Prediction Of Retention Research Articles

Predictive analytics study to determine undergraduate students at risk of dropout

In this this work, a study is presented with quantitative variables using machine learning tools to detect undergraduate students at risk of dropping out and the factors associated with this behavior. Clustering algorithms and classification methods were tested to determine the predictive power of several variables regarding the dropout phenomenon on an unbalanced database of 14,495 undergraduate students with a real dropout rate of 8.5% and a retention rate of 91.5%. The usual classification criterion that assigns individuals to a class if their probability of belonging to it is greater than 50% provided accuracies of 13.2% in the dropout prediction and 99.4% in the retention prediction. Among eight classifiers, Random Forest was selected and applied along with Threshold Probability, which allowed us to gradually increase the dropout precision to more than 50%, while maintaining retention and global precisions above 70%. Through this study, it was found that the main variables associated with student dropouts were their academic performance during the early weeks of the first semester, their average grade in the previous academic levels, the previous mathematics score, and the entrance exam score. Other important variables were the number of class hours being taken, student age, funding status of scholarships, English level, and the number of dropped subjects in the early weeks. Given the trade-off between dropout and retention precisions, our results can guide educational institutions to focus on the most appropriate academic support strategies to help students at real risk of dropping out.

Role of Abdominal Ultrasound to Detect Benign Prostatic Hyperplasia with Significant Factor of Age

Mostly in men, benign prostatic hyperplasia is serious issue. The volume of BPH increasing withurinary retention and is mostly determined by transabdominal sonography.Prostatic volumeinteractednegatively withhigh density lipoprotein cholesterol. Mostly occurrenceof BPHinmen with age of 45 to 60 years. BPH have lower urinary tract symptoms. About 20 recentpublicationssuitableforliteraturereviewwereretrievedfromdifferentdatabaseslikePubMed, sci.hub, and google scholar. The search on databases & search engines identified 27articles related to topic, among which only 20 articles were as per requirement. Only 20 articleswere usedfor extraction of data related to role of Abdominalultrasound todetectbenignprostatic hyperplasia. The current study looked at the detection of BPH with age. This review hasconclusionsthatultrasoundisamodalityofchoiceforevaluatingthepatientwithBPH,ithelpsin diagnosis as well as prediction of urinary retention and age has a significant factor in BPH.

Pre-operative prediction of post-operative urinary retention in lumbar surgery: a prospective validation of machine learning model.

Predicting urinary retention is difficult. The aim of this study is to prospectively validate a previously developed model using machine learning techniques. Patients were recruited from pre-operative clinic. Prediction of urinary retention was completed pre-operatively by 4 individuals and compared to ground truth POUR outcomes. Inter-rater reliability was calculated with intercorrelation coefficient (2,1). 171 patients were included with age 63 ± 14years, 58.5% (100/171) male, BMI 30.4 ± 5.9kg/m2, American Society of Anesthesiologists class 2.6 ± 0.5, 1.7 ± 1.0 levels, 56% (96/171) fusions. The observed rate of POUR was 25.7%. The model's performance was found to be 0.663 (0.567-0.759). With a regression model probability cutoff of 0.24 and a neural network cutoff of 0.23, the following predictive power was achieved: specificity 90.6%, sensitivity 22.7%, negative predictive value 77.2%, positive predictive value 45.5%, and accuracy 73.1%. Intercorrelation coefficient for the regression aspect of the model was found to be 0.889 and intercorrelation coefficient for the neural network aspect of the model was found to be 0.874. This prospective study confirms performance of the prediction model for POUR developed with retrospective data, showing great correlation. This supports the use of machine learning techniques in the prediction of postoperative complications such as urinary retention.

Compendium of Chromatographic Behavior of Post-translationally and Chemically Modified Peptides in Bottom-Up Proteomic Experiments.

We have systematically evaluated the chromatographic behavior of post-translationally/chemically modified peptides using data spanning over 70 of the most relevant modifications. These retention properties were measured for standard bottom-up proteomic settings (fully porous C18 separation media, 0.1% formic acid as ion-pairing modifier) using collections of modified/nonmodified peptide pairs. These pairs were generated by spontaneous degradation, chemical or enzymatic treatment, analysis of synthetic peptides, or the cotranslational incorporation of noncanonical proline analogues. In addition, these measurements were validated using external data acquired for synthetic peptides and enzymatically induced citrullination. Working in units of hydrophobicity index (HI, % ACN) and evaluating the average retention shifts (ΔHI) represent the simplest approach to describe the effect of modifications from a didactic point of view. Plotting HI values for modified (y-axis) vs nonmodified (x-axis) counterparts generates unique slope and intercept values for each modification defined by the chemistry of the modifying moiety: its hydrophobicity, size, pKa of ionizable groups, and position of the altered residue. These composition-dependent correlations can be used for coarse incorporation of PTMs into models for prediction of peptide retention. More accurate predictions would require the development of specific sequence-dependent algorithms to predict ΔHI values.

