Forest Algorithm Research Articles

DNA Methylation-Based Classification of Hairy Cell Leukemia and Splenic B Cell Lymphoma

Improved classification of rare lymphoid neoplasms would be aided by a deeper understanding of their underlying molecular features and is important for diagnosis, prognosis and therapy. Tumor entities classified within the WHO category of splenic B cell lymphomas and leukemias often exhibit heterogenous, transecting features, and include hairy cell leukemia (HCL), splenic diffuse red pulp lymphoma (SDRPL), splenic marginal zone lymphoma (SMZL), and the newly described WHO entity, splenic B cell lymphoma/leukemia with prominent nucleoli (SBLPN); the latter including patients formerly classified as HCL-variant (HCL-V). Genome-wide epigenetic information provides a tumor cell fingerprint combining cell-of-origin and tumor-specific events. Here we used DNA methylation to perform an unbiased molecular subclassification and to explore novel biological aspects of these patients. Samples from patients with a pathological diagnosis of HCL, HCL-V, SDRPL and SMZL (made prior to the 5 th WHO revision and ICC classifications) were obtained from 19 institutions across 9 countries, totaling 367 patients. Cells were FACS-purified where necessary and DNA was analyzed by 450/850K Illumina DNA methylation arrays. Genetic mutations were assessed by whole-exome or targeted sequencing, IGHV-D-J sequences by Sanger sequencing, and copy number alterations (CNAs) by Illumina arrays. The 1000 most variable CpG methylation sites were used for k-means clustering. Recursive feature elimination/random forest algorithms were used to develop a classifier for DNA methylation-based subgroups with 98% accuracy. Unsupervised clustering of 197 patients diagnosed with HCL, HCL-V or SDRPL revealed 5 distinct DNA methylation (M) subgroups ( Figure 1). Subgroup assignment was stable throughout longitudinal sampling (including pre/post-treatment) and consistent between splenic, bone marrow and PBMC derived cells. A subgroup with universally clonal BRAF-V600E mutations and majority diagnosed as HCL was termed the M-HCL subgroup ( Table 1). Four other groups termed M-SBLPN1-4 contained all HCL-V and SDRPL diagnosed samples and were devoid of BRAF-V600E mutations. M-SBLPN1 comprised MAP2K1 mutations (91%) and was enriched for CREBBP, ARIDIA and TERT-promoter mutations. These patients displayed an HCL-like immunophenotype (64.3% CD25+) with 1/3 diagnosed as HCL. M-SBLPN2 exhibited the highest prevalence of TP53 mutations and concomitant genomic instability. Patients in M-SBLPN1,2 were enriched in unmutated IGHV4-34 rearrangements. M-SBLPN3,4 subgroups displayed an immunophenotype more dissimilar to HCL, mutated IGHV genes, and enrichment of IGLL5, SYK and BIRC3 mutations. M-SBLPN4 contained the most SDRPL samples, suggesting it may represent the SDRPL entity retained by the WHO. We next uncovered that 29/170 SMZL patients displayed DNA methylation patterns mapping to M-SBLPN2-4. These patients were phenotypically and molecularly similar to SBLPN (70% displaying villous morphology and depleted in IGHV1-2*04, NOTCH2, KLF2 mutations), likely representing SMZL patients suggested for reassignment to SBLPN in the updated WHO classification. To elucidate molecular pathways governing the biology of M-SBLPN subgroups, transcription factor motif enrichment analysis in hypomethylated genomic regions revealed selective activation of AP-1 in M-HCL along with ETS in M-SBLPN1,2. Both transcription factors are downstream of MAPK signaling, consistent with activating BRAF and MAP2K1 mutations in these subgroups. However, we observed strong ETS enrichment in the absence of MAP2K1 in M-SBLPN along with mutual exclusivity of MAP2K1 and TP53 mutations, suggesting TP53 mutations are driving ETS activation. Although lymphoid neoplasms rarely exhibit TERT promoter mutations, 83% of M-SBLPN1 patients showed the c.-124C>T mutation commonly observed in other cancers producing an ETS binding site and ectopic TERT activation. ETS activation and gain of an ETS site by mutation implies oncogenesis involves aberrant TERT activation in this subgroup. In summary, we have developed a DNA methylation-based classifier that resolves 4 SBLPN subgroups with distinct molecular features, and reclassifies a subset of SMZL and HCL patients, adding further information to the updated WHO/ICC entities. We reveal distinct biological pathways operating in M-SBLPN subgroups that may aid targeted therapy approaches.

