Machine Learning Methods Applied to the Prediction of Pseudo-nitzschia spp. Blooms in the Galician Rias Baixas (NW Spain)

This work presents new prediction models based on recent developments in machine learning methods, such as Random Forest (RF) and AdaBoost, and compares them with more classical approaches, i.e., support vector machines (SVMs) and neural networks (NNs). The models predict Pseudo-nitzschia spp. blooms in the Galician Rias Baixas. This work builds on a previous study by the authors (doi.org/10.1016/j.pocean.2014.03.003) but uses an extended database (from 2002 to 2012) and new algorithms. Our results show that RF and AdaBoost provide better prediction results compared to SVMs and NNs, as they show improved performance metrics and a better balance between sensitivity and specificity. Classical machine learning approaches show higher sensitivities, but at a cost of lower specificity and higher percentages of false alarms (lower precision). These results seem to indicate a greater adaptation of new algorithms (RF and AdaBoost) to unbalanced datasets. Our models could be operationally implemented to establish a short-term prediction system.

Download Full-text

Classification models using circulating neutrophil transcripts can detect unruptured intracranial aneurysm

Journal of Translational Medicine ◽

10.1186/s12967-020-02550-2 ◽

2020 ◽

Vol 18 (1) ◽

Author(s):

Kerry E. Poppenberg ◽

Vincent M. Tutino ◽

Lu Li ◽

Muhammad Waqas ◽

Armond June ◽

...

Keyword(s):

Machine Learning ◽

Random Forest ◽

Prediction Models ◽

Model Performance ◽

Supervised Machine Learning ◽

Support Vector ◽

Learning Methods ◽

Training Cohort ◽

Network Analyses ◽

Machine Learning Methods

Abstract Background Intracranial aneurysms (IAs) are dangerous because of their potential to rupture. We previously found significant RNA expression differences in circulating neutrophils between patients with and without unruptured IAs and trained machine learning models to predict presence of IA using 40 neutrophil transcriptomes. Here, we aim to develop a predictive model for unruptured IA using neutrophil transcriptomes from a larger population and more robust machine learning methods. Methods Neutrophil RNA extracted from the blood of 134 patients (55 with IA, 79 IA-free controls) was subjected to next-generation RNA sequencing. In a randomly-selected training cohort (n = 94), the Least Absolute Shrinkage and Selection Operator (LASSO) selected transcripts, from which we constructed prediction models via 4 well-established supervised machine-learning algorithms (K-Nearest Neighbors, Random Forest, and Support Vector Machines with Gaussian and cubic kernels). We tested the models in the remaining samples (n = 40) and assessed model performance by receiver-operating-characteristic (ROC) curves. Real-time quantitative polymerase chain reaction (RT-qPCR) of 9 IA-associated genes was used to verify gene expression in a subset of 49 neutrophil RNA samples. We also examined the potential influence of demographics and comorbidities on model prediction. Results Feature selection using LASSO in the training cohort identified 37 IA-associated transcripts. Models trained using these transcripts had a maximum accuracy of 90% in the testing cohort. The testing performance across all methods had an average area under ROC curve (AUC) = 0.97, an improvement over our previous models. The Random Forest model performed best across both training and testing cohorts. RT-qPCR confirmed expression differences in 7 of 9 genes tested. Gene ontology and IPA network analyses performed on the 37 model genes reflected dysregulated inflammation, cell signaling, and apoptosis processes. In our data, demographics and comorbidities did not affect model performance. Conclusions We improved upon our previous IA prediction models based on circulating neutrophil transcriptomes by increasing sample size and by implementing LASSO and more robust machine learning methods. Future studies are needed to validate these models in larger cohorts and further investigate effect of covariates.