A multi-target QSRR approach to model retention times of small molecules in RPLC

Quantitative structure-retention relationship models (QSRR) have been utilized as an alternative to costly and time-consuming separation analyses and associated experiments for predicting retention time. However, achieving 100 % accuracy in retention prediction is unrealistic despite the existence of various tools and approaches. The limitations of vast data availability and time complexity hinder the use of most algorithms for retention prediction. Therefore, in this study, we examined and compared two approaches for modelling retention time using a dataset of small molecules with retention times obtained at multiple conditions, referred to as multi-targets (five pH levels: 2.7, 3.5, 5, 6.5, and 8 at gradient times of 20 min of mobile phase). The first approach involved developing separate models for predicting retention time at each condition (single-target approach), while the second approach aimed to learn a single model for predicting retention across all conditions simultaneously (multi-target approach). Our findings highlight the advantages of the multi-target approach over the single-target modelling approach. The multi-target models are more efficient in terms of size and learning speed compared to the single-target models. These retention prediction models offer two-fold benefits. Firstly, they enhance knowledge and understanding of retention times, identifying molecular descriptors that contribute to changes in retention behaviour under different pH conditions. Secondly, these approaches can be extended to address other multi-target property prediction problems, such as multi-quantitative structure Property(X) relationship studies (mt-QS(X)R).

Anomalous Retention Prediction Using Modelling Software in Gradient Reversed-Phase Liquid Chromatography: Why it Can Occur and How to Prevent It

Anomalous Retention Prediction Using Modelling Software in Gradient Reversed-Phase Liquid Chromatography: Why it Can Occur and How to Prevent It

Flavor-protein interactions for four plant protein isolates and whey protein isolate with aldehydes

Aldehydes are important flavor molecules to consider in plant-based products. Here, the flavor retention of a series of saturated aldehydes and mono-unsaturated aldehydes (2-alkenals) with different chain lengths (C4, C6, C8, and C10) in dispersions with protein isolates of pea, soy, fava bean, chickpea, and whey (as reference) was analyzed with APCI-TOF-MS. The headspace concentrations of alkenals were lower than aldehydes, meaning alkenals were retained more than saturated aldehydes. The retention was modeled by assuming hydrophobic interactions and covalent interactions. The ratio between the hydrophobic interaction parameter and the covalent parameter showed that covalent interactions are mainly important for butanal and butenal (C4). For the other aldehydes, hydrophobic interactions became increasingly important. Correlations were found between the chemical interaction parameters and the cysteine and methionine content of the different proteins. The obtained model parameters for each set of proteins and flavors allow the prediction of flavor retention when developing a flavored product with high protein content.

Clinical research nurse predictions of trial failure, recruitment and retention: a case for their early inclusion in trial design

BackgroundClinical research nurses are a key part of the clinical trial team but typically get involved later in the trial, usually during recruitment. The purpose of our study was to establish if CRNs who read the trial protocol can predict the performance of the trial.MethodsWe randomly selected 18 trial protocols with three statuses, terminated, withdrawn, and completed, from ClinicalTrials.gov, between 2014 and 2018 inclusive. We gave the protocols to five CRNs, asked them to make a judgement and provide a reason for that judgement (via a 12-item questionnaire) on the status of the trial (terminated, withdrawn or completed), if the trial met its recruitment target, if it recruited on time, and if it retained its participants. We also asked if it was likely a CRN was involved in the design of the trial. The CRNs were blinded to the study outcomes, did not receive any training on how to read a protocol and were prohibited from using/abstained from using the internet while completing the task.ResultsTwenty-three questionnaires on 23 trial protocols (18 different trials) were completed by 5 CRNs. The CRNs correctly predicted the trial status 48%, 95% CI: 29–67% (11/23) of the time; successful/unsuccessful recruitment 74%, 95% CI: 54–87% (17/23) of the time; on-time recruitment 70%, 95% CI: 49–84% (16/23) of the time; and participant retention 52%, 95% CI: 33–71% (12/23). CRNs identified 100% (sensitivity) of sites that hit their target and 63%, 95% CI: 36–84% (specificity) of sites that missed their target.ConclusionsCRNs are very good judges of trial recruitment and site performance issues and are a vital part of the clinical trial team. Taken with the ESP (Estimating Site Performance) study, we have made a strong case for broadening the trial team at the trial design stage. Early engagement of a broad skillset can potentially offset problems of recruitment, retention and trial failure.

