Gradient Boosting Tree Algorithm Research Articles

Correlation of gut microbiota with leukopenia after chemotherapy in patients with colorectal cancer

BackgroundThe most common toxic side effect after chemotherapy, one of the main treatments for colorectal cancer (CRC), is myelosuppression.ObjectiveTo analyze the correlation between gut microbiota and leukopenia after chemotherapy in CRC patients.MethodsStool samples were collected from 56 healthy individuals and 55 CRC patients. According to the leukocytes levels in peripheral blood, the CRC patients were divided into hypoleukocytes group (n = 13) and normal leukocytes group (n = 42). Shannon index, Simpson index, Ace index, Chao index and Coverage index were used to analyze the diversity of gut microbiota. LDA and Student's t-test(St test) were used for analysis of differences. Six machine learning algorithms, including logistic regression (LR) algorithm, random forest (RF) algorithm, neural network (NN) algorithm, support vector machine (SVM) algorithm, catboost algorithm and gradient boosting tree algorithm, were used to construct the prediction model of gut microbiota with leukopenia after chemotherapy for CRC.ResultsCompared with healthy group, the microbiota alpha diversity of CRC patients was significantly decreased (p < 0.05). After analyzing the gut microbiota differences of the two groups, 15 differential bacteria, such as Bacteroides, Faecalibacterium and Streptococcus, were screened. RF prediction model had the highest accuracy, and the gut microbiota with the highest predictive value were Peptostreptococcus, Faecalibacterium, and norank_f__Ruminococcaceae, respectively. Compared with normal leukocytes group, the microbiota alpha diversity of hypoleukocytes group was significantly decreased (p < 0.05). The proportion of Escherichia-Shigella was significantly decreased in the hypoleukocytes group. After analyzing the gut microbiota differences of the two groups, 9 differential bacteria, such as Escherichia-Shigella, Fusicatenibacter and Cetobacterium, were screened. RF prediction model had the highest accuracy, and the gut microbiota with the highest predictive value were Fusicatenibacte, Cetobacterium, and Paraeggerthella.ConclusionGut microbiota is related to leukopenia after chemotherapy. The gut microbiota may provide a novel method for predicting myelosuppression after chemotherapy in CRC patients.

Improving risk stratification of recurrent myocardial infarction in a large real-world dataset using machine learning

Abstract Background Patient (pt) risk stratification for cardiovascular (CV) events is key to determine the appropriate approach to secondary prevention. Several risk scores have been developed (SCORE2, EUROASPIRE, Thrombolysis in Myocardial Infarction [TIMI] risk score for secondary prevention [TRS 2°P]); these are based on few predictors and have low discriminatory power. Machine learning methods using real-word data (RWD) from electronic health records (EHR) databases and capable of handling a multitude of commonly available variables can serve as powerful tools to improve the discriminatory power and generate efficient and clinically relevant risk scores. Purpose This study used a non-linear machine learning model to build an improved risk score for secondary CV events from a large Israeli EHR database. Methods Records of pts with a first record of atherosclerotic CV disease (ASCVD) admission or coronary procedure and available follow-up data (mean: 1052 days) from the period of 01/2005–11/2020 were considered (n=77432 [67% coronary artery disease, 21% carotid artery disease, 12% peripheral artery disease], a stratified 1:1 randomization by cause of admission yielding in training set [n=38716] and test set [n=38716]). The model’s performance was assessed by comparison with TRS 2°P score on the test set independent of the training set. Twenty-six covariates commonly used in clinical routine and widely available in EHR have been selected and a gradient boosting trees algorithm was applied to the pt data at baseline after the original event to predict recurrent myocardial infarction (MI). The ability of the model to discriminate pts was evaluated using the concordance index (c-index), which expresses how well a predicted risk score describes the observed events (1 implies perfect concordance, 0.5 implies random concordance). Results Prediction accuracy of our model showed a c-index for a future MI of 0.74 (95% confidence interval [CI]: 0.72, 0.76) which is significantly higher than 0.57 (95% CI: 0.56, 0.58) obtained with TRS 2°P. Our 2-tier risk score has successfully discriminated a low-risk group (74% of cohort) with a 3-year MI prevalence of 5% from a high-risk group (26% of cohort) that had a 5-fold higher prevalence (26.6%) (Figure 1). In comparison, the 3-tier TRS 2°P yielded an unbalanced distribution of group sizes with a disproportionate intermediate risk group (22% low, 67% intermediate, 11% of high-risk) with a limited clinical discriminatory power (7.1%, 10,2% and 15.6% of observed events, respectively) (Figure 2). Conclusions Our machine learning based risk score model created on a large set of RWD with a long follow-up seems to offer substantial gain of performance compared to the TRS 2°P. The present 2-tier model shows significant gain of accuracy and clinical discriminatory power paving the way for more individualized prevention strategies.Figure 1Figure 2