Download Full-text

Development of a Probabilistic Seismic Performance Assessment Model of Slope Using Machine Learning Methods

Sustainability ◽

10.3390/su12083269 ◽

2020 ◽

Vol 12 (8) ◽

pp. 3269

Author(s):

Shinyoung Kwag ◽

Daegi Hahm ◽

Minkyu Kim ◽

Seunghyun Eem

Keyword(s):

Machine Learning ◽

Regression Analysis ◽

Linear Regression ◽

Seismic Performance ◽

Prediction Models ◽

Linear Regression Analysis ◽

Assessment Model ◽

Support Vector ◽

Learning Methods ◽

Machine Learning Methods

The objective of this study is to propose a model that can predict the seismic performance of slope relatively accurately and efficiently by using machine learning methods. Probabilistic seismic fragility analyses of the slope had been carried out in other studies, and a closed-form equation for slope seismic performance was proposed through a multiple linear regression analysis. However, the traditional statistical linear regression analysis showed a limit that could not accurately represent such nonlinear slope seismic performances. To overcome this limit, in this study, we used three machine learning methods (i.e., support vector machine (SVM), artificial neural network (ANN), Gaussian process regression (GPR)) to generate prediction models of the slope seismic performance. The models obtained through the machine learning methods basically showed better performance compared to the models of the traditional statistical methods. The results of the SVM showed no significant performance difference compared with the results of the nonlinear regression analysis method, but the results based on the ANN and GPR showed a remarkable improvement in the prediction performance over the other models. Furthermore, this study confirmed that the GPR-based model predicted relatively accurate seismic performance values compared with the model through the ANN.

Download Full-text

Machine Learning Methods Applied to Predict Ventilator-Associated Pneumonia with Pseudomonas aeruginosa Infection via Sensor Array of Electronic Nose in Intensive Care Unit

Sensors ◽

10.3390/s19081866 ◽

2019 ◽

Vol 19 (8) ◽

pp. 1866 ◽

Cited By ~ 13

Author(s):

Liao ◽

Wang ◽

Zhang ◽

Abbod ◽

Shih ◽

...

Keyword(s):

Machine Learning ◽

Intensive Care Unit ◽

Pseudomonas Aeruginosa ◽

Intensive Care ◽

Prediction Models ◽

Support Vector ◽

Ventilator Associated Pneumonia ◽

Learning Methods ◽

Machine Learning Methods ◽

Pseudomonas Aeruginosa Infection

One concern to the patients is the off-line detection of pneumonia infection status after using the ventilator in the intensive care unit. Hence, machine learning methods for ventilator-associated pneumonia (VAP) rapid diagnose are proposed. A popular device, Cyranose 320 e-nose, is usually used in research on lung disease, which is a highly integrated system and sensor comprising 32 array using polymer and carbon black materials. In this study, a total of 24 subjects were involved, including 12 subjects who are infected with pneumonia, and the rest are non-infected. Three layers of back propagation artificial neural network and support vector machine (SVM) methods were applied to patients’ data to predict whether they are infected with VAP with Pseudomonas aeruginosa infection. Furthermore, in order to improve the accuracy and the generalization of the prediction models, the ensemble neural networks (ENN) method was applied. In this study, ENN and SVM prediction models were trained and tested. In order to evaluate the models’ performance, a fivefold cross-validation method was applied. The results showed that both ENN and SVM models have high recognition rates of VAP with Pseudomonas aeruginosa infection, with 0.9479 ± 0.0135 and 0.8686 ± 0.0422 accuracies, 0.9714 ± 0.0131, 0.9250 ± 0.0423 sensitivities, and 0.9288 ± 0.0306, 0.8639 ± 0.0276 positive predictive values, respectively. The ENN model showed better performance compared to SVM in the recognition of VAP with Pseudomonas aeruginosa infection. The areas under the receiver operating characteristic curve of the two models were 0.9842 ± 0.0058 and 0.9410 ± 0.0301, respectively, showing that both models are very stable and accurate classifiers. This study aims to assist the physician in providing a scientific and effective reference for performing early detection in Pseudomonas aeruginosa infection or other diseases.

Download Full-text

Supervised machine learning methods in psychology: A practical introduction with annotated R code

10.31234/osf.io/s72vu ◽

2019 ◽

Author(s):

Hannes Rosenbusch ◽

Felix Soldner ◽

Anthony M Evans ◽

Marcel Zeelenberg

Keyword(s):