Development and validation of risk prediction models for large for gestational age infants using logistic regression and two machine learning algorithms.

Large for gestational age (LGA) is one of the adverse outcomes during pregnancy that endangers the life and health of mothers and offspring. We aimed to establish prediction models for LGA at late pregnancy. Data were obtained from an established Chinese pregnant women cohort of 1285 pregnant women. LGA was diagnosed as >90th percentile of birth weight distribution of Chinese corresponding to gestational age of the same-sex newborns. Women with gestational diabetes mellitus (GDM) were classified into three subtypes according to the indexes of insulin sensitivity and insulin secretion. Models were established by logistic regression and decision tree/random forest algorithms, and validated by the data. A total of 139 newborns were diagnosed as LGA after birth. The area under the curve (AUC) for the training set is 0.760 (95% confidence interval [CI] 0.706-0.815), and 0.748 (95% CI 0.659-0.837) for the internal validation set of the logistic regression model, which consisted of eight commonly used clinical indicators (including lipid profile) and GDM subtypes. For the prediction models established by the two machine learning algorithms, which included all the variables, the training set and the internal validation set had AUCs of 0.813 (95% CI 0.786-0.839) and 0.779 (95% CI 0.735-0.824) for the decision tree model, and 0.854 (95% CI 0.831-0.877) and 0.808 (95% CI 0.766-0.850) for the random forest model. We established and validated three LGA risk prediction models to screen out the pregnant women with high risk of LGA at the early stage of the third trimester, which showed good prediction power and could guide early prevention strategies.

Prediction of Heart Disease using Forest Algorithm over K-nearest neighbors using Machine Learning with Improved Accuracy

Aim: To perform Predicting heart disease using the Forest algorithm and comparing its feature extraction precision with the K-nearest neighbors algorithm for working on the precision of the forecast. Materials and Methods: In the proposed work, Predicting heart disease was carried out using machine learning algorithms such as K-nearest neighbors algorithm (n=10) and Forest Algorithm (n=10). Here the pretest power examination was done with gpower 80% and the sample size for the two gatherings was 20. Results: From The implemented experiment, the Forest algorithm accuracy is significantly better and it is 90.0% than the K-nearest neighbors algorithm 83.00%. There is a measurable 2-tailed significant distinction in exactness for two calculations is 0.001 (p<0.05) by performing Independent samples T-tests. Conclusion: The Forest algorithm got better Accuracy and classification of digits better than K-nearest neighbors algorithm for Predicting heart disease.

Prediction of Heart Disease using Forest Algorithm over Linear Regression Algorithm using Machine Learning With Improved Accuracy

Aim: To perform Predicting heart disease using the Forest algorithm and comparing its feature extraction precision with the Linear Regression Algorithm for improving the accuracy of the prediction. Methods and Materials: In the proposed work, Predicting heart disease was carried out using machine learning algorithms such as Linear Regression (n=10)and Forest algorithm(n=10). Here the pretest power analysis was carried out with 80% and the sample size for the two groups are 20. Results: From The implemented experiment, the Forest algorithm accuracy is 90.32% and the Linear Regression Algorithm 77.21%. There is a statistical 2-tailed significant difference in accuracy for two algorithms is 0.001 (p<0.05) Conclusion: This study concludes that the Forest algorithm on patients healthcare analysis is significantly better than the Linear Regression Algorithm.