A physics-guided eXtreme gradient boosting model for predicting the initial productivity of oil wells

The initial productivity of oil wells is an important index in oilfield development. Due to the nonlinear relationship among the numerous affecting factors of productivity, productivity prediction methods based on seepage theory are difficult to characterize this nonlinear relationship. The machine learning method provides a novel theory to solve this problem. However, the data-driven machine learning method neglects the physical-based seepage theory. For balancing the advantage of the seepage model and the data-driven model, the productivity formula was used to guide the machine learning algorithm for achieving nonlinear regression in this study. Therefore, an innovative physics-guided eXtreme Gradient Boosting (XGBoost) trees algorithm was proposed to predict the initial productivity. The productivity formula was employed as the physical constraint and combined with the mean-square error (MSE) loss function of the XGBoost algorithm. A hyperparameter was introduced to balance the productivity formula and the MSE formula for constructing this adaptive physics-guided loss function. Moreover, the random subsampling cross-validation approach was adopted to train the XGBoost model. The grid search approach was introduced to optimize the hyperparameters of the model. The results show that based on the numerical simulator data, the initial productivity prediction accuracy of physics-guided XGBoost was verified as 94.67%, which was higher than the completely data-driven model. Moreover, the robustness of the novel adaptive physics-guided loss function for improving forecasting accuracy had been validated with sample sizes between 50 and 20000. By introducing the physics-guided sampling process and loss function, the applicability of the XGBoost algorithm has been extended to small sample conditions. Furthermore, the practical accuracy of physics-guided XGBoost was verified as 84.74% by 162 oil wells in the Penglai oilfield. Therefore, the application value of combining the physical knowledge with the XGBoost model has been confirmed in oilfield situations.

Nonmodel rapid seismic assessment of eccentrically braced frames incorporating masonry infills using machine learning techniques

This study investigates the seismic behavior of eccentrically braced frames (EBFs) taking into account the influence of masonry infill walls using a nonmodel scenario-based machine learning framework. Predicting engineering demand parameters (EDPs) by exploiting data-driven approaches and developing the associated fragility curves of infilled EBFs are the main objectives of this research. To this end, a set of 4- and 8-story archetype EBF structures considering 12 distinct infill properties, a total number of 4 bare and 48 infilled EBF models, are created in OpenSees software. A nonlinear pushover analysis is then conducted to assess the overall impact of infills on 4- and 8-story EBFs. Incremental dynamic analysis under 44 far-field ground motions is performed to collect an extensive raw database consisting of 37,164 and 461,480 data points for each variable of bare and infilled EBFs, respectively. It encompasses seismic intensity measure (Saavg), wavelet-based robust damage sensitive feature (rDSF), derived from roof absolute acceleration, frame geometric information, and infill parameter as input features and also EDPs containing peak and residual story drift ratios and peak floor accelerations (PFA) as response variables. After data preprocessing, both regression analysis and machine learning (ML) techniques are utilized to estimate EDPs under two scenarios of Input-Output and Output-Only based on the availability of Saavg. Comparing error measures, particularly in the Extreme Gradient Boosted Tree (ExGBT) algorithm, reveals strong observed-predicted EDP compatibility. Linear equations from bare frame data predict infilled frame EDPs without modification coefficients. Data preprocessing demonstrates that infills decrease drift-based EDPs due to stiffness improvement, while their impact on PFA remains inconclusive. Saavg and rDSF-based fragility curves at three damage states show accurate predictions and effectively reduce vulnerability in infilled 4-story EBFs. However, infills undesirably affect non-ductile behavior in 8-story EBFs, leading to increased global damage as supported by pushover analysis.

Early prediction of body composition parameters on metabolically unhealthy in the Chinese population via advanced machine learning.

