Feature selection in Bayesian classifiers for the prognosis of survival of cirrhotic patients treated with TIPS

CondiS web app: imputation of censored lifetimes for machine learning-based survival analysis

SUMMARY

In the era of big data, machine learning techniques are widely applied to every area in biomedical research including survival analysis. It is well recognized that censoring, which is a common missing issue in survival time data, hampers the direct usage of these machine learning techniques. Here, we present CondiS, a web toolkit with graphical user interface to help impute the survival times for censored observations and predict the survival times for future enrolled patients. CondiS imputes a censored survival time based on its distribution conditional on its observed part. When covariates are available, CondiS-X incorporates this information to further increase the imputation accuracy. Users can also upload data of newly enrolled patients and predict their survival times. As the first web-app tool with an imputation function for censored lifetime data, CondiS web can facilitate conducting survival analysis with machine learning approaches.

AVAILABILITY AND IMPLEMENTATION

CondiS is an open-source application implemented with Shiny in R, available free at: https://biostatistics.mdanderson.org/shinyapps/CondiS/.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

CondiS: A conditional survival distribution-based method for censored data imputation overcoming the hurdle in machine learning-based survival analysis

Data analyses by machine learning (ML) algorithms are gaining popularity in biomedical research. When time-to-event data are of interest, censoring is common and needs to be properly addressed. Most ML methods cannot conveniently and appropriately take the censoring information into consideration, potentially leading to inaccurate or biased results. We aim to develop a general-purpose method for imputing censored survival data, facilitating downstream ML analysis. In this study, we propose a novel method of imputing the survival times for censored observations. The proposal is based on their conditional survival distributions (CondiS) derived from Kaplan-Meier estimators. CondiS can replace censored observations with their best approximations from the statistical model, allowing for direct application of ML methods. When covariates are available, we extend CondiS by incorporating the covariate information through ML modeling (CondiS-X), which further improves the accuracy of the imputed survival time. Compared with existing methods with similar purposes, the proposed methods achieved smaller prediction errors and higher concordance with the underlying true survival times in extensive simulation studies. We also demonstrated the usage and advantages of the proposed methods through two real-world cancer datasets. The major advantage of CondiS is that it allows for the direct application of standard ML techniques for analysis once the censored survival times are imputed. We present a user-friendly R package to implement our method, which is a useful tool for ML-based biomedical research in this era of big data.

A novel approach to fully representing the diversity in conditional dependencies for learning Bayesian network classifier

Bayesian network classifiers (BNCs) have proved their effectiveness and efficiency in the supervised learning framework. Numerous variations of conditional independence assumption have been proposed to address the issue of NP-hard structure learning of BNC. However, researchers focus on identifying conditional dependence rather than conditional independence, and information-theoretic criteria cannot identify the diversity in conditional (in)dependencies for different instances. In this paper, the maximum correlation criterion and minimum dependence criterion are introduced to sort attributes and identify conditional independencies, respectively. The heuristic search strategy is applied to find possible global solution for achieving the trade-off between significant dependency relationships and independence assumption. Our extensive experimental evaluation on widely used benchmark data sets reveals that the proposed algorithm achieves competitive classification performance compared to state-of-the-art single model learners (e.g., TAN, KDB, KNN and SVM) and ensemble learners (e.g., ATAN and AODE).

Evaluation of the impact of body mass index on venous thromboembolism risk factors

In this paper, we investigate the interaction impacts of body mass index (BMI) on the other important risk factors for venous thromboembolism (VTE), using deep venous thrombosis (DVT) patient data from the International Warfarin Pharmacogenetics Consortium (IWPC). We apply eight machine learning techniques, including naive Bayes classifier (NB), support vector machine (SVM), elastic net regression (ENET), logistic regression (LR), lasso regression (LAR), multivariate adaptive regression splines (MARS), boosted regression tree (BRT) and random forest model (RF). The RF method is selected as the best model for classification. Out of 33 features considered in this study, we identify 12 variables as relatively important risk factors for VTE. Finally, we examine the interaction impacts of BMI on these important VTE risk factors. We conclude that the impacts of risk factors on VTE incidence are varying across different BMI groups, and the variations are different for different risk factors. Therefore the interaction impacts of BMI on the other risk factors have to be taken into account in order to better understand the incidence of VTE.

Structure Learning of Bayesian Network Based on Adaptive Thresholding

Direct dependencies and conditional dependencies in restricted Bayesian network classifiers (BNCs) are two basic kinds of dependencies. Traditional approaches, such as filter and wrapper, have proved to be beneficial to identify non-significant dependencies one by one, whereas the high computational overheads make them inefficient especially for those BNCs with high structural complexity. Study of the distributions of information-theoretic measures provides a feasible approach to identifying non-significant dependencies in batch that may help increase the structure reliability and avoid overfitting. In this paper, we investigate two extensions to the k-dependence Bayesian classifier, MI-based feature selection, and CMI-based dependence selection. These two techniques apply a novel adaptive thresholding method to filter out redundancy and can work jointly. Experimental results on 30 datasets from the UCI machine learning repository demonstrate that adaptive thresholds can help distinguish between dependencies and independencies and the proposed algorithm achieves competitive classification performance compared to several state-of-the-art BNCs in terms of 0–1 loss, root mean squared error, bias, and variance.

Metabolic Syndrome Prediction Using Machine Learning Models with Genetic and Clinical Information from a Nonobese Healthy Population

The prevalence of metabolic syndrome (MS) in the nonobese population is not low. However, the identification and risk mitigation of MS are not easy in this population. We aimed to develop an MS prediction model using genetic and clinical factors of nonobese Koreans through machine learning methods. A prediction model for MS was designed for a nonobese population using clinical and genetic polymorphism information with five machine learning algorithms, including naïve Bayes classification (NB). The analysis was performed in two stages (training and test sets). Model A was designed with only clinical information (age, sex, body mass index, smoking status, alcohol consumption status, and exercise status), and for model B, genetic information (for 10 polymorphisms) was added to model A. Of the 7,502 nonobese participants, 647 (8.6%) had MS. In the test set analysis, for the maximum sensitivity criterion, NB showed the highest sensitivity: 0.38 for model A and 0.42 for model B. The specificity of NB was 0.79 for model A and 0.80 for model B. In a comparison of the performances of models A and B by NB, model B (area under the receiver operating characteristic curve [AUC] = 0.69, clinical and genetic information input) showed better performance than model A (AUC = 0.65, clinical information only input). We designed a prediction model for MS in a nonobese population using clinical and genetic information. With this model, we might convince nonobese MS individuals to undergo health checks and adopt behaviors associated with a preventive lifestyle.

RDE: A novel approach to improve the classification performance and expressivity of KDB

Bayesian network classifiers (BNCs) have demonstrated competitive classification performance in a variety of real-world applications. A highly scalable BNC with high expressivity is extremely desirable. This paper proposes Redundant Dependence Elimination (RDE) for improving the classification performance and expressivity of k-dependence Bayesian classifier (KDB). To demonstrate the unique characteristics of each case, RDE identifies redundant conditional dependencies and then substitute/remove them. The learned personalized k-dependence Bayesian Classifier (PKDB) can achieve high-confidence conditional probabilities, and graphically interpret the dependency relationships between attributes. Two thyroid cancer datasets and four other cancer datasets from the UCI machine learning repository are selected for our experimental study. The experimental results prove the effectiveness of the proposed algorithm in terms of zero-one loss, bias, variance and AUC.

Using multiple classifiers for predicting the risk of endovascular aortic aneurysm repair re-intervention through hybrid feature selection

Feature selection is essential in medical area; however, its process becomes complicated with the presence of censoring which is the unique character of survival analysis. Most survival feature selection methods are based on Cox's proportional hazard model, though machine learning classifiers are preferred. They are less employed in survival analysis due to censoring which prevents them from directly being used to survival data. Among the few work that employed machine learning classifiers, partial logistic artificial neural network with auto-relevance determination is a well-known method that deals with censoring and perform feature selection for survival data. However, it depends on data replication to handle censoring which leads to unbalanced and biased prediction results especially in highly censored data. Other methods cannot deal with high censoring. Therefore, in this article, a new hybrid feature selection method is proposed which presents a solution to high level censoring. It combines support vector machine, neural network, and K-nearest neighbor classifiers using simple majority voting and a new weighted majority voting method based on survival metric to construct a multiple classifier system. The new hybrid feature selection process uses multiple classifier system as a wrapper method and merges it with iterated feature ranking filter method to further reduce features. Two endovascular aortic repair datasets containing 91% censored patients collected from two centers were used to construct a multicenter study to evaluate the performance of the proposed approach. The results showed the proposed technique outperformed individual classifiers and variable selection methods based on Cox's model such as Akaike and Bayesian information criterions and least absolute shrinkage and selector operator in p values of the log-rank test, sensitivity, and concordance index. This indicates that the proposed classifier is more powerful in correctly predicting the risk of re-intervention enabling doctor in selecting patients' future follow-up plan.

Feature selection through validation and un-censoring of endovascular repair survival data for predicting the risk of re-intervention

Background

Feature selection (FS) process is essential in the medical area as it reduces the effort and time needed for physicians to measure unnecessary features. Choosing useful variables is a difficult task with the presence of censoring which is the unique characteristic in survival analysis. Most survival FS methods depend on Cox’s proportional hazard model; however, machine learning techniques (MLT) are preferred but not commonly used due to censoring. Techniques that have been proposed to adopt MLT to perform FS with survival data cannot be used with the high level of censoring. The researcher’s previous publications proposed a technique to deal with the high level of censoring. It also used existing FS techniques to reduce dataset dimension. However, in this paper a new FS technique was proposed and combined with feature transformation and the proposed uncensoring approaches to select a reduced set of features and produce a stable predictive model.

Methods

In this paper, a FS technique based on artificial neural network (ANN) MLT is proposed to deal with highly censored Endovascular Aortic Repair (EVAR). Survival data EVAR datasets were collected during 2004 to 2010 from two vascular centers in order to produce a final stable model. They contain almost 91% of censored patients. The proposed approach used a wrapper FS method with ANN to select a reduced subset of features that predict the risk of EVAR re-intervention after 5 years to patients from two different centers located in the United Kingdom, to allow it to be potentially applied to cross-centers predictions. The proposed model is compared with the two popular FS techniques; Akaike and Bayesian information criteria (AIC, BIC) that are used with Cox’s model.

Results

The final model outperforms other methods in distinguishing the high and low risk groups; as they both have concordance index and estimated AUC better than the Cox’s model based on AIC, BIC, Lasso, and SCAD approaches. These models have p-values lower than 0.05, meaning that patients with different risk groups can be separated significantly and those who would need re-intervention can be correctly predicted.

Conclusion

The proposed approach will save time and effort made by physicians to collect unnecessary variables. The final reduced model was able to predict the long-term risk of aortic complications after EVAR. This predictive model can help clinicians decide patients’ future observation plan.

Electronic supplementary material

The online version of this article (doi:10.1186/s12911-017-0508-3) contains supplementary material, which is available to authorized users.

Fuzzy Evidential Network and Its Application as Medical Prognosis and Diagnosis Models

Uncertainty is one of the important facts of the medical knowledge. Medical prognosis and diagnosis, as the essential parts of medical knowledge, is affected by different aspects of uncertainty, which must be managed. In the previous studies, different theories such as Bayesian probability theory, evidence theory, and fuzzy set theory have been developed to represent and manage different aspects of uncertainty. Recently, hybrid frameworks are suggested to deal with various types of uncertainty in a single framework. Evidential networks are general frameworks for dealing explicitly with total and partial ignorance and offer powerful combination rule of contradictory evidence. In this framework, the fuzziness of linguistic variables is neglected while these variables commonly appear in the medical domain knowledge and different sources of medical information. In addition, the evidential network parameters are determined based on the experts' knowledge and no data-driven algorithm is provided to learn these parameters. In this study, a novel hybrid framework called fuzzy evidential network was suggested to manage the imprecision and epistemic uncertainty of medical prognosis and diagnosis. Also, a data-driven algorithm based on the fuzzy set theory and the fuzzy maximum likelihood is provided to learn the network parameters from clinical databases. The performance of the proposed framework as various prognosis and diagnosis models, compared with well-known machine learning algorithms and the results showed its superiority. Also, an evidential method is suggested to handle the missing values and its results were compared with KNN imputation method.

Multi-stage classification method oriented to aerial image based on low-rank recovery and multi-feature fusion sparse representation

Automatic classification of terrain surfaces from an aerial image is essential for an autonomous unmanned aerial vehicle (UAV) landing at an unprepared site by using vision. Diverse terrain surfaces may show similar spectral properties due to the illumination and noise that easily cause poor classification performance. To address this issue, a multi-stage classification algorithm based on low-rank recovery and multi-feature fusion sparse representation is proposed. First, color moments and Gabor texture feature are extracted from training data and stacked as column vectors of a dictionary. Then we perform low-rank matrix recovery for the dictionary by using augmented Lagrange multipliers and construct a multi-stage terrain classifier. Experimental results on an aerial map database that we prepared verify the classification accuracy and robustness of the proposed method.

Adapting machine learning techniques to censored time-to-event health record data: A general-purpose approach using inverse probability of censoring weighting

Models for predicting the probability of experiencing various health outcomes or adverse events over a certain time frame (e.g., having a heart attack in the next 5years) based on individual patient characteristics are important tools for managing patient care. Electronic health data (EHD) are appealing sources of training data because they provide access to large amounts of rich individual-level data from present-day patient populations. However, because EHD are derived by extracting information from administrative and clinical databases, some fraction of subjects will not be under observation for the entire time frame over which one wants to make predictions; this loss to follow-up is often due to disenrollment from the health system. For subjects without complete follow-up, whether or not they experienced the adverse event is unknown, and in statistical terms the event time is said to be right-censored. