Theprediction Accuracy Research Articles

Identification of genetically plastic forms among Belarusian ancient flax (Linum usitatissimum convar. elongatum Vav. et Ell.) varieties using the Linum Insertion Sequence LIS-1

The Linum Insertion Sequence 1 (LIS-1) occurs in the genetically plastic flax genotypes in response to the lack or excess of mineral and water nutrition, but also naturally, and can be transmitted to the progeny. We have analyzed 21 ancient Belarusian varieties of flax Linum usitatissimum convar. elongatum Vav. et Ell. The LIS-1 presence or absence was checked for individual plants in at minimum two generations with primer-specific polymerase chain reaction (PCR) and agarose gel electrophoresis. The studied flax varieties formed four groups: non-responsive varieties (LIS-1 was not found, group NR); responsive, which formed and completely lost the insertion (group R0); responsive, which formed and retained LIS-1 (group R1); and responsive unstable (group R2). A statistically significant difference was found in ‘plant height’ (p < 0.05), ‘technical length of the stem’ (p < 0.05) between R0 and NR, and R2 and NR LIS-1 groups. The machine learning algorithm random forest classifier was used to predict the presence, absence or heterozygosity of LIS-1 in flax plants based on their growth and reproductive characteristics. As a result, the accuracy of the prediction was 98% on test data. In terms of sources for the selection of fibre flax varieties adaptive to environmental challenges, the most promising group consists of responsive varieties that have formed LIS-1 insertion (R0, R1 and R2 groups).

Size, shape, and dimension effects on the melting temperature of metallic nanocrystals

Melting is the most common phenomenon in nature and one of the most important properties of metallic materials. Exploring the size D, shape α, and dimension d effects on the melting temperature T m of nanocrystals is of great significance for the design, fabrication, and application of quantum devices. In this work, by redefining the critical diameter D 0 and introducing shape factor α, a unified model without any adjustable parameters has been developed to describe the T m(D, α, d) function. The model is compared with the available experimental and simulation data of Cu, Pd, In, Pb, Au, Ag, and Ni nanocrystals and other theoretical works, and a consistent agreement is obtained, which verifies the accuracy of the prediction. This model is also compared with other theoretical works, and we find that it agrees well with Lu’s model, while the BOLS method underestimates the melting point. This work not only gives a new perspective on the relationship between size, shape, dimension, and melting temperature but also provides theoretical guidance for the design and optimization of low-dimensional quantum devices.

Immune-inflammatory biomarkers for the occurrence of MACE in patients with myocardial infarction with non-obstructive coronary arteries.

Neutrophil-to-high-density lipoprotein cholesterol ratio (NHR), monocyte-to-high-density lipoprotein cholesterol ratio (MHR), lymphocyte-to-high-density lipoprotein cholesterol ratio (LHR), platelet-to-high-density lipoprotein cholesterol ratio (PHR), systemic immune-inflammation index (SII), systemic inflammation response index (SIRI), and aggregate index of systemic inflammation (AISI) have been identified as immune-inflammatory biomarkers associated with the prognosis of cardiovascular diseases. However, the relationship of these biomarkers with the prognosis of myocardial infarction with non-obstructive coronary arteries (MINOCA) remains unclear. Patients with MINOCA who underwent coronary angiography at the 920th Hospital of Joint Logistics Support Force were included in our study. Clinical baseline characteristics and laboratory testing data were collected from the hospital record system. The patients were divided into two groups on the basis of major adverse cardiovascular events (MACE) occurrence. Multiple logistic regression analysis was conducted to assess the relationship between NHR, MHR, LHR, PHR, SII, SIRI, AISI, and MACE. Receiver operating characteristic (ROC) curves were generated to evaluate the predictive value of NHR, MHR, LHR, PHR, SII, SIRI, and AISI for MACE in patients with MINOCA. The accuracy of the prediction was indicated by the area under the curve (AUC) value. The study included 335 patients with MINOCA. (81 in the MACE group and 254 in the No-MACE group). The MACE group had higher levels of NHR, MHR, LHR, PHR, SII, SIRI, and AISI than the No-MACE group. Multiple logistic regression analysis adjusted for confounding factors indicated that the higher levels of NHR, MHR, PHR, SII, SIRI, and AISI were associated with the occurrence of MACE in patients with MINOCA (P < 0.001). The AUC values for NHR, MHR, PHR, SII, SIRI, and AISI were 0.695, 0.747, 0.674, 0.673, 0.688, and 0.676, respectively. The combination of NHR, MHR, PHR, SII, SIRI, and AISI improved the accuracy of predicting MACE in patients with MINOCA (AUC = 0.804). Higher levels of NHR, MHR, PHR, SII, SIRI, and AISI were associated with the occurrence of MACE, and the combination of NHR, MHR, PHR, SII, SIRI, and AISI improved the accuracy for predicting the incidence of MACE events in patients with MINOCA.