Metabolic syndrome (Mets) is considered a global epidemic of the 21st century, predisposing to cardiometabolic diseases. This study aims to describe and compare the body composition profiles between metabolic healthy (MH) and metabolic unhealthy (MU) phenotype in normal and obesity population in China, and to explore the predictive ability of body composition indices to distinguish MU by generating machine learning algorithms. A cross-sectional study was conducted and the subjects who came to the hospital to receive a health examination were enrolled. Body composition was assessed using bioelectrical impedance analyser. A model generator with a gradient-boosting tree algorithm (LightGBM) combined with the SHapley Additive exPlanations method was adapted to train and interpret the model. Receiver-operating characteristic curves were used to analyze the predictive value. We found the significant difference in body composition parameters between the metabolic healthy normal weight (MHNW), metabolic healthy obesity (MHO), metabolic unhealthy normal weight (MUNW) and metabolic unhealthy obesity (MUO) individuals, especially among the MHNW, MUNW and MUO phenotype. MHNW phenotype had significantly lower whole fat mass (FM), trunk FM and trunk free fat mass (FFM), and had significantly lower visceral fat areas compared to MUNW and MUO phenotype, respectively. The bioimpedance phase angle, waist-hip ratio (WHR) and free fat mass index (FFMI) were found to be remarkably lower in MHNW than in MUNW and MUO groups, and lower in MHO than in MUO group. For predictive analysis, the LightGBM-based model identified 32 status-predicting features for MUNW with MHNW group as the reference, MUO with MHO as the reference and MUO with MHNW as the reference, achieved high discriminative power, with area under the curve (AUC) values of 0.842 [0.658, 1.000] for MUNW vs. MHNW, 0.746 [0.599, 0.893] for MUO vs. MHO and 0.968 [0.968, 1.000] for MUO and MHNW, respectively. A 2-variable model was developed for more practical clinical applications. WHR > 0.92 and FFMI > 18.5 kg/m2 predict the increased risk of MU. Body composition measurement and validation of this model could be a valuable approach for the early management and prevention of MU, whether in obese or normal population.

Concurrent prediction of metallic glasses’ global energy and internal structural heterogeneity by interpretable machine learning

Predicting glasses’ properties from their structures is a formidable challenge because of the inherently disordered atomic configurations. Here we tackle the problem using a new two-stage (encoding/interpreting) machine learning pipeline. First, local environments are encoded by the Smooth Overlap of Atomic Positions (SOAP) descriptors, which are then fed into extreme gradient boosting tree algorithm to train/predict given samples’ configurational energy. 40 important unique local environments (ULEs) most responsible for the global energy of ZrCu-based glasses are identified. Markedly, we discover that the same short-range orders of Voronoi cells, when embedded in various ULEs, could impact the sample's global stability in qualitatively different manners. These new findings thus reveal a profound connection between short-range orders and medium-range orders. In the second stage, a designed interpreting stage is employed to decompose a sample's 3 N degrees-of-freedom configuration into a 40-dimension probability spectrum barcode via frequency mapping of those ULEs. We demonstrate that, in addition to the global energy prediction, by analyzing barcode-elements’ occupational fractions and fluctuations, one can simultaneously assess samples’ structural heterogeneity, which is known as a crucial quantity to dictate metallic glasses’ deformation behaviors. The implications of our findings to a barcode-mediated new strategy of inverse engineering design of metallic glasses are also discussed.

Research on aircraft automation warning model based on gradient boosting tree algorithm

Flight safety is the basis for the survival and development of civil aviation transportation industry, and with the rapid development of China's civil aviation industry, the study of aircraft safety issues is becoming more and more important. Serious flight accidents can lead to great threat to life as well as economic loss, so this paper focuses on flight safety issues, and analyzes and investigates flight key parameters segment data, aircraft overruns, and flight parameters data. In this paper, based on the relevant data from QAR , t-test was performed on the processed data through data analysis and data cleaning to ensure the reliability of the data. And the principal component components of the key data were extracted using the random forest algorithm, and the 6-bit eigenvalues were selected as the key parameters, which had the greatest influence on the landing G value of the aircraft. Then, the data set was divided according to the ratio of 8:2, and the pilot qualification was used as the label value, and the machine learning classification prediction was performed by the gradient boosting tree GBDT algorithm to obtain the pilot qualification prediction model, and the highest accuracy reached 76.3% in the test set. This paper also compares the extreme gradient boosting XGBoost algorithm, with the highest current parameter reaching 73.4%. In this paper, the gradient boosting tree GBDT is chosen to construct a flight crew qualification assessment model.