Machine Learning ◽

Prediction Models ◽

Psychological Research ◽

Supervised Machine Learning ◽

Support Vector ◽

Learning Methods ◽

Comprehensive Overview ◽

K Nearest Neighbors ◽

Machine Learning Methods ◽

Out Of Sample

Machine learning methods for pattern detection and prediction are increasingly prevalent in psychological research. We provide a comprehensive overview of machine learning, its applications, and how to implement models for research. We review fundamental concepts of machine learning, such as prediction accuracy and out-of-sample evaluation, and summarize four standard prediction algorithms: linear regressions, ridge regressions, decision trees, and random forests (plus k-nearest neighbors, Naïve Bayes classifiers, and support vector machines in the supplementary material). This selection provides a set of powerful models that are implemented regularly in machine learning projects. We demonstrate each method with examples and annotated R code, and discuss best practices for determining sample sizes; comparing model performances; tuning prediction models; preregistering prediction models; and reporting results. Finally, we discuss the value of machine learning methods in maintaining psychology’s status as a predictive science.

Download Full-text

Modeling Traders’ Behavior with Deep Learning and Machine Learning Methods: Evidence from BIST 100 Index

Complexity ◽

10.1155/2020/8285149 ◽

2020 ◽

Vol 2020 ◽

pp. 1-16

Author(s):

Afan Hasan ◽

Oya Kalıpsız ◽

Selim Akyokuş

Keyword(s):

Machine Learning ◽

Deep Learning ◽

Financial Market ◽

Performance Metrics ◽

Confusion Matrix ◽

Support Vector ◽

Human Beings ◽

Learning Methods ◽

Technical Indicators ◽

Machine Learning Methods

Although the vast majority of fundamental analysts believe that technical analysts’ estimates and technical indicators used in these analyses are unresponsive, recent research has revealed that both professionals and individual traders are using technical indicators. A correct estimate of the direction of the financial market is a very challenging activity, primarily due to the nonlinear nature of the financial time series. Deep learning and machine learning methods on the other hand have achieved very successful results in many different areas where human beings are challenged. In this study, technical indicators were integrated into the methods of deep learning and machine learning, and the behavior of the traders was modeled in order to increase the accuracy of forecasting of the financial market direction. A set of technical indicators has been examined based on their application in technical analysis as input features to predict the oncoming (one-period-ahead) direction of Istanbul Stock Exchange (BIST100) national index. To predict the direction of the index, Deep Neural Network (DNN), Support Vector Machine (SVM), Random Forest (RF), and Logistic Regression (LR) classification techniques are used. The performance of these models is evaluated on the basis of various performance metrics such as confusion matrix, compound return, and max drawdown.

Download Full-text

Applications of Machine Learning Methods in Drug Toxicity Prediction

Current Topics in Medicinal Chemistry ◽

10.2174/1568026618666180727152557 ◽

2018 ◽

Vol 18 (12) ◽

pp. 987-997 ◽

Cited By ~ 26

Author(s):

Li Zhang ◽

Hui Zhang ◽

Haixin Ai ◽

Huan Hu ◽

Shimeng Li ◽

...

Keyword(s):

Machine Learning ◽

Computational Methods ◽

Drug Toxicity ◽

Prediction Models ◽

New Drugs ◽

Toxicity Tests ◽

Support Vector ◽

Toxicity Prediction ◽

Learning Methods ◽

Machine Learning Methods

Toxicity evaluation is an important part of the preclinical safety assessment of new drugs, which is directly related to human health and the fate of drugs. It is of importance to study how to evaluate drug toxicity accurately and economically. The traditional in vitro and in vivo toxicity tests are laborious, time-consuming, highly expensive, and even involve animal welfare issues. Computational methods developed for drug toxicity prediction can compensate for the shortcomings of traditional methods and have been considered useful in the early stages of drug development. Numerous drug toxicity prediction models have been developed using a variety of computational methods. With the advance of the theory of machine learning and molecular representation, more and more drug toxicity prediction models are developed using a variety of machine learning methods, such as support vector machine, random forest, naive Bayesian, back propagation neural network. And significant advances have been made in many toxicity endpoints, such as carcinogenicity, mutagenicity, and hepatotoxicity. In this review, we aimed to provide a comprehensive overview of the machine learning based drug toxicity prediction studies conducted in recent years. In addition, we compared the performance of the models proposed in these studies in terms of accuracy, sensitivity, and specificity, providing a view of the current state-of-the-art in this field and highlighting the issues in the current studies.

