkable accuracy, exhibiting an impressive area under curve (AUC) (The overall AUC of the model constructed by the five key genes was 0.916, and the validation set was 0.722). Furthermore, a statistically significant variance (p < 0.05) was observed in T cell CD4 naive and eosinophil cells across different diagnostic groups. Exploring interaction networks uncovered intricate connections, including gene-miRNA interactions (164 interactions involving 148 miRNAs), gene-transcription factor (TF) interactions (22 interactions with 20 TFs), and gene-drug small molecule interactions (15 interactions involving 15 drugs). Notably, IL1B and S100A8 demonstrated interactions with specific drugs. Conclusion: In the realm of TLE, we have successfully pinpointed noteworthy differentially expressed apoptosis-related genes (DEARGs), including CD38, FAIM2, IL1B, PAWR, and S100A8. A comprehensive understanding of the implications associated with these identified genes not only opens avenues for advancing our comprehension of the underlying pathophysiology but also bears considerable potential in guiding the development of innovative diagnostic methodologies and therapeutic interventions for the effective management of epilepsy in the future. [ABSTRACT FROM AUTHOR]

This paper proposes experimental results and a modeling procedure to analyze the factors affecting the defects of centrifugal pumps. The centrifugal pump is an essential mechanical element. Pump malfunction can cause problems in the entire system because most production facilities are equipped with various centrifugal pumps. Therefore, the diagnosis of pump failure is essential, and several studies have proposed data-driven methodologies for detecting faults in pump systems. Most studies have performed fault classification using only the vibration signal of the pump. However, this study approached the problem in three ways to improve FDC performance. Firstly, this study tried to collect various signals such as pressure, flow, and motor current as well as vibration signals. That is because various signals are likely to indicate the performance of the pump system considering the mechanical characteristics of the centrifugal pump. Secondly, this study presented a methodology for extracting features that can be used in machine learning models from the signals. For objective analysis, not only various statistics were calculated, but also features in the frequency domain were extracted. Using the extracted features, random forest modeling was applied and higher performance was achieved in experiments. Lastly, variables that significantly affected the anomaly detection of the pump were suggested. In order to analyze experimental results of feature importance, we have presented a procedure to introduce a correlation matrix of cosine similarity, which is a new method that has not been attempted in previous research. [ABSTRACT FROM AUTHOR]

The prolonged duration of chronic low back pain (cLBP) inevitably leads to changes in the cognitive, attentional, sensory and emotional processing brain regions. Currently, it remains unclear how these alterations are manifested in the interplay between brain functional and structural networks. This study aimed to predict the Oswestry Disability Index (ODI) in cLBP patients using multimodal brain magnetic resonance imaging (MRI) data and identified the most significant features within the multimodal networks to aid in distinguishing patients from healthy controls (HCs). We constructed dynamic functional connectivity (dFC) and structural connectivity (SC) networks for all participants (n = 112) and employed the Connectome-based Predictive Modeling (CPM) approach to predict ODI scores, utilizing various feature selection thresholds to identify the most significant network change features in dFC and SC outcomes. Subsequently, we utilized these significant features for optimal classifier selection and the integration of multimodal features. The results revealed enhanced connectivity among the frontoparietal network (FPN), somatomotor network (SMN) and thalamus in cLBP patients compared to HCs. The thalamus transmits pain-related sensations and emotions to the cortical areas through the dorsolateral prefrontal cortex (dlPFC) and primary somatosensory cortex (SI), leading to alterations in whole-brain network functionality and structure. Regarding the model selection for the classifier, we found that Support Vector Machine (SVM) best fit these significant network features. The combined model based on dFC and SC features significantly improved classification performance between cLBP patients and HCs (AUC=0.9772). Finally, the results from an external validation set support our hypotheses and provide insights into the potential applicability of the model in real-world scenarios. Our discovery of enhanced connectivity between the thalamus and both the dlPFC (FPN) and SI (SMN) provides a valuable supplement to prior research on cLBP.