Prediction quality of patients’ composited outcomes of interest and how much variations on the outcomes explained by the patients’ sodium levels

Background: Hypernatremia can be caused by medications, glucose/urea-related hyperosmolarity, overdiuresis, compartmental fluid distribution imbalances, iatrogenic hypertonic saline excess, renal concentrating performance defects, overdiuresis, and hypermetabolic states. Patients who are in critical condition, encompassing ailments of the heart, liver, lungs, brain, and kidneys, frequently manifest hyponatremia (sodium &lt; 135 mEq/L). Similar to hypernatremia, hyponatremia signifies the severity of the illness and a bleak prognosis. Hyponatremia is less lethal than central pontine myelinolysis, which overcorrects sodium concentrations urgently. Aim: The objective of this research endeavour was to investigate the strength of correlations, the extent to which the independent variables account for the overall variability in the dependent variable, and the accuracy of the dependent variable's predictions. In addition, the purpose of this test was to extract the coefficients required to present the Binary Logistic Regression models that were investigated. Methods: Between May 2018 and May 2021, this study examined two Na-based prognosticators for overall mortality in critically ill patients admitted to the King Hussein Medical Centre. An investigation was conducted on 2155 cases utilising Receiver Operating Characteristic (ROC) and Sensitivity Indices Test analyses to ascertain the most effective cut-off points, specificities, sensitivity, predictive values, likelihood ratios, Youden and accuracy indices, and mortality rates. The research discovered that serial termination points with lower values suggested more robust evidence for a positive actual state, whereas higher values suggested less robust evidence for a negative actual state. A Binary Logistic Regression (BLgR) test was performed to examine correlations, the extent of total variations in the dependent variable, and the accuracy of the prediction for each Na-related mortality prognosticator in relation to overall mortality. The analysis was conducted utilising version 25 of SPSS for Windows. Results: The research revealed that the AUROC for cNa was considerably greater in magnitude compared to its prognosticator that is genealogically related to Na. 129.35 mEq/l and 131.05 mEq/l were the optimal cut-off points, TPRs, TNRs, PPVs, NLRs, YIs, and AIs for the two comparative Na investigated statutes. The BLgR models were developed in order to simulate the relationship between the corrected and measured Na levels of patients and their overall mortality rate. For critically ill patients, the probabilities of overall mortality at the optimal cutoff operating Na levels were 79.59 and 81.62 percent, respectively.

PENGARUH ROA, ROE, LDR, CR, DAN NPL TERHADAP KONDISI FINANCIAL DISTRESS PADA PERUSAHAAN PERBANKAN YANG TERDAFTAR DI BURSA EFEK INDONESIA PERIODE 2019-2022”

The study aims to predict financial distress through the financial ratio variables Return on Asset, Return On Equity, Loan to Deposit Ratio, Current Ratio, and Non-Performing Loan, from one, two and three years before occurring in banking companies listed on the Indonesian Stock Exchange. The population in this study is 47 banking companies. Sampling techniques are purposive sampling according to certain criteria so that 35 samples are obtained from the company. Data obtained from the Indonesian Stock Exchange for the years 2019-2022. The data analysis technique used in this study is logistic regression analysis. The results of the research show that the return on asset, return on equity, Loan to Deposit Ratio, Current ratio, and Non-Performing Loan, together, significantly influenced the probability of financial distress in banking companies listed on the Indonesian Stock Exchange for one, two and three years before it occurred. The degree of accuracy of the prediction of the influence of the financial variable ratio on the probability of financial distress for one, two and three years before having suffered financial distresses was 88.6 per cent, 88.6, and 82.9 per cent. The test results partially showed that the Return on Asset from the first, second and third years before the financial distress has a positive and non-significant effect on the likelihood of financial distress, the Return on Equity from one, two and three years before the financial distress had a negative and significant impact on the probability of the financial distress, the Loan to Deposit Ratio from one, two and three years before the financial distress had a negative and significant impact on the probability of the financial distress, the Ratio of Current Ratio from one and two years before financial distress has a negative and no significant effect on financial distress probability, and the Non-Performing Loan from three years prior to financial distress have a significant negative and positive impact on financial distress probability.