Guidance on modeling circulating miRNA to distinguish multiple cancer types by an observation of large-scale open data.

e13537 Background: Cell-free miRNAs (cf-miRNA), circulated in body fluids such as plasma or serum, have shown their ability to detect, diagnose, and monitor cancers. Combining machine learning (ML) technology with these biomarkers facilitates early detection of cancers, which increases the accuracy of clinical decisions and empowers people to take control of their health status. However, the data of the cf-miRNAs has characteristics, which will affect the results of ML. Therefore, this study tries to expound on them in different aspects and to build a reasonable model. Methods: We downloaded large-scale datasets of the platform GPL21263 from the Gene Expression Omnibus for modeling experiments. We curated 8,174 subjects with 2,565 miRNA targets across 7 cancer types of different cf-miRNA-based cancer studies. Moreover, we used principal component analysis (PCA) to observe the datasets, recursive feature elimination (RFE) for feature selection, and tree-based algorithms to build the prediction model. Results: The characteristics of the cf-miRNA we like to share are: (1) Cancer subjects express more cf-miRNAs than control subjects. In the control group, there were 294 and 327 miRNAs with missing rates under 50% and 25%, respectively. In contrast, there were 395 and 485 miRNAs with the same thresholds in the cancer group; (2) Dividing subjects into cancer and control groups is simpler than distinguishing specific cancer types. In the PCA, the average Euclidean distance between the control group and each cancer type is 98.48, while it is 20.23 within each cancer type; (3) For obtaining cancer-specific biomarkers, we suggested that other non-target cancer subjects should be considered as negative controls. We modeled 7 cancer types and compared the proportion of cancer-specific biomarkers, not selected by any other models. The proportion increased from 30.0% to 57.8% after we added other non-target cancer subjects to the control group; next, we focus on multi-cancer modeling: (4) We need at least 400 samples to distinguish seven cancer types. In our experiment, we kept increasing the size of the training data, a hundred at a time. As data was added to the model, the accuracy increased but plateaued after adding 400 samples; (5) Based on RFE, 120 miRNAs is a reasonable number to distinguish seven cancers. Moreover, we found some of these miRNAs are only expressed in cancer subjects. We might lose this kind of biomarkers if we filtered out them by the missing rate; (6) The 10-fold cross validation accuracy of the multi-cancer model can achieve 93.0% using the gradient-boosted trees algorithm. Conclusions: In this study, we showed the guidance for modeling miRNAs in different aspects, including labeling strategy, sample and feature sizes, and the high-accuracy multi-cancer model we can achieve. We hope this guidance would inspire researchers on cf-miRNA-related machine learning applications.

Classification of the immune microenvironment associated with 12 cell death modes and construction of a prognostic model for squamous cell lung cancer.

An increasing number of patients with lung squamous cell carcinoma (LUSC) are benefiting from immunotherapy. However, the individual immune profile of patients who respond to treatment is unclear. Multiple programmed cell death (PCD) patterns play an important role in the proliferation and differentiation of tumor cells, predicting the efficacy of immunotherapy using a risk model for programmed cell death gene combinations LUSC risk model. Genes associated with 12 types of PCD were analyzed to establish a prognostic model. Risk scores were calculated using PCDG-based expression profiles, and LUSC patients were classified into two groups. Tumor immune microenvironment (TIME) characteristics and immunotherapy responses were compared between the two groups. Finally, staging was predicted using the extreme gradient boosting tree algorithm (eXtreme Gradient Boosting, XGBoost), and an algorithmic model was constructed to predict the prognosis of LUSC patients based on the PCDG risk score. A stepwise downscaling of 1256 PCDGs was performed to screen out 16 genes associated with LUSC prognosis to construct a risk model. Immune cell infiltration levels, the immunotherapy response, and prognostic differences were different between these two groups of patients. The classification prediction model based on the XGBoost algorithm and the prognostic model based on the risk score were able to distinguish the risk subtypes and individual prognosis of LUSC patients, respectively. PCD patterns exert a crucial effect on the development of LUSC. An evaluation of different PCD patterns in LUSC improves the understanding of the characteristics of infiltrating immune cells and mutational features of the TIME, distinguishes LUSC patients who might benefit from immunotherapy, and predicts their future survival.