Depressive mood states in healthy populations are prevalent but often under-reported. Biases exist in self-reporting of depression in otherwise healthy individuals. Gait and balance control can serve as objective markers for identifying those individuals, particularly in real-world settings. We utilized inertial measurement units (IMU) to measure gait and balance control. An exploratory, cross-sectional design was used to compare individuals who reported feeling depressed at the moment (n = 49) with those who did not (n = 84). The Quality Assessment Tool for Observational Cohort and Cross-sectional Studies was employed to ensure internal validity. We recruited 133 participants aged between 18–36 years from the university community. Various instruments were used to evaluate participants' present depressive symptoms, sleep, gait, and balance. Gait and balance variables were used to detect depression, and participants were categorized into three groups: not depressed, mild depression, and moderate–high depression. Participant characteristics were analyzed using ANOVA and Kruskal–Wallis tests, and no significant differences were found in age, height, weight, BMI, and prior night's sleep between the three groups. Classification models were utilized for depression detection. The most accurate model incorporated both gait and balance variables, yielding an accuracy rate of 84.91% for identifying individuals with moderate–high depression compared to non-depressed individuals. [ABSTRACT FROM AUTHOR]

Remote sensing of nutrient disorders has become more common in recent years. Most research has considered one or two nutrient disorders and few studies have sought to distinguish among multiple macronutrient deficiencies. This study was conducted to provide a baseline spectral characterization of macronutrient deficiencies in flue-cured tobacco (Nicotiana tabacum L.). Reflectance measurements were obtained from greenhouse-grown nutrient-deficient plants at several stages of development. Feature selection methods including information entropy and first and second derivatives were used to identify wavelengths useful for discriminating among these deficiencies. Detected variability was primarily within wavelengths in the visible spectrum, while near-infrared and shortwave-infrared radiation contributed little to the observed variability. Principal component analysis was used to reduce data dimensionality and the selected components were used to develop linear discriminant analysis models to classify the symptoms. Classification models for young, intermediate, and mature plants had overall accuracies of 92%, 82%, and 75%, respectively, when using 10 principal components. Nitrogen, sulfur, and magnesium deficiencies exhibited greater classification accuracies, while phosphorus and potassium deficiencies demonstrated poor or inconsistent results. This study demonstrates that spectral analysis of flue-cured tobacco is a promising methodology to improve current scouting methods. [ABSTRACT FROM AUTHOR]

There is an increasing demand to introduce Introductory Business Analytics (IBA) courses into undergraduate business education. Many real‐world business contexts require predictive analytics to understand the determinants of a dichotomous outcome; hence, IBA courses should include binary logistic regression analysis. This article provides our reflective discussions on the design of learning activities and assessments to assist business students in learning binary logistic regression in an IBA course. Data on student engagement and learning outcomes are used to shed light on the impacts of teaching logistic regression on student learning and experience. Notably, students opt to focus their assessment work more on logistic regression than on multiple regression analysis, showing the potential attraction of students toward binary logistic regression analysis. We also observed several challenges, mainly related to the use of Excel, that require special attention from instructors. [ABSTRACT FROM AUTHOR]

Depressive mood states in healthy populations are prevalent but often under-reported. Biases exist in self-reporting of depression in otherwise healthy individuals. Gait and balance control can serve as objective markers for identifying those individuals, particularly in real-world settings. We utilized inertial measurement units (IMU) to measure gait and balance control. An exploratory, cross-sectional design was used to compare individuals who reported feeling depressed at the moment (n = 49) with those who did not (n = 84). The Quality Assessment Tool for Observational Cohort and Cross-sectional Studies was employed to ensure internal validity. We recruited 133 participants aged between 18–36 years from the university community. Various instruments were used to evaluate participants’ present depressive symptoms, sleep, gait, and balance. Gait and balance variables were used to detect depression, and participants were categorized into three groups: not depressed, mild depression, and moderate–high depression. Participant characteristics were analyzed using ANOVA and Kruskal–Wallis tests, and no significant differences were found in age, height, weight, BMI, and prior night’s sleep between the three groups. Classification models were utilized for depression detection. The most accurate model incorporated both gait and balance variables, yielding an accuracy rate of 84.91% for identifying individuals with moderate–high depression compared to non-depressed individuals.

