Boosting for high-dimensional two-class prediction

Prediction of low Apgar score at five minutes following labor induction intervention in vaginal deliveries: machine learning approach for imbalanced data at a tertiary hospital in North Tanzania

Background

Prediction of low Apgar score for vaginal deliveries following labor induction intervention is critical for improving neonatal health outcomes. We set out to investigate important attributes and train popular machine learning (ML) algorithms to correctly classify neonates with a low Apgar scores from an imbalanced learning perspective.

Methods

We analyzed 7716 induced vaginal deliveries from the electronic birth registry of the Kilimanjaro Christian Medical Centre (KCMC). 733 (9.5%) of which constituted of low (< 7) Apgar score neonates. The ‘extra-tree classifier’ was used to assess features’ importance. We used Area Under Curve (AUC), recall, precision, F-score, Matthews Correlation Coefficient (MCC), balanced accuracy (BA), bookmaker informedness (BM), and markedness (MK) to evaluate the performance of the selected six (6) machine learning classifiers. To address class imbalances, we examined three widely used resampling techniques: the Synthetic Minority Oversampling Technique (SMOTE) and Random Oversampling Examples (ROS) and Random undersampling techniques (RUS). We applied Decision Curve Analysis (DCA) to evaluate the net benefit of the selected classifiers.

Results

Birth weight, maternal age, and gestational age were found to be important predictors for the low Apgar score following induced vaginal delivery. SMOTE, ROS and and RUS techniques were more effective at improving “recalls” among other metrics in all the models under investigation. A slight improvement was observed in the F1 score, BA, and BM. DCA revealed potential benefits of applying Boosting method for predicting low Apgar scores among the tested models.

Conclusion

There is an opportunity for more algorithms to be tested to come up with theoretical guidance on more effective rebalancing techniques suitable for this particular imbalanced ratio. Future research should prioritize a debate on which performance indicators to look up to when dealing with imbalanced or skewed data.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12884-022-04534-0.

VirVACPRED: A Web Server for Prediction of Protective Viral Antigens

Viral antigens are key in the development of vaccines that prevent or eradicate infections caused by these pathogens. Bioinformatics tools are modern alternatives that facilitate the discovery of viral antigens, reducing the costs of experimental assays. We developed a bioinformatics tool called VirVACPRED, which is highly efficient in predicting viral antigens. In this study, we obtained a model based on the gradient boosting classifier, which showed high performance during the training, leave-one-out cross-validation (accuracy = 0.7402, sensitivity = 0.7319, precision = 0.7503, F1 = 0.7251, kappa = 0.4774, Matthews correlation coefficient = 0.4981) and testing (accuracy = 0.8889, sensitivity = 1.0, precision = 0.8276, F1 = 0.9057, kappa = 0.7734, Matthews correlation coefficient = 0.7941). VirVACPRED is a robust tool that can be of great help in the search and proposal of new viral antigens, which can be considered in the development of future vaccines against infections caused by viruses.

A new method for exploring gene-gene and gene-environment interactions in GWAS with tree ensemble methods and SHAP values

Background

The identification of gene–gene and gene–environment interactions in genome-wide association studies is challenging due to the unknown nature of the interactions and the overwhelmingly large number of possible combinations. Parametric regression models are suitable to look for prespecified interactions. Nonparametric models such as tree ensemble models, with the ability to detect any unspecified interaction, have previously been difficult to interpret. However, with the development of methods for model explainability, it is now possible to interpret tree ensemble models efficiently and with a strong theoretical basis.

Results

We propose a tree ensemble- and SHAP-based method for identifying as well as interpreting potential gene–gene and gene–environment interactions on large-scale biobank data. A set of independent cross-validation runs are used to implicitly investigate the whole genome. We apply and evaluate the method using data from the UK Biobank with obesity as the phenotype. The results are in line with previous research on obesity as we identify top SNPs previously associated with obesity. We further demonstrate how to interpret and visualize interaction candidates.

Conclusions

The new method identifies interaction candidates otherwise not detected with parametric regression models. However, further research is needed to evaluate the uncertainties of these candidates. The method can be applied to large-scale biobanks with high-dimensional data.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-021-04041-7.