Machine learning prediction and characterization of sigma-free high-entropy alloys

Precipitation hardening provides a great potential to achieve superior strength-ductility trade-off in high-entropy alloys (HEAs). However, the formation of some precipitates such as brittle sigma phases may significantly deteriorate the ductility. Empirical criteria such as PSFE, VEC, and Md were proposed to predict the formation of undesirable phases during aging, nevertheless, they are insufficient to describe their stability and are not dependable for the noted purpose. Accordingly, for the first time, machine-learning models were conducted for this prediction. A feature selection based on Pearson correlation and mutual information has been used to improve the performance of models. Linear regression, second-order polynomial regression, decision tree, and neural network are conducted by machine learning methods in this paper. Among the implemented methods, the neural network had the best performance which improved the accuracy of the prediction by ∼20% compared to the conventional methods. Additionally, a new parameter based on linear regression, which is more reliable and user-friendly than thermodynamic parameters was developed in this investigation. The results showed that some elements such as Cr, Mo, and V may promote sigma precipitation. A HEA with a sigma-free microstructure was developed for validation of the model. We strongly believe that this HEA has a great potential for overcoming the strength-ductility trade-off because of not only the lack of sigma precipitate but also the presence of NiAl precipitate. The results are a step forward in designing alloys with the potential of microstructure engineering to achieve superior properties.

Prediction of Postoperative Range of Motion after Mobile-Bearing Medial Unicompartmental Knee Arthroplasty from the Preoperative Range of Motion and Other Preoperative Factors.

Postoperative flexion after unicompartmental knee arthroplasty might be predicted from the preoperative range of motion and other preoperative factors, but this has not been sufficiently investigated. Between 2013 and 2017, 198 patients (198 knees) underwent unilateral knee arthroplasty with medial mobile-bearing unicompartmental knee arthroplasty. Range of motion was measured preoperatively and at the time of final follow-up. To investigate the accuracy of the prediction of preoperative to postoperative gain or loss of the flexion angle, we performed receiver operating characteristic analysis. Logistic regression analysis was used to evaluate other predictive factors. Change in flexion angle was significantly strongly and negatively correlated with the preoperative flexion angle (R = - 0.688; 95% confidence interval: -0.755 to -0.607; p < 0.001). Preoperative flexion angle was suggested to be a significant predictor of gain or loss of the flexion angle with the area under the curve of 0.781; the cutoff value calculated using the Youden index was 140 degrees. Logistic regression analysis showed that in addition to the preoperative flexion angle of the operated side, the postoperative flexion range was significantly affected by the patient's height and by the preoperative flexion angle of the contralateral knee. If the preoperative flexion angle in Oxford mobile-bearing medial unicompartmental knee arthroplasty is <140 degrees, the postoperative flexion angle may be improved; if it exceeds 140 degrees, the postoperative flexion angle may worsen. This predictive ability is further improved by consideration of the patient's height and the range of motion on the contralateral side.

Data Analysis of Student Academic Performance and Prediction of Student Academic Performance Based on Machine Learning Algorithms

With the development and popularization of education, the quality of education has become one of the key factors in the development of a country. And students' academic performance, as one of the important indicators of education quality, has been attracting much attention. This paper mines and analyzes the data affecting students' academic performance, and also conducts a predictive study of students' academic performance using logistic regression model. In this study, 30 indicators such as gender, age, family size, parental education, parental occupation, family relationship, health, and the number of drinks per this paperek and per month this paperre used as input variables, and students' academic performance was categorized into SUCCESS and FAIL, and the training and test sets this paperre divided according to the ratio of 7:3, and the logistic regression model was used for training and prediction. The results show that the logistic regression model has high prediction accuracy in predicting students' academic performance (whether they fail or not), with an accuracy of 95.8%, precision of 96.7%, recall of 95.1%, and F1 of 95.8%. This indicates that the logistic regression model has high accuracy and reliability in predicting students' academic performance. The results of this study are important for schools and educational organizations. Through the prediction of students' academic performance, schools can identify students' learning problems in time and take targeted measures to help students improve their academic performance. Meanwhile, this study also provides some useful reference information for individual students to help them better understand their learning situation, adjust their learning strategies in time and improve their learning efficiency. In the future, the method can be further explored and improved to enhance the accuracy and reliability of the prediction and to provide better support and assistance for students' learning and development.

