Feasibility of the Optimal Design of AI-Based Models Integrated with Ensemble Machine Learning Paradigms for Modeling the Yields of Light Olefins in Crude-to-Chemical Conversions.

The prediction of the yields of light olefins in the direct conversion of crude oil to chemicals requires the development of a robust model that represents the crude-to-chemical conversion processes. This study utilizes artificial intelligence (AI) and machine learning algorithms to develop single and ensemble learning models that predict the yields of ethylene and propylene. Four single-model AI techniques and four ensemble paradigms were developed using experimental data derived from the catalytic cracking experiments of various crude oil fractions in the advanced catalyst evaluation reactor unit. The temperature, feed type, feed conversion, total gas, dry gas, and coke were used as independent variables. Correlation matrix analyses were conducted to filter the input combinations into three different classes (M1, M2, and M3) based on the relationship between dependent and independent variables, and three performance metrics comprising the coefficient of determination (R2), Pearson correlation coefficient (PCC), and mean square error (MSE) were used to evaluate the prediction performance of the developed models in both calibration and validations stages. All four single models have very low R2 and PCC values (as low as 0.07) and very high MSE values (up to 4.92 wt %) for M1 and M2 in both calibration and validation phases. However, the ensemble ML models show R2 and PCC values of 0.99-1 and an MSE value of 0.01 wt % for M1, M2, and M3 input combinations. Therefore, ensemble paradigms improve the performance accuracy of single models by up to 58 and 62% in the calibration and validation phases, respectively. The ensemble paradigms predict with high accuracy the yield of ethylene and propylene in the catalytic cracking of crude oil and its fractions.

Effects of Coriandrum sativum on distant organ inflammation and apoptosis due to liver ischemia/reperfusion injury.

Liver ischemia/reperfusion (IR) often affects distant organs, such as small intestine, kidney, and lung. Coriandrum sativum (CS) has an antioxidant and anti-inflammatory effect on liver damage. The aim of this study was to investigate the anti-inflammatory and antiapoptotic effects of CS extract on small intestine, lung, and kidney after the liver IR injury. Small intestine, lung, and kidney tissues were evaluated and scored in terms of cell degeneration, inflammation, and congestion, as well as caspase-3 (Cas-3) and cluster of differentiation 31 (CD31) immunostainings were carried out. Renal enzymes, creatinine and urea levels were measured biochemically in serum. After IR, a decrease in villi size, diffuse degeneration, epithelial cell shedding and extensive congestion in the capillaries were observed. Meanwhile, the number of degenerated villi and congestion decreased in the IR+CS group. Due to IR, increased congestion was detected in the interalveolar septum of the lungs and in the capillaries between the kidney tubules. It was also observed that the positively stained cells with Cas-3 and CD31 were increased in the lung, kidney, and small intestine tissues of the IR group, and decreased in the IR+CS group. Kidney enzymes, urea and creatinine levels were significantly increased in the IR group and decreased in the IR+CS group. In conclusion, it was observed that liver IR caused changes in distant organs, especially in the small intestine, lung, and kidneys. Damaging effects of IR as well as apoptosis and inflammation were found to be decreased in the groups treated with CS.

A multicentre prospective analysis of the incidence of pemphigoid diseases in Turkey.

BACKGROUND: The differentiation between the pemphigoid diseases is essential for treatment and prognosis. In Turkey, data on the incidence of these diseases are insufficient. Our aim in this study is to determine the incidence, demographics and clinical characteristics associated with diseases of the pemphigoid group. METHODS: We prospectively analysed 295 patients with pemphigoid who visited dermatology clinics of tertiary referral hospitals in 12 different regions of Turkey within a year. The diagnosis was based on clinical, histopathological, direct immunofluorescence (DIF) and serological (multivariant enzyme-linked immunosorbent assay [ELISA], indirect immunofluorescence and mosaic-based BIOCHIP) examinations. Clinical and demographic findings, aetiological factors and concomitant diseases observed in the patients were recorded. RESULTS: A total of 295 (female/male ratio: 1.7/1) patients with pemphigoid were diagnosed in 1-year period. The overall incidence rate of pemphigoid diseases was found to be 3.55 cases per million-years. The ratio of pemphigoid group diseases to pemphigus group diseases was 1.6. The most common pemphigoid type was bullous pemphigoid (BP, 93.2%). The others were epidermolysis bullosa acquisita (3.1%), pemphigoid gestationis (2.4%), linear IgA disease (1%) and mucous membrane pemphigoid (0.3%). The most common (26.8%) possible trigger of the bullous pemphigoid was gliptin derivative drugs. The most common concomitant diseases with pemphigoid were cardiovascular (27.8%) and neurological diseases (23.7%). CONCLUSIONS: This study showed that the increased frequency of bullous pemphigoid reversed the pemphigoid/pemphigus ratio in Turkey. Further studies are warranted regarding the reasons for this increase.

Chemometrics-based models hyphenated with ensemble machine learning for retention time simulation of isoquercitrin in Coriander sativum L. using high-performance liquid chromatography.

In this research, two nonlinear models, namely; adaptive neuro-fuzzy inference system and feed-forward neural network and a classical linear model were employed for the prediction of retention time of isoquercitrin in Coriander sativum L. using the high-performance liquid chromatography technique. The prediction employed the use of composition of mobile phase and pH as the corresponding input parameters. The performance indices of the models were evaluated using root mean square error, determination co-efficient, and correlation co-efficient. The results obtained from the simple models showed that subclustering-adaptive-neuro fuzzy inference system gave the best results in both the training and testing phases and boosted the performance accuracy of the simple models. The overall comparison of the results showed that subclustering-adaptive-neuro fuzzy inference system ensemble demonstrated outstanding performance and increased the accuracy of the single models and ensemble models in the testing phase, up to 35% and 3%, respectively.

Artificial intelligence-based models for the qualitative and quantitative prediction of a phytochemical compound using HPLC method.

Isoquercitrin is a flavonoid chemical compound that can be extracted from different plant species such as Mangifera indica (mango), Rheum nobile , Annona squamosal , Camellia sinensis (tea), and coriander ( Coriandrum sativum L.). It possesses various biological activities such as the prevention of thromboembolism and has anticancer, antiinflammatory, and antifatigue activities. Therefore, there is a critical need to elucidate and predict the qualitative and quantitative properties of this phytochemical compound using the high performance liquid chromatography (HPLC) technique. In this paper, three different nonlinear models including artificial neural network (ANN), adaptive neuro-fuzzy inference system (ANFIS), and support vector machine (SVM),in addition to a classical linear model [multilinear regression analysis (MLR)], were used for the prediction of the retention time (tR) and peak area (PA) for isoquercitrin using HPLC. The simulation uses concentration of the standard, composition of the mobile phases (MP-A and MP-B), and pH as the corresponding input variables. The performance efficiency of the models was evaluated using relative mean square error (RMSE), mean square error (MSE), determination coefficient (DC), and correlation coefficient (CC). The obtained results demonstrated that all four models are capable of predicting the qualitative and quantitative properties of the bioactive compound. A predictive comparison of the models showed that M3 had the highest prediction accuracy among the three models. Further evaluation of the results showed that ANFIS-M3 outperformed the other models and serves as the best model for the prediction of PA. On the other hand, ANN-M3proved its merit and emerged as the best model for tR simulation. The overall predictive accuracy of the best models showed them to be reliable tools for both qualitative and quantitative determination.