Representation Learning for Motor Imagery Recognition with Deep Neural Network

This study describes a method for classifying electrocorticograms (ECoGs) based on motor imagery (MI) on the brain–computer interface (BCI) system. This method is different from the traditional feature extraction and classification method. In this paper, the proposed method employs the deep learning algorithm for extracting features and the traditional algorithm for classification. Specifically, we mainly use the convolution neural network (CNN) to extract the features from the training data and then classify those features by combing with the gradient boosting (GB) algorithm. The comprehensive study with CNN and GB algorithms will profoundly help us to obtain more feature information from brain activities, enabling us to obtain the classification results from human body actions. The performance of the proposed framework has been evaluated on the dataset I of BCI Competition III. Furthermore, the combination of deep learning and traditional algorithms provides some ideas for future research with the BCI systems.

Prediction of Important Factors for Bleeding in Liver Cirrhosis Disease Using Ensemble Data Mining Approach

The main motivation to conduct the study presented in this paper was the fact that due to the development of improved solutions for prediction risk of bleeding and thus a faster and more accurate diagnosis of complications in cirrhotic patients, mortality of cirrhosis patients caused by bleeding of varices fell at the turn in the 21th century. Due to this fact, an additional research in this field is needed. The objective of this paper is to develop one prediction model that determines most important factors for bleeding in liver cirrhosis, which is useful for diagnosis and future treatment of patients. To achieve this goal, authors proposed one ensemble data mining methodology, as the most modern in the field of prediction, for integrating on one new way the two most commonly used techniques in prediction, classification with precede attribute number reduction and multiple logistic regression for calibration. Method was evaluated in the study, which analyzed the occurrence of variceal bleeding for 96 patients from the Clinical Center of Nis, Serbia, using 29 data from clinical to the color Doppler. Obtained results showed that proposed method with such big number and different types of data demonstrates better characteristics than individual technique integrated into it.

Recent Development of Computational Predicting Bioluminescent Proteins

Bioluminescent Proteins (BLPs) are widely distributed in many living organisms that act as a key role of light emission in bioluminescence. Bioluminescence serves various functions in finding food and protecting the organisms from predators. With the routine biotechnological application of bioluminescence, it is recognized to be essential for many medical, commercial and other general technological advances. Therefore, the prediction and characterization of BLPs are significant and can help to explore more secrets about bioluminescence and promote the development of application of bioluminescence. Since the experimental methods are money and time-consuming for BLPs identification, bioinformatics tools have played important role in fast and accurate prediction of BLPs by combining their sequences information with machine learning methods. In this review, we summarized and compared the application of machine learning methods in the prediction of BLPs from different aspects. We wish that this review will provide insights and inspirations for researches on BLPs.

Stillbirth risk prediction using machine learning for a large cohort of births from Western Australia, 1980-2015

Quantification of stillbirth risk has potential to support clinical decision-making. Studies that have attempted to quantify stillbirth risk have been hampered by small event rates, a limited range of predictors that typically exclude obstetric history, lack of validation, and restriction to a single classifier (logistic regression). Consequently, predictive performance remains low, and risk quantification has not been adopted into antenatal practice. The study population consisted of all births to women in Western Australia from 1980 to 2015, excluding terminations. After all exclusions there were 947,025 livebirths and 5,788 stillbirths. Predictive models for stillbirth were developed using multiple machine learning classifiers: regularised logistic regression, decision trees based on classification and regression trees, random forest, extreme gradient boosting (XGBoost), and a multilayer perceptron neural network. We applied 10-fold cross-validation using independent data not used to develop the models. Predictors included maternal socio-demographic characteristics, chronic medical conditions, obstetric complications and family history in both the current and previous pregnancy. In this cohort, 66% of stillbirths were observed for multiparous women. The best performing classifier (XGBoost) predicted 45% (95% CI: 43%, 46%) of stillbirths for all women and 45% (95% CI: 43%, 47%) of stillbirths after the inclusion of previous pregnancy history. Almost half of stillbirths could be potentially identified antenatally based on a combination of current pregnancy complications, congenital anomalies, maternal characteristics, and medical history. Greatest sensitivity is achieved with addition of current pregnancy complications. Ensemble classifiers offered marginal improvement for prediction compared to logistic regression.