The squeezing behavior of surrounding rock can be described as the time-dependent large deformation during tunnel excavation, which appears in special geological conditions, such as weak rock masses and high in situ stress. Several problems such as budget increase and construction period extension can be caused by squeezing in rock mass. It is significant to propose a model for accurate prediction of rock squeezing. In this research, the support vector machine (SVM) as a machine learning model was optimized by the whale optimization algorithm (WOA), WOA-SVM, to classify the tunnel squeezing based on 114 real cases. The role of WOA in this system is to optimize the hyper-parameters of SVM model for receiving a higher level of accuracy. In the established database, five input parameters, i.e., buried depth, support stiffness, rock tunneling quality index, diameter and the percentage strain, were used. In the process of model classification, different effective parameters of SVM and WOA were considered, and the optimum parameters were designed. To examine the accuracy of the WOA-SVM, the base SVM, ANN (refers to the multilayer perceptron) and GP (refers to the Gaussian process classification) were also constructed. Evaluation of these models showed that the optimized WOA-SVM is the best model among all proposed models in classifying the tunnel squeezing. It has the highest accuracy (approximately 0.9565) than other un-optimized individual classifiers (SVM, ANN, and GP). This was obtained based on results of different performance indexes. In addition, according to sensitivity analysis, the percentage strain is highly sensitive to the model, followed by buried depth and support stiffness. That means, ɛ, H and K are the best combination of parameters for the WOA–SVM model. [ABSTRACT FROM AUTHOR]

This paper analyses method skew in India over more than two decades using a new method skew index. The analysis reveals clear regional pattern in method skew in the country. In north and east regions of the country, method skew is average and below average, but it is high or very high in the south region. We investigate roles of such factors as poverty, education, social class, and religion in deciding the method skew at the district level using the classification modeling approach and find that the roles of these factors are different in different regions of the country. An important finding of the analysis is that there is a positive correlation between the degree of the method skew and contraceptive prevalence at the district level. [ABSTRACT FROM AUTHOR]

Remote sensing of nutrient disorders has become more common in recent years. Most research has considered one or two nutrient disorders and few studies have sought to distinguish among multiple macronutrient deficiencies. This study was conducted to provide a baseline spectral characterization of macronutrient deficiencies in flue-cured tobacco (Nicotiana tabacum L.). Reflectance measurements were obtained from greenhouse-grown nutrient-deficient plants at several stages of development. Feature selection methods including information entropy and first and second derivatives were used to identify wavelengths useful for discriminating among these deficiencies. Detected variability was primarily within wavelengths in the visible spectrum, while near-infrared and shortwave-infrared radiation contributed little to the observed variability. Principal component analysis was used to reduce data dimensionality and the selected components were used to develop linear discriminant analysis models to classify the symptoms. Classification models for young, intermediate, and mature plants had overall accuracies of 92%, 82%, and 75%, respectively, when using 10 principal components. Nitrogen, sulfur, and magnesium deficiencies exhibited greater classification accuracies, while phosphorus and potassium deficiencies demonstrated poor or inconsistent results. This study demonstrates that spectral analysis of flue-cured tobacco is a promising methodology to improve current scouting methods.

Anaerobic biodegradation rates (half-lives) of organic chemicals are pivotal for environmental risk assessment and remediation. Traditional experimental evaluation, constrained by prolonged, oxygen-free conditions, struggles to keep pace with emerging contaminants. Data-driven machine learning (ML) models serve as promising complements. However, reported quantitative structure-biodegradation relationships or ML models on anaerobic biodegradation are mostly based on small data sets (<100 records) and neglect experimental conditions, usually achieving compromised predictions. This work aimed to develop ML models for predicting the biodegradation half-lives of organic pollutants in anaerobic environments (i.e., sediment/soil and sludge). Focusing on important features of both chemicals and experimental conditions, we first curated two data sets, one for sediment/soil (SED) and the other for sludge (SLD), covering 978 records for 206 chemicals from the literature, and then conducted a meta-analysis. Next, we built a binary classification (half-life of 30 days as the cutoff) model with an accuracy of 81% and a regression model with R 2 of 0.56 for SED based on LightGBM (80% and 0.31 for SLD based on Extra tree, respectively). The model interpretations underscored the significance of experimental conditions (e.g., temperature and inoculum dosage), as evidenced by their high feature importance, and the models were found to correctly capture the effects of chemical substructures, for example, branched structures and aromatic rings prolonged half-lives while methyl group and ortho-substitution on rings shortened half-lives. The applicability domains of the models were also defined, resulting in reasonable prediction for the half-lives of 41% (SED) or 67% (SLD) of over 4000 persistent, bioaccumulative, and toxic chemicals. Overall, this study pioneers ML models for predicting the anaerobic degradation half-lives, offering valuable support for future evaluation and implementation of chemical anaerobic biodegradation.