Predicting academic performance in a course by universal features of LMS Moodle digital footprint

Student retention prediction is one of the most important problems of learning analytics. In the global scope research on the topic for higher education is rather extensive, there are cases of successful implementation of education support services in universities. The literature analysis shows of the growing interest in this problem in the Russian scientific and pedagogical community. At the same time, the specifics of Russian education does not allow direct transfer of foreign experience into the domestic educational system.The study reveals that a significant contribution to predicting student retention can be made by models for predicting academic performance in educational courses of the curriculum. The authors propose a structural model of a system for predicting academic performance, which includes a universal model based on generalized indicators of the digital footprint, a course-based model that takes into account the specifics of learning in a particular discipline, and a model based on the student’s educational profile.In the empirical study we trained 5 models for early prediction of interim assessment grades based on the universal indicators of the LMS Moodle student digital footprint. The most accurate model, especially in the first half of the semester, turned out to be ensemble-averaging models of logistic regression, random forest and gradient boosting. It was found that universal models are effective for detection of at-risk students in the discipline, the directions for further improvement of the universal model of performance prediction were determined and conditions for scaling the proposed approach to create a prognostic system for student retention to other educational institutions were formulated.

The driving mechanism of soil organic carbon biodegradability in the black soil region of Northeast China

The biodegradability of soil organic carbon (BSOC), defined as soil mineralization C per unit of soil organic carbon (SOC), is considered to be an important indicator of SOC stability and is closely related to the global C cycle. However, the magnitude and driving mechanism of BSOC in farmland remain largely unexplored, especially at the regional scale. Here, we conducted regional scale sampling to investigate latitude distribution pattern of BSOC and the relative contributions of biotic (soil micro-food web) and abiotic (climate and soil) drivers to BSOC in the black soil region of Northeast China. Results showed that BSOC declined with increasing latitude, which indicates that as the latitude increases, SOC becomes more stable in the black soil region of Northeast China. Over a range of latitude from 43°N to 49°N, BSOC was negatively correlated with soil micro-food web metrics of diversity (indicated by species richness), biomass and connectance, and soil factors of soil pH and clay content (CC), while it was positively correlated with climate factors of mean annual temperature (MAT), mean annual precipitation (MAP) and soil factor of soil bulk density (SBD). Among those predictors, soil micro-food web metrics were the most direct factors contributing to the variations of BSOC, which exerted the largest total effect on BSOC (−0.809). Collectively, our results provide convincing evidence that soil micro-food web metrics play a direct vital role in determining the distribution pattern of BSOC over a range of latitudes in the black soil region of Northeast China. This highlights the necessity of considering the role of soil organisms in regulating C dynamics in prediction of SOC mineralization and retention in the terrestrial ecosystem.

SpliceAI-10k calculator for the prediction of pseudoexonization, intron retention, and exon deletion.

SpliceAI is a widely used splicing prediction tool and its most common application relies on the maximum delta score to assign variant impact on splicing. We developed the SpliceAI-10k calculator (SAI-10k-calc) to extend use of this tool to predict: the splicing aberration type including pseudoexonization, intron retention, partial exon deletion, and (multi)exon skipping using a 10 kb analysis window; the size of inserted or deleted sequence; the effect on reading frame; and the altered amino acid sequence. SAI-10k-calc has 95% sensitivity and 96% specificity for predicting variants that impact splicing, computed from a control dataset of 1212 single-nucleotide variants (SNVs) with curated splicing assay results. Notably, it has high performance (≥84% accuracy) for predicting pseudoexon and partial intron retention. The automated amino acid sequence prediction allows for efficient identification of variants that are expected to result in mRNA nonsense-mediated decay or translation of truncated proteins. SAI-10k-calc is implemented in R (https://github.com/adavi4/SAI-10k-calc) and also available as a Microsoft Excel spreadsheet. Users can adjust the default thresholds to suit their target performance values.