Improving Daily Streamflow Forecasting Using Deep Belief Net-Work Based on Flow Regime Recognition

Streamflow forecasting is of great significance for water resources planning and management. In recent years, numerous data-driven models have been widely used for streamflow forecasting. However, the traditional single data-driven model ignores the utilization of different streamflow regimes. This study proposed an integrated framework for daily streamflow forecasting based on the regime recognition of flow sequences. The framework integrates self-organizing maps (SOM) for identifying streamflow sub-sequences, the random forests (RF) algorithm to select input variables for different streamflow sub-sequences, and a deep belief network (DBN) for establishing complex relationships between the selected input variables and streamflows for different sub-sequences. Specifically, the integrated framework was applied to forecast daily streamflow at the Xiantao hydrological station in the Hanjiang River Basin, China. The results show that the developed integrated framework has higher streamflow prediction accuracy than the single data-driven model (i.e., the DBN model in this study), with Nash efficiency coefficient (NSE) of 0.91/0.81 and coefficient of determination (R2) of 0.93/0.89 for the integrated framework/DBN model during the validation period, respectively. Additionally, the prediction accuracy of the peak flood was also improved. The relative error of the peak flood derived from the integrated framework was reduced by 4.6%, compared with the single DBN model. Overall, the constructed integration framework, considering the complex characteristic of different flow regimes, could improve the accuracy for daily streamflow forecasting.

A novel 6-metabolite signature for prediction of clinical outcomes in type 2 diabetic patients undergoing percutaneous coronary intervention

BackgroundOutcome prediction tools for patients with type 2 diabetes mellitus (T2DM) undergoing percutaneous coronary intervention (PCI) are lacking. Here, we developed a machine learning-based metabolite classifier for predicting 1-year major adverse cardiovascular events (MACEs) after PCI among patients with T2DM.MethodsSerum metabolomic profiling was performed in a nested case–control study of 108 matched pairs of patients with T2DM occurring and not occurring MACEs at 1 year after PCI, then the matched pairs were 1:1 assigned into the discovery and internal validation sets. External validation was conducted using targeted metabolite analyses in an independent prospective cohort of 301 patients with T2DM receiving PCI. The function of candidate metabolites was explored in high glucose-cultured human aortic smooth muscle cells (HASMCs).ResultsOverall, serum metabolome profiles differed between diabetic patients with and without 1-year MACEs after PCI. Through VSURF, a machine learning approach for feature selection, we identified the 6 most important metabolic predictors, which mainly targeted the nicotinamide adenine dinucleotide (NAD+) metabolism. The 6-metabolite model based on random forest and XGBoost algorithms yielded an area under the curve (AUC) of ≥ 0.90 for predicting MACEs in both discovery and internal validation sets. External validation of the 6-metabolite classifier also showed good accuracy in predicting MACEs (AUC 0.94, 95% CI 0.91–0.97) and target lesion failure (AUC 0.89, 95% CI 0.83–0.95). In vitro, there were significant impacts of altering NAD+ biosynthesis on bioenergetic profiles, inflammation and proliferation of HASMCs.ConclusionThe 6-metabolite model may help for noninvasive prediction of 1-year MACEs following PCI among patients with T2DM.

Road Accidents Prediction and Classification

Abstract: Road accidents leads to death, disability and hospitalization of people across world which leads to loss of potential income of individual and also affects the economy of the country. For every 10 people killed during road accidents across world one person belongs to India. In year 2020, total of 3,66,138 road accidents occurred leading to loss of 1,31,714 persons lives, injuring 3,48,279 persons. Number of road accidents and damage caused by it can be reduced by identifying the factors leading to it. In this project, we are applying the concepts of data mining and machine learning to identify the various factors that affect road accidents and its severity. The application will take variety inputs such as age of vehicle, light condition, road surface condition, speed limit etc. and will use random forest machine learning algorithm to calculate the severity of a possible accident. The severity of a possible accident will be displayed on a scale of 1 to 3, 1 being the highest and 3 being the least severe so, that they drive safely and take precautions. This data can be used in future to analyze inputs and improves the accuracy of the system output. In case of severity 1 which is the case of possibility of fatal accident an alert message will be sent to police so that they can take any preventive measures and therefore this application can prove to be very helpful in reducing accident fatality rates in the country.

Fabricated Components Hoisting Activity Recognition and Collision Analysis Based on Inertial Measurement Unit IMU

The prefabricated components are an important aspect of prefabricated constructions. Based on three possible collision situations when the fabricated component is being lifted during the installation phase, the aim of this study is to investigate how to prevent effective collision between the hoist and the barrier, and establish the standardized operation process (SOP) of using the Inertial Measurement Unit (IMU) to detect the hoisting collision of fabricated building components. The (IMU) technology is used in the collection of lifting activities data. The hoisting activity will be divided into four situations: classification and recognition of stationary, ascending, advancing, and descending of components; use of the K-Nearest Neighbor (KNN) algorithm and the Random Forest (RF) algorithm for data processing to recognize the hoisting activities of assembled components; construction of the hoisting activity recognition model; determination of the best recognition position of the IMU; further collision analysis based on the recognition of the hoisting of assembled components. The collision is divided into direct collision, sudden stop, and detour in a specific space from obstacles. Image analysis of the three types of collision activities will be carried out to help perceive lifting activities in advance and reduce loss resulting from collisions caused by components. According to the systematic research and discussion of hoisting activity recognition and collision behavior, it provides a reasonable basis and ideas for solving hoisting collisions in prefabricated buildings and aids in the use of inertial sensors in construction to provide assistance for construction automation.

P-052 Targeted metabolomics (amino acids) profiling of seminal plasma distinguished teratozoospermia from normozoospermic samples

Abstract Study question Is there a difference in metabolic composition between teratozoospermia and normozoospermic seminal samples? Summary answer Teratozoospermia and normozoospermia in humans are distinguished by seminal fluid metabolome markers especially amino acid content. What is known already Metabolomic profiling is the study of metabolites as downstream gene expression products in physiological and pathological circumstances. Identifying the metabolome of human seminal plasma (HSP) is a new research area to screen putative (in)fertility biomarkers. Knowing the semen metabolic signature and whether there are differences in metabolic composition between the teratozoospermia sample and semen with normal morphology could enhance our understanding of aspects associated with poor sperm quality and consequent male infertility. Study design, size, duration The Ethics Committee of Islamic Azad Tehran Medical Sciences University-Pharmacy and Pharmaceutical Branches approved the case-control study. Teratozoospermia was defined based on 5th WHO (2010) as normal morphology<4%, count>15*106/ml, progressive and non-progressive motility≥40%. In contrast, in normozoospermic samples from fertile men (who had at least one child), normal morphology was greater than 4%. HSP specimens were obtained from 15 infertile patients with teratozoospermia, and 12 confirmed fertile men as the control group. Participants/materials, setting, methods Fresh semen samples were obtained through masturbation following 3–4 days of abstention. A portion of each sample was used for routine semen analysis. Remained seminal plasma was isolated by centrifugation and stored; following post-thaw, HSPs were subjected to LC-MS-MSanalysis. MetaboAnalyst 5.0 software(www.metaboanalyst.ca) was utilized to identify fundamental metabolism alterations related to the morphology status of spermatozoa. In addition, different uni/multivariate analysis was performed to ascertain the group of metabolites responsible for the overall discrimination ability. Main results and the role of chance Seventeen significantly differentially metabolites (14 decreased and three increased) were observed in teratozoospermia samples vs. fertile control. Down-regulated metabolites include Cysteine, Gamma-aminobutyric acid, Glycine, Histidine, Hydroxylysine, Hydroxyproline, Methionine, Ornithine, Proline, Tryptophan, Aspartic acid, Argininosuccinic acid, Alpha-aminoadipic acid, and Beta-aminoisobutyric acid; whereas up. In contrast, up-regulatedlutamine, Asparagine, and Glycylproline. The PLS-DA score plot demonstrated the clear separation obtained according to the sperm morphological status. The two distinct groups were well-clustered, with distinct metabolic profiles for each group. The random forest (RF) algorithm was utilized to select relevant variables for teratozoospermia status classification by estimating the importance of each metabolite to teratozoospermia. The RF analysis defined a set of 15 metabolites that constitute the best predictors of teratozoospermia. In particular, increased Glutamine, Asparagine, and decreased Cysteine, Tryptophan, Glycine, and Valine were strong predictors of teratozoospermia. Interestingly, pathway enrichment analysis revealed that the most affected metabolic pathways in teratozoospermia men were the Aminoacyle-tRANA, arginine, valine-leucine, and isoleucine biosynthesis. Among 34 metabolites, Asparagine, Glutamine, and Glycylproline negatively correlated with sperm morphology status; whereas 21 metabolites were positively correlated with morphology, including Alanine, Argininesuccinic, Aspartic, Cystine, Gamma-aminobutyric acid, Glutamic, Glycine, Histidine, Homocitrulline, Hydroxylysine, Hydroxyproline, Isoleucine, Leucine, Methionine, Ornithine, Phenylalanine, Proline, Threonine, Tryptophan, Tyrosine, and Valine. Limitations, reasons for caution Since the strict criteria were considered for teratozoospermia patients, these results are preliminary and need to be confirmed with more sample size. Wider implications of the findings The possible cause of metabolite difference in seminal plasma of teratozoospermia patients comes from the disability of spermatozoa for selectively absorption or secretion of specific amino acids into/out of the cell, which is critical for sperm function and maturation and reflects the potential of fertility. Trial registration number Not Applicable

Comparative study on the performance of different classification algorithms, combined with pre- and post-processing techniques to handle imbalanced data, in the diagnosis of adult patients with familial hypercholesterolemia.

Familial Hypercholesterolemia (FH) is an inherited disorder of cholesterol metabolism. Current criteria for FH diagnosis, like Simon Broome (SB) criteria, lead to high false positive rates. The aim of this work was to explore alternative classification procedures for FH diagnosis, based on different biological and biochemical indicators. For this purpose, logistic regression (LR), naive Bayes classifier (NB), random forest (RF) and extreme gradient boosting (XGB) algorithms were combined with Synthetic Minority Oversampling Technique (SMOTE), or threshold adjustment by maximizing Youden index (YI), and compared. Data was tested through a 10 × 10 repeated k-fold cross validation design. The LR model presented an overall better performance, as assessed by the areas under the receiver operating characteristics (AUROC) and precision-recall (AUPRC) curves, and several operating characteristics (OC), regardless of the strategy to cope with class imbalance. When adopting either data processing technique, significantly higher accuracy (Acc), G-mean and F1 score values were found for all classification algorithms, compared to SB criteria (p < 0.01), revealing a more balanced predictive ability for both classes, and higher effectiveness in classifying FH patients. Adjustment of the cut-off values through pre or post-processing methods revealed a considerable gain in sensitivity (Sens) values (p < 0.01). Although the performance of pre and post-processing strategies was similar, SMOTE does not cause model’s parameters to loose interpretability. These results suggest a LR model combined with SMOTE can be an optimal approach to be used as a widespread screening tool.

An Ensemble Framework to Forest Optimization Based Reduct Searching

Essentially, the solution to an attribute reduction problem can be viewed as a reduct searching process. Currently, among various searching strategies, meta-heuristic searching has received extensive attention. As a new emerging meta-heuristic approach, the forest optimization algorithm (FOA) is introduced to the problem solving of attribute reduction in this study. To further improve the classification performance of selected attributes in reduct, an ensemble framework is also developed: firstly, multiple reducts are obtained by FOA and data perturbation, and the structure of those multiple reducts is symmetrical, which indicates that no order exists among those reducts; secondly, multiple reducts are used to execute voting classification over testing samples. Finally, comprehensive experiments on over 20 UCI datasets clearly validated the effectiveness of our framework: it is not only beneficial to output reducts with superior classification accuracies and classification stabilities but also suitable for data pre-processing with noise. This improvement work we have performed makes the FOA obtain better benefits in the data processing of life, health, medical and other fields.

Using Background Knowledge from Preceding Studies for Building a Random Forest Prediction Model: A Plasmode Simulation Study

There is an increasing interest in machine learning (ML) algorithms for predicting patient outcomes, as these methods are designed to automatically discover complex data patterns. For example, the random forest (RF) algorithm is designed to identify relevant predictor variables out of a large set of candidates. In addition, researchers may also use external information for variable selection to improve model interpretability and variable selection accuracy, thereby prediction quality. However, it is unclear to which extent, if at all, RF and ML methods may benefit from external information. In this paper, we examine the usefulness of external information from prior variable selection studies that used traditional statistical modeling approaches such as the Lasso, or suboptimal methods such as univariate selection. We conducted a plasmode simulation study based on subsampling a data set from a pharmacoepidemiologic study with nearly 200,000 individuals, two binary outcomes and 1152 candidate predictor (mainly sparse binary) variables. When the scope of candidate predictors was reduced based on external knowledge RF models achieved better calibration, that is, better agreement of predictions and observed outcome rates. However, prediction quality measured by cross-entropy, AUROC or the Brier score did not improve. We recommend appraising the methodological quality of studies that serve as an external information source for future prediction model development.

Machine Learning-Based Model for Prediction of Hemorrhage Transformation in Acute Ischemic Stroke After Alteplase.

Hemorrhage transformation (HT) is the most dreaded complication of intravenous thrombolysis (IVT) in acute ischemic stroke (AIS). The prediction of HT after IVT is important in the treatment decision-making for AIS. We designed and compared different machine learning methods, capable of predicting HT in AIS after IVT. A total of 345 AIS patients who received intravenous alteplase between January 2016 and June 2021 were enrolled in this retrospective study. The demographic characteristics, clinical condition, biochemical data, and neuroimaging variables were included for analysis. HT was confirmed by head computed tomography (CT) or magnetic resonance imaging (MRI) within 48 h after IVT. Based on the neuroimaging results, all of the patients were divided into the non-HT group and the HT group. Then, the variables were applied in logistic regression (LR) and random forest (RF) algorithms to establish HT prediction models. To evaluate the accuracy of the machine learning models, the models were compared to several of the common scales used in clinics, including the multicenter stroke survey (MSS) score, safe implementation of treatments in stroke (SITS) score, and SEDAN score. The performance of these prediction models was evaluated using the receiver operating characteristic (ROC) curve (AUC). Forty-five patients had HT (13.0%) within 48 h after IVT. The ROC curve results showed that the AUCs of HT that were predicted by the RF model, LR model, MSS, SITS, and SEDAN scales after IVT were 0.795 (95% CI, 0.647–0.944), 0.703 (95% CI, 0.515–0.892), 0.657 (95% CI, 0.574–0.741), 0.660 (95% CI, 0.580–0.740) and 0.655 (95% CI, 0.571–0.739), respectively. The RF model performed better than the other models and scales. The top four most influential factors in the RF importance matrix plot were triglyceride, Lpa, the baseline NIHSS, and hemoglobin. The SHapley Additive exPlanation values made the RF prediction model clinically interpretable. In this study, an RF machine learning method was successfully established to predict HT in AIS patients after intravenous alteplase, which the sensitivity was 66.7%, and the specificity was 80.7%.

CREDIT CARD FRAUD DETECTION USING ISOLATION FOREST AND LOCAL OUTLIER FACTOR

Nowadays, card fraud is increasing due to the prevalence of modern technology. Thus, Automatic systems to detect and prevent against card fraud are a significant tool in the financial industries battle against card crime. Machine learning and novelty detection techniques approaches are taken in Credit Card Fraud Detection field, since they are effective technologies and methodologies and easy to apply at the same time. It is used in order to reduce fraud activities. The main aim of this project was to create a program that detects and Identifies potentially fraudulent credit card transactions from a given data set, and evaluate its performance to be compared with other classifiers with evaluation metric. The program was trained with the given data set, based on the some of the most popular Machine Learning and Deep Learning Classification algorithms which are; Random Forest, Isolation Forest and Neural Networks Algorithm Load Balancing and Feature Selection were maintained throughout the project, and it was implemented using Python programming language. However, there were no autonomous system that will be able to categorically define a transaction as fraud. The objective was to highlight those transactions that have a high probability of being fraudulent based on some criteria, known or otherwise learnt. Key Words: isolation forest and local outlier factor, Pandas, Json, Matplotlib and seaborn.

Environmental Pollution Analysis and Impact Study—A Case Study for the Salton Sea in California

A natural experiment conducted on the shrinking Salton Sea, a saline lake in California, showed that each one foot drop in lake elevation resulted in a 2.6% average increase in PM2.5 concentrations. The shrinking has caused the asthma rate continues to increase among children, with one in five children being sent to the emergency department, which is related to asthma. In this paper, several data-driven machine learning (ML) models are developed for forecasting air quality and dust emission to study, evaluate and predict the impacts on human health due to the shrinkage of the sea, such as the Salton Sea. The paper presents an improved long short-term memory (LSTM) model to predict the hourly air quality (O3 and CO) based on air pollutants and weather data in the previous 5 h. According to our experiment results, the model generates a very good R2 score of 0.924 and 0.835 for O3 and CO, respectively. In addition, the paper proposes an ensemble model based on random forest (RF) and gradient boosting (GBoost) algorithms for forecasting hourly PM2.5 and PM10 using the air quality and weather data in the previous 5 h. Furthermore, the paper shares our research results for PM2.5 and PM10 prediction based on the proposed ensemble ML models using satellite remote sensing data. Daily PM2.5 and PM10 concentration maps in 2018 are created to display the regional air pollution density and severity. Finally, the paper reports Artificial Intelligence (AI) based research findings of measuring air pollution impact on asthma prevalence rate of local residents in the Salton Sea region. A stacked ensemble model based on support vector regression (SVR), elastic net regression (ENR), RF and GBoost is developed for asthma prediction with a good R2 score of 0.978.

817-P: Identifying Modulators of Treatment Benefits of Albiglutide (GLP-1 RA) in Preventing Major Adverse Cardiovascular Events (MACE) among Individuals with Type 2 Diabetes in the HARMONY Trial Using Machine Learning

This study aimed to identify patient characteristics associated with the effectiveness of albiglutide in preventing MACE among patients with type 2 diabetes (T2D) in the HARMONY trial. The HARMONY trial was a randomized, double-blind, placebo-controlled trial to examine the effectiveness of albiglutide in reducing the risk of cardiovascular disease in 9,463 individuals with T2D. We used the causal forest (CF) and causal tree (CT) algorithm to identify participant characteristics that modulate the effectiveness of albiglutide in preventing MACE from a list of 16 candidate variables (demographics, disease histories, medications, and biomarkers) at the study baseline. A computer-driven approach identified that study participants with baseline A1c level above 8.0% (absolute risk change (ARC) : -2.77%, 95% confidence interval [-4.21% to -1.28%]; relative risk (RR) : 0.74 [0.62 to 0.87]) , or diabetes duration less than 15 years (ARC: -2.55% [ -3.85% to -1.04%]; RR: 0.70 [0.58 to 0.85]) , or DBP higher than 70 mmHg (ARC: -2.40% [-3.68% to -1.21%]; RR: 0.73 [0.62 to 0.86]) , or heart rate below 70 per hour (ARC: -2.88% [-4.76% to -1.04%]; RR: 0.70 [0.56 to 0.88]) , or ALP level below 70 U/L (ARC: -3.10% [-3.10% to -1.30%]; RR: 0.65 [0.52 to 0.81]) , or serum creatinine level below 90 umol/L (ARC: -1.99% [-3.33% to -0.81%]; RR: 0.74 [0.61 to 0.88]) benefited from albiglutide in reducing the risk of MACE. In contrast, their counterparts who did not have these characteristics did not receive statistically significant MACE risk reduction from albiglutide. Our study identified clinical features that may result in greater benefits of albiglutide on MACE than what would be expected from a population average. Clinicians may consider these individualized characteristics for better-targeted albiglutide treatment. However, future GLP-1 trials may help to confirm this approach to optimize outcomes. Disclosure H.Kianmehr: None. Y.Lin: None. L.Shi: None. V.Fonseca: Consultant; Abbott, Asahi Kasei Corporation, Bayer AG, Novo Nordisk, Sanofi, Research Support; Fractyl Health, Inc., Jaguar Gene Therapy, Stock/Shareholder; Abbott, Amgen Inc., BRAVO4Health, Mellitus Health. J.Guo: None. H.Shao: Board Member; BRAVO4HEALTH, LLC.

174-OR: Identify Effect Modulators of Intensive Glucose Control in Preventing Cognitive Function Decline among Individuals with Type 2 Diabetes: A Revisit to the ACCORD MIND Trial