• The alterations in dFC and SC networks unveil distinct characteristics associated with cLBP. • The enhanced connectivity between the FPN, SMN and thalamus signifies the most relevant changes in brain networks associated with cLBP. • The integration of different features effectively enhances the classification accuracy between cLBP patients and HC. The prolonged duration of chronic low back pain (cLBP) inevitably leads to changes in the cognitive, attentional, sensory and emotional processing brain regions. Currently, it remains unclear how these alterations are manifested in the interplay between brain functional and structural networks. This study aimed to predict the Oswestry Disability Index (ODI) in cLBP patients using multimodal brain magnetic resonance imaging (MRI) data and identified the most significant features within the multimodal networks to aid in distinguishing patients from healthy controls (HCs). We constructed dynamic functional connectivity (dFC) and structural connectivity (SC) networks for all participants (n = 112) and employed the Connectome-based Predictive Modeling (CPM) approach to predict ODI scores, utilizing various feature selection thresholds to identify the most significant network change features in dFC and SC outcomes. Subsequently, we utilized these significant features for optimal classifier selection and the integration of multimodal features. The results revealed enhanced connectivity among the frontoparietal network (FPN), somatomotor network (SMN) and thalamus in cLBP patients compared to HCs. The thalamus transmits pain-related sensations and emotions to the cortical areas through the dorsolateral prefrontal cortex (dlPFC) and primary somatosensory cortex (SI), leading to alterations in whole-brain network functionality and structure. Regarding the model selection for the classifier, we found that Support Vector Machine (SVM) best fit these significant network features. The combined model based on dFC and SC features significantly improved classification performance between cLBP patients and HCs (AUC=0.9772). Finally, the results from an external validation set support our hypotheses and provide insights into the potential applicability of the model in real-world scenarios. Our discovery of enhanced connectivity between the thalamus and both the dlPFC (FPN) and SI (SMN) provides a valuable supplement to prior research on cLBP. [ABSTRACT FROM AUTHOR]

Purpose: the formation of mathematical models and algorithms for predicting the development of depressive disorders in patients based on their individual medical and social characteristics. Material and methods. The substantiation of the urgency of the problem is given. The methods and the course of the research are described. Results and conclusions. On the basis of the data obtained, a set of the most significant medical and social characteristics of patients was formed, an integral indicator for assessing the degree of risk of depressive disorders was developed, a computer program was created to predict the probability of development of affective pathology. Appropriate conclusions had been made.

Background: Epilepsy stands as an intricate disorder of the central nervous system, subject to the influence of diverse risk factors and a significant genetic predisposition. Within the pathogenesis of temporal lobe epilepsy (TLE), the apoptosis of neurons and glial cells in the brain assumes pivotal importance. The identification of differentially expressed apoptosis-related genes (DEARGs) emerges as a critical imperative, providing essential guidance for informed treatment decisions., Methods: We obtained datasets related to epilepsy, specifically GSE168375 and GSE186334. Utilizing differential expression analysis, we identified a set of 249 genes exhibiting significant variations. Subsequently, through an intersection with apoptosis-related genes, we pinpointed 16 genes designated as differentially expressed apoptosis-related genes (DEARGs). These DEARGs underwent a comprehensive array of analyses, including enrichment analyses, biomarker selection, disease classification modeling, immune infiltration analysis, prediction of miRNA and transcription factors, and molecular docking analysis., Results: In the epilepsy datasets examined, we successfully identified 16 differentially expressed apoptosis-related genes (DEARGs). Subsequent validation in the external dataset GSE140393 revealed the diagnostic potential of five biomarkers (CD38, FAIM2, IL1B, PAWR, S100A8) with remarkable accuracy, exhibiting an impressive area under curve (AUC) (The overall AUC of the model constructed by the five key genes was 0.916, and the validation set was 0.722). Furthermore, a statistically significant variance ( p < 0.05) was observed in T cell CD4 naive and eosinophil cells across different diagnostic groups. Exploring interaction networks uncovered intricate connections, including gene-miRNA interactions (164 interactions involving 148 miRNAs), gene-transcription factor (TF) interactions (22 interactions with 20 TFs), and gene-drug small molecule interactions (15 interactions involving 15 drugs). Notably, IL1B and S100A8 demonstrated interactions with specific drugs., Conclusion: In the realm of TLE, we have successfully pinpointed noteworthy differentially expressed apoptosis-related genes (DEARGs), including CD38, FAIM2, IL1B, PAWR, and S100A8. A comprehensive understanding of the implications associated with these identified genes not only opens avenues for advancing our comprehension of the underlying pathophysiology but also bears considerable potential in guiding the development of innovative diagnostic methodologies and therapeutic interventions for the effective management of epilepsy in the future., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2024 Wang, Ren, Xu, Zhang, Wang, Wang, Zhong, Wang and Yao.)