A Microfluidic and Numerical Analysis of Non-equilibrium Phase Behavior of Gas Condensates

Conventional assumptions about multiphase flow in gas condensate reservoirs often do not correlate with field production. This discrepancy stems from the various mechanisms influencing the multiphase process, which are inadequately represented in numerical models. One of the least understood mechanisms is the influence of the non-equilibrium thermodynamics on the flow in the wellbore region, where the reservoir pressure is below the dew point pressure. To address this problem, experimental and mathematical analyses were conducted using a microfluidic device designed to replicate the flow dynamics in a gas condensate system. The experimental results showed an 11% deviation from the initial pressure of condensate saturation when compared with the conventional assumption of local equilibrium in numerical models. Similarly, there is a 14% deviation between the experimental and simulated volumes of the condensate. These findings underscore the inadequacy of existing models to accurately predict the saturation profile of the condensate phase. A mathematical model based on a relaxation parameter was applied to account for non-equilibrium phase separation and the fog state of the aerosol as observed in the microfluidic experiment. Incorporating a relaxation parameter (τ\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ au$$\\end{document}) enhanced the accuracy of the prediction of the initial pressure of the condensate saturation and an improvement in the prediction of the condensate volumes from 76% to 97.2%. Consequently, it provides a valuable framework and insight on the non-equilibrium phase behavior of gas condensate systems under constant flow regimes.

A multi-channel spatial-temporal transformer model for traffic flow forecasting

Traffic flow forecasting is a crucial task in transportation management and planning. The main challenges for traffic flow forecasting are that (1) as the length of prediction time increases, the accuracy of prediction will decrease; (2) the predicted results greatly rely on the extraction of temporal and spatial dependencies from the road networks. To overcome the challenges mentioned above, we propose a multi-channel spatial-temporal transformer model for traffic flow forecasting, which improves the accuracy of the prediction by fusing results from different channels of traffic data. Our approach leverages graph convolutional network to extract spatial features from each channel while using a transformer-based architecture to capture temporal dependencies across channels. We introduce an adaptive adjacency matrix to overcome limitations in feature extraction from fixed topological structures. Experimental results on six real-world datasets demonstrate that introducing a multi-channel mechanism into the temporal model enhances performance and our proposed model outperforms state-of-the-art models in terms of accuracy.

An advancing method for web service reliability and scalability using ResNet convolution neural network optimized with Zebra Optimization Algorithm

AbstractWeb service reliability and scalability is an important mission that keeps web services running normally. Within web service, the web services invoked by users not only depend on the service itself, but also on web load condition. Due to the features of web dynamics, traditional reliability and scalability methods have become inappropriate; at the same time, the web condition parameter sparsity problem will cause inaccurate reliability prediction. To address these challenges, Web Service Reliability and Scalability Determination Using ResNet Convolutional Neural Network optimized with Zero Optimization Algorithm (WRS‐ResNetCNN‐ZOA) is proposed in this manuscript. Initially, the input data is collected from WSRec dataset. The ResNet convolutional neural network (ResNetCNN) with Business Process Execution Language (BPEL) specification is introduced to forecast the reliability and scalability of web service. The results are categorized as right and wrong based on ResNetCNN. The weight parameters of the ResNetCNN is optimized by Zebra Optimization Algorithm to improve accuracy of the prediction. The performance of the proposed method is examined under some performance metrics, like F‐measure, reliability, scalability, accuracy, sensitivity, specificity, and precision. The proposed technique attains 15.36%, 35.39%, 23.87%, 20.67% better reliability, 42.39%, 11.39%, 34.16%, 25.78% better accuracy when analyzed to the existing methods, like Web Reliability based on K‐clustering, (WRS‐KClustering), Web Reliability prediction based on AdaBoostM1 and J48 (WRS‐AdaM1‐J48), Web Reliability prediction based on Online service Reliability (WRS‐OPUN), and Web Reliability prediction based on Dynamic Bayesian Network (WRS‐DBNS), respectively.

Using nonlinear auto-regressive with exogenous input neural network (NNARX) in blood glucose prediction

BackgroundPredicting of future blood glucose (BG) concentration is important for diabetes control. Many automatic BG monitoring or controlling systems use BG predictors. The accuracy of the prediction for long prediction time is a major factor affecting the performance of the control system. The predicted BG can be used for glycemia management in the form of early hypoglycemic/hyperglycemic alarms or adjusting insulin injections. Recent developments in continuous glucose monitoring (CGM) devices open new opportunities for glycemia management of diabetic patients. Many of those systems need prediction for long prediction horizons to avoid going through hypo or hyperglycemia.MethodsIn this article a nonlinear autoregressive exogenous input neural network (NNARX) is proposed to predict the glucose concentration for longer prediction horizons (PHs) than that was obtained previously with an established recurrent neural network (RNN). The proposed NNARX is a modified version from our previously published RNN with different initialization and building technique but has the same architecture. The modification is based on starting with building nonlinear autoregressive exogenous input model using MATLAB and train it, then close the loop to get NNARX network.ResultsThe results of using the proposed NNARX indicate that the proposed NNARX is better in prediction and stability than unmodified RNN as PH becomes higher than 45 minutes.ConclusionsModification in RNN building extends the ability of the prediction till 100 minutes. It performs statistically significant improvements in the FIT and RMSE values for 100 minutes prediction. It also decreases root mean squared error (RMSE) for both 45 and 60 minutes of prediction.

Combustion Characteristic Prediction of Dual Direct Injection Fuel (Diesel-Propane) on RCEM Based on an Artificial Neural Network Approach

Studies from around the world show that engines using biofuel, LPG, and CNG emit fewer pollutants than those using conventional fuels. Experimental research has focused on a rapid compression and expansion machine (RCEM) that resembles a compression ignition (CI) engine. It uses dual direct injection fuel, diesel and propane (DP), with propane injection timing varying from 0 to 40 before top dead center (BTDC) and diesel injection timing remaining at 10 BTDC. The compression ratio was changed at points 17 and 19 by adjusting the RCEM connecting rod. A converge simulation program was used to run the simulation model, which was used to examine how the fire and inflow inside the chamber developed. The ANN method was used to predict pressure, temperature, power, TKE, and ITE data output based on propane energy fraction, compression ratio, and SOI of propane as input data parameters. It was noticed that the ANN prediction on experimental data has a higher accuracy compared to the simulation prediction. The R and MSE values were used to identify the accuracy of the prediction on output parameter data. ANN generalization capability is comparatively high when trained with large enough databases. The highest accuracy of prediction was produced on TKE, which had an MSE of 0.003715 and R value of 0.99981 from 287900 sample data. This shows that the ANN model is quite accurate in forecasting output experimental data.

A novel Bayesian optimization prediction framework for four-axis industrial robot joint motion state

Robot joints are the main structure for controlling the motion of the machine body, where the motion state of them directly affects the performance of the industrial robot. Due to the difficulty of obtaining the joint torque information of industrial robots, it is very hard to monitor the motion state of them. Based on the velocity and force driven by current of motors, we propose a novel Bayesian optimization framework to predict the joint motion state of industrial robot in this paper. Based on the temporal correlation of joint current and the correlation between the current and motion state of joint, we use the LSTM and BiLSTM to regressing prediction of the current and state of joint motor first. Then, the Bayesian optimization method is used to adjust the hyperparameters of our network, which realize the analysis of the joint motor current under different motion states and improve the accuracy of the prediction of joint motion states. Finally, we design the joint current acquisition platform of industrial robot based on Hall current sensors, which can collect joint currents without contact and generate experimental dataset. Comparing with the popular intelligent methods, the results show that our Bayesian optimization framework realizes a more accurate prediction of motion state for the four-axis industrial robot on the basis of contact-less current acquisition.

Artificial neural network-assisted prediction of radiobiological indices in head and neck cancer.