Download Full-text

Metabolic Syndrome Prediction Models Using Machine Learning and Sasang Constitution Type

Evidence-based Complementary and Alternative Medicine ◽

10.1155/2021/8315047 ◽

2021 ◽

Vol 2021 ◽

pp. 1-7

Author(s):

Ji-Eun Park ◽

Sujeong Mun ◽

Siwoo Lee

Keyword(s):

Machine Learning ◽

Nearest Neighbor ◽

Prediction Models ◽

Support Vector ◽

K Nearest Neighbor ◽

Learning Methods ◽

Machine Learning Methods ◽

Sasang Constitution ◽

Constitution Type ◽

Conventional Regression

Background. Machine learning may be a useful tool for predicting metabolic syndrome (MetS), and previous studies also suggest that the risk of MetS differs according to Sasang constitution type. The present study investigated the development of MetS prediction models utilizing machine learning methods and whether the incorporation of Sasang constitution type could improve the performance of those prediction models. Methods. Participants visiting a medical center for a health check-up were recruited in 2005 and 2006. Six kinds of machine learning were utilized (K-nearest neighbor, naive Bayes, random forest, decision tree, multilayer perceptron, and support vector machine), as was conventional logistic regression. Machine learning-derived MetS prediction models with and without the incorporation of Sasang constitution type were compared to investigate whether the former would predict MetS with higher sensitivity. Age, sex, education level, marital status, body mass index, stress, physical activity, alcohol consumption, and smoking were included as potentially predictive factors. Results. A total of 750/2,871 participants had MetS. Among the six types of machine learning methods investigated, multiplayer perceptron and support vector machine exhibited the same performance as the conventional regression method, based on the areas under the receiver operating characteristic curves. The naive-Bayes method exhibited the highest sensitivity (0.49), which was higher than that of the conventional regression method (0.39). The incorporation of Sasang constitution type improved the sensitivity of all of the machine learning methods investigated except for the K-nearest neighbor method. Conclusion. Machine learning-derived models may be useful for MetS prediction, and the incorporation of Sasang constitution type may increase the sensitivity of such models.

Download Full-text

Classification Models using Circulating Neutrophil Transcripts Can Detect Unruptured Intracranial Aneurysm

10.21203/rs.3.rs-17161/v2 ◽

2020 ◽

Author(s):

Kerry E Poppenberg ◽

Vincent M Tutino ◽

Lu Li ◽

Muhammad Waqas ◽

Armond June ◽

...

Keyword(s):

Machine Learning ◽

Random Forest ◽

Prediction Models ◽

Model Performance ◽

Supervised Machine Learning ◽

Support Vector ◽

Learning Methods ◽

Training Cohort ◽

Network Analyses ◽

Machine Learning Methods

Abstract Background: Intracranial aneurysms (IAs) are dangerous because of their potential to rupture. We previously found significant RNA expression differences in circulating neutrophils between patients with and without unruptured IAs and trained machine learning models to predict presence of IA using 40 neutrophil transcriptomes. Here, we aim to develop a predictive model for unruptured IA using neutrophil transcriptomes from a larger population and more robust machine learning methods. Methods: Neutrophil RNA extracted from the blood of 134 patients (55 with IA, 79 IA-free controls) was subjected to next-generation RNA sequencing. In a randomly-selected training cohort (n=94), the Least Absolute Shrinkage and Selection Operator (LASSO) selected transcripts, from which we constructed prediction models via 4 well-established supervised machine-learning algorithms (K-Nearest Neighbors, Random Forest, and Support Vector Machines with Gaussian and cubic kernels). We tested the models in the remaining samples (n=40) and assessed model performance by receiver-operating-characteristic (ROC) curves. Real-time quantitative polymerase chain reaction (RT-qPCR) of 9 IA-associated genes was used to verify gene expression in a subset of 49 neutrophil RNA samples. We also examined the potential influence of demographics and comorbidities on model prediction. Results: Feature selection using LASSO in the training cohort identified 37 IA-associated transcripts. Models trained using these transcripts had a maximum accuracy of 90% in the testing cohort. The testing performance across all methods had an average area under ROC curve (AUC)=0.97, an improvement over our previous models. The Random Forest model performed best across both training and testing cohorts. RT-qPCR confirmed expression differences in 7 of 9 genes tested. Gene ontology and IPA network analyses performed on the 37 model genes reflected dysregulated inflammation, cell signaling, and apoptosis processes. In our data, demographics and comorbidities did not affect model performance. Conclusions: We improved upon our previous IA prediction models based on circulating neutrophil transcriptomes by increasing sample size and by implementing LASSO and more robust machine learning methods. Future studies are needed to validate these models in larger cohorts and further investigate effect of covariates.