Big Data Analytics Using Machine Learning Techniques for Prediction on Datasets

Data analytics is the process of performing scientific and statistical analysis on raw data in order to transform it into information that can be used for gaining knowledge. A recently emerging trend in feature abstraction is the combination of computational techniques and big data analysis. This requires gaining knowledge from trustworthy data sources, being able to digest information quickly, and making accurate predictions about the future. The primary objective of this study is to locate the machine learning strategies that produce the most accurate prediction by utilising the model that has been proposed. The supervised and unsupervised strategies have been implemented in a variety of different ways using the MapReduce methodology; however, the suggested model makes use of the Apache Spark framework in order to compare the many existing methods. In this study, the emphasis is placed on elucidating the characteristics of datasets in order to conduct the most accurate analysis possible using machine learning techniques. For the purpose of conducting an analysis of the data sets, machine learning methods such as linear regression, decision trees, random forests, and gradient boosting tree algorithms are utilised. In light of the findings of this research, it is possible to draw the conclusion that when the Spark framework is applied on top of Machine Learning methods, the efficiency of the model is improved by a factor of seventy percent in comparison to the MapReduce paradigm. Keyword : Apache Spark Framework, Big Data Analytics, Machine Learning Algorithms, MapReduce Paradigm.

GB-score: Minimally designed machine learning scoring function based on distance-weighted interatomic contact features.

In recent years, thanks to advances in computer hardware and dataset availability, data-driven approaches (like machine learning) have become one of the essential parts of the drug design framework to accelerate drug discovery procedures. Constructing a new scoring function, a function that can predict the binding score for a generated protein-ligand pose during docking procedure or a crystal complex, based on machine and deep learning has become an active research area in computer-aided drug design. GB-Score is a state-of-the-art machine learning-based scoring function that utilizes distance-weighted interatomic contact features, PDBbind-v2019 general set, and Gradient Boosting Trees algorithm to the binding affinity prediction. The distance-weighted interatomic contact featurization method used the distance between different ligand and protein atom types for numerical representation of the protein-ligand complex. GB-Score attains Pearson's correlation 0.862 and RMSE 1.190 on the CASF-2016 benchmark test in the scoring power metric. GB-Score's codes are freely available on the web at https://github.com/miladrayka/GB_Score.

The role of advanced machine learning approach in predicting multiple sclerosis development and progression

This study aims to utilize a fine-tuned gradient boosting trees algorithm to predict the onset and progression of Multiple Sclerosis (MS) based on a comprehensive set of demographic and clinical variables. The goal was to enhance early diagnosis and enable individualized treatment approaches using artificial intelligence. The research utilized Dataset, a publicly accessible dataset derived from a prospective cohort study of individuals of Mexican mestizo descent diagnosed with Clinically Isolated Syndrome (CIS). The study spanned from 2006 to 2010, systematically collecting and analyzing data on various individual traits to explore correlations with MS development. The gradient boosting trees algorithm was employed to construct predictive models, harnessing patient-specific variables, including demographic factors and clinical data. The classifier exhibited outstanding performance, with a mean accuracy of 99.63% and minimal standard deviation. The confusion matrix indicated one false positive and no false negatives. Key metrics such as precision, recall, and AUC all approached 1, demonstrating the classifier's ability to distinguish between the two classes with high confidence. Comparative analysis with similar studies in the literature revealed superior performance, highlighting the classifier's accuracy and effectiveness in predicting MS. The application of the gradient boosting trees algorithm to predict MS based on demographic and clinical variables offers a promising avenue for early diagnosis and tailored treatment. This research demonstrates the potential of AI to transform healthcare, particularly in the context of MS. The predictive models developed have the capacity to enhance early detection, improve patient quality of life, and pave the way for further AI-based solutions in healthcare.

Finding undiagnosed patients with hepatitis C virus: an application of machine learning to US ambulatory electronic medical records

AimsTo develop and validate a machine learning (ML) algorithm to identify undiagnosed hepatitis C virus (HCV) patients, in order to facilitate prioritisation of patients for targeted HCV screening.MethodsThis retrospective study...

Optimal mode of delivery in pregnancy: Individualized predictions using national vital statistics data.