Comparison of Machine Learning Techniques for Prediction of Hospitalization in Heart Failure Patients

The present study aims to compare the performance of eight Machine Learning Techniques (MLTs) in the prediction of hospitalization among patients with heart failure, using data from the Gestione Integrata dello Scompenso Cardiaco (GISC) study. The GISC project is an ongoing study that takes place in the region of Puglia, Southern Italy. Patients with a diagnosis of heart failure are enrolled in a long-term assistance program that includes the adoption of an online platform for data sharing between general practitioners and cardiologists working in hospitals and community health districts. Logistic regression, generalized linear model net (GLMN), classification and regression tree, random forest, adaboost, logitboost, support vector machine, and neural networks were applied to evaluate the feasibility of such techniques in predicting hospitalization of 380 patients enrolled in the GISC study, using data about demographic characteristics, medical history, and clinical characteristics of each patient. The MLTs were compared both without and with missing data imputation. Overall, models trained without missing data imputation showed higher predictive performances. The GLMN showed better performance in predicting hospitalization than the other MLTs, with an average accuracy, positive predictive value and negative predictive value of 81.2%, 87.5%, and 75%, respectively. Present findings suggest that MLTs may represent a promising opportunity to predict hospital admission of heart failure patients by exploiting health care information generated by the contact of such patients with the health care system.

Comparing deep belief networks with support vector machines for classifying gene expression data from complex disorders

Genomics data provide great opportunities for translational research and the clinical practice, for example, for predicting disease stages. However, the classification of such data is a challenging task due to their high dimensionality, noise, and heterogeneity. In recent years, deep learning classifiers generated much interest, but due to their complexity, so far, little is known about the utility of this method for genomics. In this paper, we address this problem by studying a computational diagnostics task by classification of breast cancer and inflammatory bowel disease patients based on high‐dimensional gene expression data. We provide a comprehensive analysis of the classification performance of deep belief networks (DBNs) in dependence on its multiple model parameters and in comparison with support vector machines (SVMs). Furthermore, we investigate combined classifiers that integrate DBNs with SVMs. Such a classifier utilizes a DBN as representation learner forming the input for a SVM. Overall, our results provide guidelines for the complex usage of DBN for classifying gene expression data from complex diseases.

Exploring the clinical features of narcolepsy type 1 versus narcolepsy type 2 from European Narcolepsy Network database with machine learning

Narcolepsy is a rare life-long disease that exists in two forms, narcolepsy type-1 (NT1) or type-2 (NT2), but only NT1 is accepted as clearly defined entity. Both types of narcolepsies belong to the group of central hypersomnias (CH), a spectrum of poorly defined diseases with excessive daytime sleepiness as a core feature. Due to the considerable overlap of symptoms and the rarity of the diseases, it is difficult to identify distinct phenotypes of CH. Machine learning (ML) can help to identify phenotypes as it learns to recognize clinical features invisible for humans. Here we apply ML to data from the huge European Narcolepsy Network (EU-NN) that contains hundreds of mixed features of narcolepsy making it difficult to analyze with classical statistics. Stochastic gradient boosting, a supervised learning model with built-in feature selection, results in high performances in testing set. While cataplexy features are recognized as the most influential predictors, machine find additional features, e.g. mean rapid-eye-movement sleep latency of multiple sleep latency test contributes to classify NT1 and NT2 as confirmed by classical statistical analysis. Our results suggest ML can identify features of CH on machine scale from complex databases, thus providing 'ideas' and promising candidates for future diagnostic classifications.

The parameter sensitivity of random forests

Background

The Random Forest (RF) algorithm for supervised machine learning is an ensemble learning method widely used in science and many other fields. Its popularity has been increasing, but relatively few studies address the parameter selection process: a critical step in model fitting. Due to numerous assertions regarding the performance reliability of the default parameters, many RF models are fit using these values. However there has not yet been a thorough examination of the parameter-sensitivity of RFs in computational genomic studies. We address this gap here.

Results

We examined the effects of parameter selection on classification performance using the RF machine learning algorithm on two biological datasets with distinct p/n ratios: sequencing summary statistics (low p/n) and microarray-derived data (high p/n). Here, p, refers to the number of variables and, n, the number of samples. Our findings demonstrate that parameterization is highly correlated with prediction accuracy and variable importance measures (VIMs). Further, we demonstrate that different parameters are critical in tuning different datasets, and that parameter-optimization significantly enhances upon the default parameters.

Conclusions

Parameter performance demonstrated wide variability on both low and high p/n data. Therefore, there is significant benefit to be gained by model tuning RFs away from their default parameter settings.

Electronic supplementary material

The online version of this article (doi:10.1186/s12859-016-1228-x) contains supplementary material, which is available to authorized users.