Highlights • Daily pattern prediction-based modeling approach for electricity price forecasting. • Weighted voting mechanism (WVM) to better generate the final day-ahead DPP. • The proposed approach is validated in the case study using the data from PJM. Abstract Day-ahead electricity price forecasting (DAEPF) plays a very important role in the decision-making optimization of electricity market participants, the dispatch control of independent system operators (ISOs) and the strategy formulation of energy trading. Unified modeling that only fits a single mapping relation between the historical data and future data usually produces larger errors because the different fluctuation patterns in electricity price data show different mapping relations. A daily pattern prediction (DPP) based classification modeling approach for DAEPF is proposed to solve this problem. The basic idea is that first recognize the price pattern of the next day from the "rough" day-ahead forecasting results provided by conventional forecasting methods and then perform classification modeling to further improve the forecasting accuracy through building a specific forecasting model for each pattern. The proposed approach consists of four steps. First, K-means is utilized to group all the historical daily electricity price curves into several clusters in order to assign each daily curve a pattern label for the training of the following daily pattern recognition (DPR) model and classification modeling. Second, a DPP model is proposed to recognize the price pattern of the next day from the forecasting results provided by multiple conventional forecasting methods. A weighted voting mechanism (WVM) method is proposed in this step to combine multiple day-ahead pattern predictions to obtain a more accurate DPP result. Third, the classification forecasting model of each different daily pattern can be established according to the clustering results in step 1. Fourth, the credibility of DPP result is checked to eventually determine whether the proposed classification DAEPF modeling approach can be adopted or not. A case study using the real electricity price data from the PJM market indicates that the proposed approach presents a better performance than unified modeling for a certain daily pattern whose DPP results show high reliability and accuracy. [ABSTRACT FROM AUTHOR]

Fermented soybean paste is widely used in food sauces and stews in East Asia. The main ingredients are soybeans, salt, and water. Therefore, it provides both proteins and inorganic nutrients. Fermented soybean paste consumed in South Korea is produced in South Korea and China. In this study, for 166 fermented soybean pastes collected in Korean markets, nine inorganic elements were analyzed by inductively-coupled plasma optical emission spectroscopy (ICP-OES), and partial least squares-discriminant analysis (PLS-DA) models were developed to classify the samples with respect to their provenances. Among these elements, Mg and Sr were effective variables for discriminating between fermented soybean pastes produced in South Korea and those produced in China. The model trained using Mg and Sr exhibited a classification accuracy of 92.7%. Discrimination power of Mg was rationalized by the difference in Mg concentration between the salts used in South Korea (sea salts) and those used in China (refined salts). Our results indicated that the elemental content of fermented soybean paste was significantly influenced by the salts used as the main ingredients. The combination of ICP-OES and PLS-DA has been suggested as a reliable methodology for discriminating fermented soybean pastes produced in South Korea and those produced in China. • Fermented soybean pastes were classified according to their producing countries. • Discrimination capabilities of inorganic elements were investigated. • Mg, Sr, and Zn have significant discrimination capabilities. • Salt used for the soybean pastes is the source of Mg with discrimination power. [ABSTRACT FROM AUTHOR]