This study aimed to identify patient characteristics associated with the effectiveness of intensive glucose control (i.e., A1c&amp;lt;6.0% vs. A1c 7-8%) in preventing cognitive function decline (CFD) among individuals with type 2 diabetes (T2D) in the ACCORD trial. The Memory in Diabetes (MIND) study was sub trial of the ACCORD study, including 2,977 participants without a history of cognitive impairment or dementia. We applied the causal forest (CF) and causal tree (CT) algorithm to identify participant characteristics that modulate the effectiveness of intensive glucose control in preventing CFD from a list of 58 candidate variables (demographics, disease histories, medications, and biomarkers) at the study baseline. The study outcome was the change of cognitive function from the baseline to the 40th month, assessed through the Digit Symbol Substitution Test (DSST) , Rey Auditory Verbal Learning Test (RAVLT) , Mini-Mental State Examination (MMSE) , and Stroop test. A computer-driven approach identified a close relationship between renal function and the effectiveness of intensive glycemic control in preventing CFD: study participants with the urinary albumin less than 0.4 mg/dl (absolute function change (AFC) : 0.43 in MSSE) or urinary creatinine 120-200 mg/dl (AFC: 0.30 in RAVLT) had slower CFD associated with intensive glycemic control, while participants with GFR&amp;lt;60 (AFC:-0.51 in RVALT) , or GFR&amp;lt;90 (AFC:-1.7 in Stroop) had a faster CFD associated with intensive glycemic control. In addition, only participants with age below 60 years old benefited from the intensive glucose control (AFC:1.23 in DSST) in preventing CFD. All p&amp;lt;0.05. Our study tied renal function to the effectiveness of intensive glycemic control in preventing CFD. Clinicians may consider these individualized characteristics for better-targeted A1c goal setting, especially those at risk for CFD. Disclosure H.Kianmehr: None. J.Guo: None. V.Fonseca: Consultant; Abbott, Asahi Kasei Corporation, Bayer AG, Novo Nordisk, Sanofi, Research Support; Fractyl Health, Inc., Jaguar Gene Therapy, Stock/Shareholder; Abbott, Amgen Inc., BRAVO4Health, Mellitus Health. L.Shi: None. H.Shao: Board Member; BRAVO4HEALTH, LLC.

Talent management by predicting employee attrition using enhanced weighted forest optimization algorithm with improved random forest classifier

Talent management by predicting employee attrition using enhanced weighted forest optimization algorithm with improved random forest classifier

An Automatic Pronunciation Error Detection and Correction Mechanism in English Teaching Based on an Improved Random Forest Model

Teachers in traditional English classes focus more on writing and grammar instruction, while oral language instruction is neglected. In exam-oriented education, most Chinese students can master English written test skills, but only a few students can communicate effectively in English daily. People are progressively realizing that language is a tool for communication and communication in recent years, as the frequency of international exchanges has increased and that language learning should focus on oral language education. However, there are numerous issues with teaching oral English. When students perform individual oral practice after class, for example, they are unable to determine whether their pronunciation is correct. As a result, a computer-assisted study into automatic pronunciation of spoken English has become a viable solution to these issues. However, the present spoken English pronunciation mistake correction model’s accuracy and stability have not yet reached an optimal level. Based on this background, this work provides an enhanced random forest model and uses it to detect and correct automatic pronunciation errors in English classes. The improved random forest (RF) algorithm is used to classify and detect whether the learner’s pronunciation is correct. Mel cepstral coefficient (MFCC) is used for feature extraction, and principal component analysis (PCA) is used for dimensionality reduction of feature data. The experimental structure demonstrates that by using a combination classification framework based on MFCC, PCA, and RF, the learner’s pronunciation difficulty may be resolved. This allows for different error categories to receive feedback corrections.

Bottom-up estimates of reactive nitrogen loss from Chinese wheat production in 2014

Excessive use of synthetic nitrogen (N) for Chinese wheat production results in high loss of reactive N loss (Nr; all forms of N except N2) into the environment, causing serious environmental issues. Quantifying Nr loss and its spatial variations therein is vital to optimize N management and mitigate loss. However, accurate, high spatial resolution estimations of Nr from wheat production are lacking due to limitations of data generation and estimation methods. Here, we applied the random forest (RF) algorithm to bottom-up N application rate data, obtained through a survey of millions of farmers, to estimate the Nr loss from wheat production in 2014. The results showed that the average total Nr loss was 52.5 kg N ha−1 (range: 4.6-157.8 kg N ha−1), which accounts for 26.1% of the total N applied. The hotspots for high Nr loss are the same as those high applied N, including northwestern Xinjiang, central-southern Hebei, Shandong, central-northern Jiangsu, and Hubei. Our database could guide regional N management and be used in conjunction with biogeochemical models.