Download Full-text

An Empirical Comparison of Machine-Learning Methods on Bank Client Credit Assessments

Sustainability ◽

10.3390/su11030699 ◽

2019 ◽

Vol 11 (3) ◽

pp. 699 ◽

Cited By ~ 13

Author(s):

Lkhagvadorj Munkhdalai ◽

Tsendsuren Munkhdalai ◽

Oyun-Erdene Namsrai ◽

Jong Lee ◽

Keun Ryu

Keyword(s):

Machine Learning ◽

Neural Networks ◽

Random Forest ◽

Deep Neural Networks ◽

Credit Scoring ◽

Support Vector ◽

Learning Approaches ◽

Learning Methods ◽

Human Expert ◽

Machine Learning Methods

Machine learning and artificial intelligence have achieved a human-level performance in many application domains, including image classification, speech recognition and machine translation. However, in the financial domain expert-based credit risk models have still been dominating. Establishing meaningful benchmark and comparisons on machine-learning approaches and human expert-based models is a prerequisite in further introducing novel methods. Therefore, our main goal in this study is to establish a new benchmark using real consumer data and to provide machine-learning approaches that can serve as a baseline on this benchmark. We performed an extensive comparison between the machine-learning approaches and a human expert-based model—FICO credit scoring system—by using a Survey of Consumer Finances (SCF) data. As the SCF data is non-synthetic and consists of a large number of real variables, we applied two variable-selection methods: the first method used hypothesis tests, correlation and random forest-based feature importance measures and the second method was only a random forest-based new approach (NAP), to select the best representative features for effective modelling and to compare them. We then built regression models based on various machine-learning algorithms ranging from logistic regression and support vector machines to an ensemble of gradient boosted trees and deep neural networks. Our results demonstrated that if lending institutions in the 2001s had used their own credit scoring model constructed by machine-learning methods explored in this study, their expected credit losses would have been lower, and they would be more sustainable. In addition, the deep neural networks and XGBoost algorithms trained on the subset selected by NAP achieve the highest area under the curve (AUC) and accuracy, respectively.

Download Full-text

DEVELOPMENT AND VALIDATION OF A MODEL FOR THE PREDICTION OF MORTALITY IN CHILDREN UNDER FIVE YEARS WITH CLINICAL PNEUMONIA IN RURAL GAMBIA

10.1101/2021.08.04.21260737 ◽

2021 ◽

Author(s):

Alexander Jarde ◽

David Jeffries ◽

Grant A Mackenzie

Keyword(s):

Machine Learning ◽

Predictive Model ◽

Prediction Models ◽

Characteristic Curve ◽

Machine Learning Algorithms ◽

Support Vector ◽

Learning Methods ◽

Final Model ◽

Machine Learning Methods ◽

Prediction Of Mortality

Background: Pneumonia is the leading cause of death in children aged 1-59 months. Prediction models for child pneumonia mortality have been developed using regression methods but their performance is insufficient for clinical use. Methods: We used a variety of machine learning methods to develop a predictive model for mortality in children with clinical pneumonia enrolled in population-based surveillance in the Basse Health and Demographic Surveillance System in rural Gambia (n=11,012). Four machine learning algorithms (support vector machine, random forest, artifical neural network, and regularized logistic regression) were implemented, fitting all possible combinations of two or more of 16 selected features. Models were shortlisted based on their training set performance , the number of included features, and the reliability of feature measurement. The final model was selected considering its clinical interpretability. Results: When we applied the final model to the test set (55 deaths), the area under the Receiver Operating Characteristic Curve was 0.88 (95% confidence interval: 0.84, 0.91), sensitivity was 0.78 and specificity was 0.77. Conclusions: Our evaluation of multiple machine learning methods combined with minimal and pragmatic feature selection led to a predictive model with very good performance. We plan further validation of our model in different populations.

Download Full-text