In the last decade essential oils have attracted scientists with a constant increase rate of more than 7% as witnessed by almost 5000 articles. Among the prominent studies essential oils are investigated as antibacterial agents alone or in combination with known drugs. Minor studies involved essential oil inspection as potential anticancer and antiviral natural remedies. In line with the authors previous reports the investigation of an in-house library of extracted essential oils as a potential blocker of HSV-1 infection is reported herein. A subset of essential oils was experimentally tested in an in vitro model of HSV-1 infection and the determined IC50s and CC50s values were used in conjunction with the results obtained by gas-chromatography/mass spectrometry chemical analysis to derive machine learning based classification models trained with the partial least square discriminant analysis algorithm. The internally validated models were thus applied on untested essential oils to assess their effective predictive ability in selecting both active and low toxic samples. Five essential oils were selected among a list of 52 and readily assayed for IC50 and CC50 determination. Interestingly, four out of the five selected samples, compared with the potencies of the training set, returned to be highly active and endowed with low toxicity. In particular, sample CJM1 from Calaminta nepeta was the most potent tested essential oil with the highest selectivity index (IC50 = 0.063 mg/mL, SI > 47.5). In conclusion, it was herein demonstrated how multidisciplinary applications involving machine learning could represent a valuable tool in predicting the bioactivity of complex mixtures and in the near future to enable the design of blended essential oil possibly endowed with higher potency and lower toxicity.

Shale gas production from horizontal wells may encounter potential problems because of gas leakage from various zones of a well into the air and ground water reserves. Stopping such leaks is a serious challenge faced by the shale gas industry. To stop gas communication between various zones of the well, lifelong integrity of the cemented annulus between the metal casing and the borehole wall must be ensured. Various physical factors i.e. casing properties (internal diameter, centralizers and casing-hole relationship), cement and drilling mud properties (density, viscosity, additives), and other variables, such as temperature and pressure, affect the quality of cement job. The quality of cement job was analyzed in terms of sustained casing pressure. Sustained casing pressure or SCP results from sustained pressure on an annulus seen at surface from fluid or gas leaking from a lower formation as a result of poor zonal isolation. The SCP value was used in categorizing horizontal shale gas wells as leaking or not leaking. The statistical classification model was built to predict whether a well will leak or not, under the effect of various physical factors. The multivariate statistical technique PLS-DA (Partial Least Square Discriminant Analysis) was used to build this model and as the input (predictor) variables, the model used dimensionless groups of the physical factors. The VIP (variable importance in projection) variable selection method was used to identify highly impacting physical factors and the optimal model structure was determined using the 10-fold cross validation method. The model was able to correctly classify 81% of the classified wells in cross-validation tests for intermediate casing. The types of data-driven models developed are helpful in predicting whether annular gas leakage will occur under the influence of physical factors and based on the model feedback, the responsible factors can be regulated to perform better cement job, which would result in reduced gas leakage and less remedial cementing cost. [ABSTRACT FROM AUTHOR]

The prediction of biologically active compounds plays a very important role for high-throughput screening approaches in drug discovery. Most computational models, in this area, concentrate on measuring structural similarities between chemical elements. There are various methods to predict anti-HIV activity, such as artificial neural network and support vector machine, but generally using shallow machine learning with low accuracies and less samples. In this work, one of deep learning methods, stacked auto-encoder (SAE), is proposed to predict anti-HIV activity of a broad group of compounds for the first time. Through contrasting experiments of artificial neural network, support vector machine, and SAE under the same condition, the accuracy after descriptors screening is higher than using raw descriptors, and SAE performs better than the other two methods to achieve the perfect forecast of anti-HIV activity. It has a great significance on promoting anti-HIV drug design, which therefore can reduce research and development costs and improve the efficiency of anti-HIV drug discovery. [ABSTRACT FROM AUTHOR]

Wheat durum pasta represents one of the most typical Italian food products. Many factories realize different class products regarding the use of organic wheat or not and the application of traditional or industrial production parameters. Being not subordinate to regulations, these classifications are only indicative of the real processes representing different quantitative levels for the same class. Aim of this study is to set up a rapid and nondestructive method to distinguish between different declared production parameters of pasta brands (spaghetti) such as industrial (high temperature, short-time drying, Teflon drawn) and traditional (low temperature, long-time drying, bronze drawn) processes and the use of organic wheat or not. A hyperspectral system operating within visible and near-infrared spectra was used to acquire images of spaghetti bundles (of two different years). Hyperspectral information was statistically analyzed by multivariate provisional soft independent modeling of class analogy (SIMCA). The results report a percentage of correct classification equal to 75.3 % for the first year and to 73.9 % for the second year. For both sampling years, all the traditional brands are the most distant from the origin (i.e., full industrial model). The Spearman's cross-correlation test performed on the SIMCA distances indicates a statistically significant correlation between the 2 years of analysis confirming the system repeatability. The results demonstrate that the differentiation of pasta does not depend only on the raw material (i.e., organic and not) but also on the times and types of processing (i.e., short- and long-time drying, bronze or Teflon wire drawing). [ABSTRACT FROM AUTHOR]