An RRAM retention prediction framework using a convolutional neural network based on relaxation behavior

The long-time retention issue of resistive random access memory (RRAM) brings a great challenge in the performance maintenance of large-scale RRAM-based computation-in-memory (CIM) systems. The periodic update is a feasible method to compensate for the accuracy loss caused by retention degradation, especially in demanding high-accuracy applications. In this paper, we propose a selective refresh strategy to reduce the updating cost by predicting the devices’ retention behavior. A convolutional neural network-based retention prediction framework is developed. The framework can determine whether an RRAM device has poor retention that needs to be updated according to its short-time relaxation behavior. By reprogramming these few selected devices, the method can recover the accuracy of the RRAM-based CIM system effectively. This work provides a valuable retention coping strategy with low time and energy costs and new insights for analyzing the physical connection between the relaxation and retention behavior of the RRAM device.

Atomistic Details of Peptide Reversed-Phase Liquid Chromatography from Molecular Dynamics Simulations

Peptide separations by reversed-phase liquid chromatography (RPLC) are an integral part of bottom-up proteomics. These separations typically employ C18 columns with water/acetonitrile gradient elution in the presence of formic acid. Despite the widespread use of such workflows, the exact nature of peptide interactions with the stationary and mobile phases is poorly understood. Here, we employ microsecond molecular dynamics (MD) simulations to uncover details of peptide RPLC. We examined two tryptic peptides, a hydrophobic and a hydrophilic species, in a slit pore lined with C18 chains that were grafted onto SiO2 support. Our simulations explored peptide trapping, followed by desorption and elution. Trapping in an aqueous mobile phase was initiated by C18 contacts with Lys butyl moieties. This was followed by extensive anchoring of nonpolar side chains (Leu/Ile/Val) in the C18 layer. Exposure to water/acetonitrile triggered peptide desorption in a stepwise fashion; charged sites close to the termini were the first to lift off, followed by the other residues. During water/acetonitrile elution, both peptides preferentially resided close to the pore center. The hydrophilic peptide exhibited no contacts with the stationary phase under these conditions. In contrast, the hydrophobic species underwent multiple transient Leu/Ile/Val binding interactions with C18 chains. These nonpolar interactions represent the foundation of differential peptide retention, in agreement with the experimental elution behavior of the two peptides. Extensive peptide/formate ion pairing was observed in water/acetonitrile, particularly at N-terminal sites. Overall, this work uncovers an unprecedented level of RPLC molecular details, paving the way for MD simulations as a future tool for improving retention prediction algorithms and for the design of novel column materials.

Novel computer-assisted approach to quick prediction and optimization of gradient separation for online enrichment-reversed phase liquid chromatography tandem system

An algorithm capable of predicting and optimizing the gradient separation of LC × LC system was developed in this paper. Two groups of structural analogues, five ginsenosides as well as eight bisphenols, which were difficult to discriminate in routine analysis, were used to verify the effectiveness of the proposed algorithm in fast separation optimization. Average errors of retention times below 1% were found in the retention prediction for all types of gradient programs, implying that the theory could lead to high quality in prediction of the retention times under gradients elution. Meanwhile, 84% of relative average deviations (RADs) between the predicted peak width and the measured ones were less than 20%. The larger deviation occurred at the time when the peak appeared while the gradient of the mobile phase changed, which led the deviations increased to 20%–42%. In all, method development and optimization for LC × LC tandem system was realized by the homemade user-friendly software. The present protocol may turn on great opportunities for the convenient method development in analysis of trace structural analogues in environmental, food and biological samples.

Standardized method for mechanistic modeling of multimodal anion exchange chromatography in flow through operation

Multimodal chromatography offers an increased selectivity compared to unimodal chromatographic methods and is often employed for challenging separation tasks in industrial downstream processing (DSP). Unfortunately, the implementation of multimodal polishing into a generic downstream platform can be hampered by non-robust platform conditions leading to a time and cost intensive process development. Mechanistic modeling can assist experimental process development but readily applicable and easy to calibrate multimodal chromatography models are lacking. In this work, we present a mechanistic modeling aided approach that paves the way for an accelerated development of anionic mixed-mode chromatography (MMC) for biopharmaceutical purification. A modified multimodal isotherm model was calibrated using only three chromatographic experiments and was employed in the retention prediction of four antibody formats including a Fab, a bispecific, as well as an IgG1 and IgG4 antibody subtype at pH 5.0 and 6.0. The chromatographic experiments were conducted using the anionic mixed-mode resin Capto adhere at industrial relevant process conditions to enable flow through purification. An existing multimodal isotherm model was reduced to hydrophobic interactions in the linear range of the adsorption isotherm and successfully employed in the simulation of six chromatographic experiments per molecule in concert with the transport dispersive model (TDM). The model reduction to only three parameters did prevent structural parameter non-identifiability and enabled an analytical isotherm parameter determination that was further refined by incorporation of size exclusion effects of the selected multimodal resin. During the model calibration, three linear salt gradient elution experiments were performed for each molecule followed by an isotherm parameter uncertainty assessment. Lastly, each model was validated with a set of step and isocratic elution experiments. This standardized modeling approach facilitates the implementation of multimodal chromatography as a key unit operation for the biopharmaceutical downstream platform, while increasing the mechanistic insight to the multimodal adsorption behavior of complex biologics.