Child birth via Cesarean section accounts for approximately 32% of all births each year in the United States. A variety of risk factors and complications can lead caregivers and patients to plan for a Cesarean delivery in advance before onset of labor. However, a non-trivial subset of Cesarean sections (∼25%) are unplanned and occur after an initial trial of labor is attempted. Unfortunately, patients who deliver via unplanned Cesarean sections have increased maternal morbidity and mortality rates and higher rates of neonatal intensive care admissions. In an effort to develop models aimed at improving health outcomes in labor and delivery, this work seeks to explore the use of national vital statistics data to quantify the likelihood of an unplanned Cesarean section based on 22 maternal characteristics. Machine learning techniques are used to ascertain influential features, train and evaluate models, and assess accuracy against available test data. Based on cross-validation results from a large training cohort (n = 6,530,467 births), the gradient-boosted tree algorithm was identified as the best performer and was evaluated on a large test cohort (n = 10,613,877 births) for two prediction scenarios. Area under the receiver operating characteristic curves of 0.77 or higher and recall scores of 0.78 or higher were obtained and the resulting models are well calibrated. Combined with feature importance analysis to explain why certain maternal characteristics lead to a specific prediction in individual patients, the developed analysis pipeline provides additional quantitative information to aid in the decision process on whether to plan for a Cesarean section in advance, a substantially safer option among women at a high risk of unplanned Cesarean delivery during labor.

Analysis and Prediction of Telecom Customer Churn based on Machine Learning

As the telecommunications market becomes increasingly saturated, major operators are facing an increasingly severe problem of soaring customer churn rates. How to identify high-risk churn customers is the most concerned issue for operators. Thanks to the rapid development of pattern recognition technology, existing machine learning algorithms provide key technical support for telecom customer churn prediction. However, how to choose an appropriate forecasting method combined with the characteristics of the application data is still an open question. To this end, based on the analysis and comparison of the feature correlation between telecom customer data and churn, this paper compares the differences in the prediction results of different machine algorithms, so as to choose the method that best fits the characteristics of the application data to build the final customer churn prediction model. Specifically, the Spearman correlation coefficient is used to calculate the correlation between variables in the dataset, the random forest algorithm is used to score the importance of all variables, and the prediction generated by the gradient boosting tree algorithm is introduced. Finally, the gradient boosting tree algorithm is evaluated by five performance indicators: precision rate, recall rate, precision rate, F1 score and AUC (Area under the ROC curve).

Abstract 9842: Machine Learning-Based Prediction Model of COVID-19 Deterioration Using Wearable Core Temperature and Oxygen Saturation Sensor Data

Background: Accurate prediction of COVID-19 deterioration could enable clinicians to make optimal decisions on medical interventions or hospitalization. A machine learning-based approach using vital data is recognized as a useful tool to predict the exacerbation of infectious disease. However, in COVID-19, effective prediction models are still emerging. Here we developed an algorithm to predict clinical deterioration in non-severe COVID-19 patients using machine learning techniques with continuous core temperature (CT), SpO 2 and pulse rate (PR) data obtained via wearable sensors. Methods: Fifty mild to moderate COVID-19 patients admitted to Nishi-Kobe Medical Center between December 2020 and August 2021 were enrolled. Disease severity was judged by WHO criteria. All patients wore a Moni-Patch ®️ , a wireless CT monitoring patch and a ring-type SpO 2 monitoring device. CT, SpO 2 and PR were continuously collected to a smartphone via Bluetooth ®️ . Data from 41 patients were of sufficient duration and quality to be analyzed and preprocessed to adjust sampling windows before fed into the framework. Additional features were derived by modifying the sensor data using patients’ characteristics. To develop classifiers to distinguish a disease deteriorating into severe one, XGBoost, a gradient-boosted tree algorithm, with 3-fold cross-validation was used. Based on the algorithm, deterioration into severe disease within 24 hour was inferred every single hour using 12-hour time frame data. The area under the curve (AUC) of receiver operating characteristics was analyzed based on the classifier’s prediction threshold. Results: After the enrollment, 13 out of 41 patients developed severe diseases. At the enrollment, baseline characteristics were comparable between mild to moderate and severe disease, except for body temperature that was significantly higher in severe disease. The algorithm developed using the sensor data was able to achieve an AUC of 0.92. Conclusions: Our algorithm could well distinguished COVID-19 deteriorating into severe disease. The results indicates that machine learning approach with continuous CT, SpO 2 and PR measurement using wearable sensors is beneficial to predict the exacerbation of COVID-19.