Prevention of drug dispensing errors is an importance topic in medical care. In this paper, we propose a risk management approach, namely Hybrid Data Mining (HDM), to prevent the problem of drug dispensing errors. An intelligent drug dispensing errors prevention system based on the proposed approach is then implemented. The proposed approach consists of two main procedures: First, the classification modeling and logistic regression approaches are used to derive decision tree and regression function from the given dispensing errors cases and drug databases. In the second procedure, similar drugs are then gathered together into clusters by combing clustering technique ( PoCluster) and the extracted logistic regression function. The drugs that may cause dispensing errors will then be alerted through the clustering results and the decision tree. Through experimental evaluation on real datasets in a medical center, the proposed approach was shown to be capable of discovering the potential dispensing errors effectively. Hence, the proposed approach and implemented system serve as very useful application of data mining techniques for risk management in healthcare fields. [ABSTRACT FROM AUTHOR]

This paper analyses method skew in India over more than two decades using a new method skew index. The analysis reveals clear regional pattern in method skew in the country. In north and east regions of the country, method skew is average and below average, but it is high or very high in the south region. We investigate roles of such factors as poverty, education, social class, and religion in deciding the method skew at the district level using the classification modeling approach and find that the roles of these factors are different in different regions of the country. An important finding of the analysis is that there is a positive correlation between the degree of the method skew and contraceptive prevalence at the district level., (© 2021 The Population Council, Inc.)

• Metabolic biomarker profiles were used for the prediction of clinical diagnoses. • k-NN classifiers were fitted to demonstrate the reliability of µPADs derived data. • Predictors were associated with glucose, lactate, and uric acid concentrations. • Metabolic disorders were suitably predicted with 95.0% overall accuracy. • The modeling addressed here proved that the µPADs could be applied in telemedicine. The development of microfluidic devices for point-of-care (POC) testing has been widely applied for food quality control, environmental monitoring, and clinical diagnoses. Regarding the latter, microfluidic paper-based analytical devices (µPADs) have a great potential in developing portable and disposable platforms for fast disease-relevant biomarker detection in remote areas. In this work, k-nearest neighbors (k-NN) classifiers were applied to demonstrate the reliability of using chemical data obtained from µPADs for diagnosis of diseases like diabetes mellitus and hyperuricemia. Predictors based on glucose, lactate, and uric acid biomarkers were evaluated to reach this goal. Healthy adults were differentiated from the diagnosed patients with a perfect success rate. Metabolic disorders were suitably predicted with 95.0% overall accuracy. This model also showed sensitivities and specificities in the ranges 75–100% and 93–100%, respectively, indicating that the classification protocol is robust. The k-NN modeling proved the potential of the µPADs to be applied in telemedicine for reliable off-site diagnoses. [ABSTRACT FROM AUTHOR]

In the last decade essential oils have attracted scientists with a constant increase rate of more than 7% as witnessed by almost 5000 articles. Among the prominent studies essential oils are investigated as antibacterial agents alone or in combination with known drugs. Minor studies involved essential oil inspection as potential anticancer and antiviral natural remedies. In line with the authors previous reports the investigation of an in-house library of extracted essential oils as a potential blocker of HSV-1 infection is reported herein. A subset of essential oils was experimentally tested in an in vitro model of HSV-1 infection and the determined IC50s and CC50s values were used in conjunction with the results obtained by gas-chromatography/mass spectrometry chemical analysis to derive machine learning based classification models trained with the partial least square discriminant analysis algorithm. The internally validated models were thus applied on untested essential oils to assess their effective predictive ability in selecting both active and low toxic samples. Five essential oils were selected among a list of 52 and readily assayed for IC50 and CC50 determination. Interestingly, four out of the five selected samples, compared with the potencies of the training set, returned to be highly active and endowed with low toxicity. In particular, sample CJM1 from Calaminta nepeta was the most potent tested essential oil with the highest selectivity index (IC50 = 0.063 mg/mL, SI > 47.5). In conclusion, it was herein demonstrated how multidisciplinary applications involving machine learning could represent a valuable tool in predicting the bioactivity of complex mixtures and in the near future to enable the design of blended essential oil possibly endowed with higher potency and lower toxicity. [ABSTRACT FROM AUTHOR]