Current trends in chromatographic prediction using artificial intelligence and machine learning.

Artificial intelligence (AI) and machine learning (ML) gained tremendous growth and are rapidly becoming popular in various fields of prediction due to their potential abilities, accuracy, and speed. Machine learning algorithms employ historical data to analyze or predict information using patterns or trends. AI and ML were most employed in chromatographic predictions and particularly attractive options for liquid chromatography method development, as they can help achieve desired results faster, more accurately, and more efficiently. This review aims at exploring various AI and ML models employed in the determination of chromatographic characteristics. This review also aims to provide deep insight into reported artificial neural network (ANN) associated techniques which maintained better accuracy and significant possibilities for chromatographic characteristics prediction in liquid chromatography over classical linear models and also emphasizes the integration of a fuzzy system with an ANN, as this integrated study provides more efficient and accurate methods in chromatographic prediction than other linear models. This study also focuses on the retention prediction of a target molecule employing QSRR methodology combined with an ANN, highlighting a more effective technique than the QSRR alone. This approach showed the benefits of combining AI or ML algorithms with the QSRR to obtain more accurate retention predictions, emphasizing the potential of artificial intelligence and machine learning for overcoming adversities in analytical chemistry.

Latest Algorithms in Machine and Deep Learning Methods to Predict Retention Rates and Dropout in Higher Education: A Literature Review

External factors, such as global impact, or internal factors, such as educational services or the quality of learning, can affect the Retention rate or Number of Dropouts (DO) of students in higher education. Higher education institutions must have a strategy to manage retention rates properly. They can take an initial approach by knowing the estimated retention rate or the number of DOs so they can anticipate it by determining the right strategy. Several researchers have researched retention prediction or DO using specific methods and algorithms. This literature review aims to provide an overview and analysis of the methods and algorithms used to predict retention rates or the Number of DOs of students in higher education and to know the latest trends in developing the algorithms used in predicting retention rates or DO. The method used in this research is the traditional literature review. We have identified Twenty-one articles according to the theme. From the article, there are 21 machine learning algorithms, 13 deep learning algorithms used, seven time-series algorithms, four feature selection algorithms, and three combinations of algorithms used. The review results show that deep learning algorithms provide higher accuracy values than machine learning algorithms, and the recent trend of using algorithms to predict retention levels or DO is towards using time series algorithms in deep learning methods.

Research on Customer Retention Prediction Model of VOD Platform Based on Machine Learning

Research on Customer Retention Prediction Model of VOD Platform Based on Machine Learning

Unified Linear and Nonlinear Models for Retention Prediction of Aliphatic Aldehydes and Ketones in Different Columns and Temperatures: Application of Atom-Type-Based AI Topological Indices

Unified Linear and Nonlinear Models for Retention Prediction of Aliphatic Aldehydes and Ketones in Different Columns and Temperatures: Application of Atom-Type-Based AI Topological Indices

Predicting Retention in Higher Education from high-stakes Exams or School GPA

ABSTRACT The transition from secondary to tertiary education varies from country to country. In many countries, secondary school is concluded with high-stakes, national exams, or high-stakes entry tests are used for admissions to tertiary education. In other countries, secondary-school grade point average (GPA) is the determining factor. In the Netherlands, both play a role. With administrative data of close to 180,000 students, we investigated whether national exam scores or secondary school GPA was a better predictor of tertiary first-year retention. For both university education and higher professional education, secondary school GPA was the better prediction of retention, to the extent that national exams did not explain any additional variance. Moreover, for students who failed their exam, being held back by the secondary school for an additional year and entering tertiary education one year later, GPA in the year of failure remained as predictive as for students who had passed their exams and started tertiary education immediately. National exam scores, on the other hand, had no predictive value at all for these students. It is concluded that secondary school GPA measures aspects of student performance that is not included in high-stakes national exams, but that are predictive of subsequent success in tertiary education.