Predicting electricity infrastructure induced wildfire risk in California

This paper examines the use of risk models to predict the timing and location of wildfires caused by electricity infrastructure. Our data include historical ignition and wire-down points triggered by grid infrastructure collected between 2015–2019 in Pacific Gas & Electricity territory along with various weather, vegetation, and very high resolution data on grid infrastructure including location, age, and materials. With these data we explore a range of machine learning methods and strategies to manage training data imbalance. The best area under the receiver operating characteristic we obtain is 0.776 for distribution feeder ignitions and 0.824 for transmission line wire-down events, both using the histogram-based gradient boosting tree algorithm with under-sampling. We then use these models to identify which information provides the most predictive value. After line length, we find that weather and vegetation features dominate the list of top important features for ignition or wire-down risk. Distribution ignition models show more dependence on slow-varying vegetation variables such as burn index, energy release content, and tree height, whereas transmission wire-down models rely more on primary weather variables such as wind speed and precipitation. These results point to the importance of improved vegetation modeling for feeder ignition risk models, and improved weather forecasting for transmission wire-down models. We observe that infrastructure features make small but meaningful improvements to risk model predictive power.

Predicting the Risk of Overweight and Obesity in Madrid—A Binary Classification Approach with Evolutionary Feature Selection

In this paper, we experimented with a set of machine-learning classifiers for predicting the risk of a person being overweight or obese, taking into account his/her dietary habits and socioeconomic information. We investigate with ten different machine-learning algorithms combined with four feature-selection strategies (two evolutionary feature-selection methods, one feature selection from the literature, and no feature selection). We tackle the problem under a binary classification approach with evolutionary feature selection. In particular, we use a genetic algorithm to select the set of variables (features) that optimize the accuracy of the classifiers. As an additional contribution, we designed a variant of the Stud GA, a particular structure of the selection operator of individuals where a reduced set of elitist solutions dominate the process. The genetic algorithm uses a direct binary encoding, allowing a more efficient evaluation of the individuals. We use a dataset with information from more than 1170 people in the Spanish Region of Madrid. Both evolutionary and classical feature-selection methods were successfully applied to Gradient Boosting and Decision Tree algorithms, reaching values up to 79% and increasing the average accuracy by two points, respectively.

Academic Performance Evaluation Using Data Mining in Times of Pandemic

This work focuses on studying the relationship that existed between the use of the learning management system (LMS) and the academic performance of the students of the Jorge Basadre Grohmann National University of Tacna-Perú. For this, we use the data provided by the LMS (access virtual classroom) and the university's academic management system (grades). For that, we perform various classification machine learning algorithms to predict academic performance with two classes SATISFACTORY or POOR where Gradient Boosted Trees algorithm had the best accuracy 91.79%. However, with three classes, SATISFACTORY, REGULAR AND POOR, Random Forest algorithm had the best accuracy of 89.26%.

Prediction of Admission Costs Following Anterior Cervical Discectomy and Fusion Utilizing Machine Learning.

Retrospective case series. Predict cost following anterior cervical discectomy and fusion (ACDF) within the 90-day global period using machine learning models. The incidence of ACDF has been increasing with a disproportionate decrease in reimbursement. As bundled payment models become common, it is imperative to identify factors that impact the cost of care. The Nationwide Readmissions Database (NRD) was accessed in 2018 for all primary ACDFs by the International Classification of Diseases 10th Revision (ICD-10) procedure codes. Costs were calculated by utilizing the total hospital charge and each hospital's cost-to-charge ratio. Hospital characteristics, such as volume of procedures performed and wage index, were also queried. Readmissions within 90 days were identified, and cost of readmissions was added to the total admission cost to represent the 90-day healthcare cost. Machine learning algorithms were used to predict patients with 90-day admission costs >1SD from the mean. There were 42,485 procedures included in this investigation with an average age of 57.7±12.3 years with 50.6% males. The average cost of the operative admission was $24,874±25,610, the average cost of readmission was $25,371±11,476, and the average total cost was $26,977±28,947 including readmissions costs. There were 10,624 patients who were categorized as high cost. Wage index, hospital volume, age, and diagnosis-related group severity were most correlated with the total cost of care. Gradient boosting trees algorithm was most predictive of the total cost of care (area under the curve=0.86). Bundled payment models utilize wage index and diagnosis-related groups to determine reimbursement of ACDF. However, machine learning algorithms identified additional variables, such as hospital volume, readmission, and patient age, that are also important for determining the cost of care. Machine learning can improve cost-effectiveness and reduce the financial burden placed upon physicians and hospitals by implementing patient-specific reimbursement.