We proposed an artificial neural network model to predict radiobiological parameters for the head and neck squamous cell carcinoma patients treated with radiation therapy. The model uses the tumor specification, demographics, and radiation dose distribution to predict the tumor control probability and the normal tissue complications probability. These indices are crucial for the assessment and clinical management of cancer patients during treatment planning. Two publicly available datasets of 31 and 215 head and neck squamous cell carcinoma patients treated with conformal radiation therapy were selected. The demographics, tumor specifications, and radiation therapy treatment parameters were extracted from the datasets used as inputs for the training of perceptron. Radiobiological indices are calculated by open-source software using dosevolume histograms from radiation therapy treatment plans. Those indices were used as output in the training of a single-layer neural network. The distribution of data used for training, validation, and testing purposes was 70, 15, and 15%, respectively. The best performance of the neural network was noted at epoch number 32 with the mean squared error of 0.0465. The accuracy of the prediction of radiobiological indices by the artificial neural network in training, validation, and test phases were determined to be 0.89, 0.87, and 0.82, respectively. We also found that the percentage volume of parotid inside the planning target volume is the significant parameter for the prediction of normal tissue complications probability. We believe that the model has significant potential to predict radiobiological indices and help clinicians in treatment plan evaluation and treatment management of head and neck squamous cell carcinoma patients.

Optimized Bags of Artificial Neural Networks Can Predict the Viability of Organisms Exposed to Nanoparticles.

Prediction of organismal viability upon exposure to a nanoparticle in varying environments─as fully specified at the molecular scale─has emerged as a useful figure of merit in the design of engineered nanoparticles. We build on our earlier finding that a bag of artificial neural networks (ANNs) can provide such a prediction when such machines are trained with a relatively small data set (with ca. 200 examples). Therein, viabilities were predicted by consensus using the weighted means of the predictions from the bags. Here, we confirm the accuracy and precision of the prediction of nanoparticle viabilities using an optimized bag of ANNs over sets of data examples that had not previously been used in the training and validation process. We also introduce the viability strip, rather than a single value, as the prediction and construct it from the viability probability distribution of an ensemble of ANNs compatible with the data set. Specifically, the ensemble consists of the ANNs arising from subsets of the data set corresponding to different splittings between training and validation, and the different bags (k-folds). A machine uses a single partition (or bag) of k - 1 ANNs each trained on 1/k of the data to obtain a consensus prediction, and a k-bag machine quorum samples the k possible machines available for a given partition. We find that with increasing k in the k-bag or machines, the viability strips become more normally distributed and their predictions become more precise. Benchmark comparisons between ensembles of 4-bag machines and fraction machines suggest that the fraction machine has similar accuracy while overcoming some of the challenges arising from divergent ANNs in the 4-bag machines.

Serum cholesterol level as a predictive biomarker for prognosis of Neuroblastoma

BackgroundNeuroblastoma (NB), a type of solid tumor in children, has a poor prognosis. Few blood biomarkers can accurately predict the prognosis, including recurrence and survival, in children with NB. In this study, we found that the serum total cholesterol (Tchol) level was associated with the prognosis of patients through a retrospective study.MethodsMultivariate Cox regression model was used to identify the independent risk factors in the children with NB. Kaplan–Meier method was used to analyze the correlation between the common biomarkers, including the serum Tchol level, and the prognosis of the patients. ROC curves were used to predict the accuracy of the International Neuroblastoma Staging System (INSS) stage and Children’s Oncology Group (COG) risk stratification after adding the serum Tchol level.ResultsCompared with the other patients, serum Tchol level was significantly increased in the relapsed and died patients (P < 0.05). Subsequently, serum Tchol level was found as an independent risk factor to affect the outcome of patients (P < 0.05). Finally, we added serum Tchol level into traditional stage and risk classification system to form the new INSS stage and COG risk classification system. It was found that the areas under the ROC curve (AUC) of recurrence-free survival in the new INSS stage and COG risk classification system were increased to 0.691 (95%CI: 0.535–0.847) and 0.748 (95%CI: 0.622–0.874), respectively. Moreover, the AUC areas of overall survival in the new INSS stage and COG risk classification system were increased to 0.722 (95%CI: 0.561–0.883) and 0.668 (95%CI: 0.496–0.819), respectively.ConclusionWe found that serum Tchol level, a clinical biomarker, is a risk factor for recurrence and death among the children with NB. The serum Tchol level could significantly increase the accuracy of the prediction for NB prognosis.

Early prediction of ventricular peritoneal shunt dependency in aneurysmal subarachnoid haemorrhage patients by recurrent neural network-based machine learning using routine intensive care unit data

Aneurysmal subarachnoid haemorrhage (aSAH) can lead to complications such as acute hydrocephalic congestion. Treatment of this acute condition often includes establishing an external ventricular drainage (EVD). However, chronic hydrocephalus develops in some patients, who then require placement of a permanent ventriculoperitoneal (VP) shunt. The aim of this study was to employ recurrent neural network (RNN)-based machine learning techniques to identify patients who require VP shunt placement at an early stage. This retrospective single-centre study included all patients who were diagnosed with aSAH and treated in the intensive care unit (ICU) between November 2010 and May 2020 (n = 602). More than 120 parameters were analysed, including routine neurocritical care data, vital signs and blood gas analyses. Various machine learning techniques, including RNNs and gradient boosting machines, were evaluated for their ability to predict VP shunt dependency. VP-shunt dependency could be predicted using an RNN after just one day of ICU stay, with an AUC-ROC of 0.77 (CI: 0.75–0.79). The accuracy of the prediction improved after four days of observation (Day 4: AUC-ROC 0.81, CI: 0.79–0.84). At that point, the accuracy of the prediction was 76% (CI: 75.98–83.09%), with a sensitivity of 85% (CI: 83–88%) and a specificity of 74% (CI: 71–78%). RNN-based machine learning has the potential to predict VP shunt dependency on Day 4 after ictus in aSAH patients using routine data collected in the ICU. The use of machine learning may allow early identification of patients with specific therapeutic needs and accelerate the execution of required procedures.

A novel fatigue life prediction methodology based on energy dissipation in viscoelastic materials, synergistic effects of stress level, stress ratio, and temperature

This work introduces a new methodology to predict the fatigue life of viscoelastic materials by considering the creep effect on fatigue behavior under the concurrent effects of stress level, stress ratio, and temperature. The model established based on the total amount of energy dissipated during fatigue loading. To estimate the amount of dissipated energy under varying stress ratios and temperatures, two shift factors, ψcyclic(R,T) and ψcreep(R,T), were derived, attributed to the cyclic and creep parts of fatigue loading. These shift factors were subsequently incorporated into defined equilibrium equations to create the relationship between estimated dissipated energy and fatigue life. The input data for the model consisted of the dissipated energy and cyclic creep values obtained from experiments conducted at a reference stress ratio of 0.5 and reference temperature of 20 °C, together with the storage and loss moduli measured from one dynamic mechanical analysis (DMA) experiment in the temperature range of 15 °C–60 °C on a fully-cured epoxy adhesive. To validate the accuracy of the results, the predicted fatigue life at three temperatures of 20 °C, 40 °C and 55 °C, each loaded under three stress ratios of 0.1, 0.5, and 0.9 were compared with the experiments conducted under the same conditions. Almost all predicted results were in good agreement with the experiments, nevertheless; at the stress ratio of 0.9 at 20 °C, due to the significant change in the cyclic creep behavior, the accuracy of the prediction was lower. The developed model was used as a new constant life diagram (CLD) formulation, which afterwards was further developed to plot three-dimensional constant life diagrams (3-D CLDs) in which the constant life surfaces were a synergistic function of stress ratio and temperature.

Zero-Field Splitting Tensor of the Triplet Excited States of Aromatic Molecules: A Valence Full-π Complete Active Space Self-Consistent Field Study.

A method to predict the D tensor in the molecular frame with multiconfigurational wave functions in large active space was proposed, and the spin properties of the lowest triplets of aromatic molecules were examined with full-π active space; such calculations were challenging because the size of active space grows exponentially with the number of π electrons. In this method, the exponential growth of complexity is resolved by the density matrix renormalization group (DMRG) algorithm. From the D tensor, we can directly determine the direction of the magnetic axes and the ZFS parameters, D- and E-values, of the phenomenological spin Hamiltonian with their signs, which are not usually obtained in ESR experiments. The method using the DMRG-CASSCF wave function can give correct results even when the sign of D- and E-values is sensitive to the accuracy of the prediction of the D tensor and existing methods fail to predict the correct magnetic axes.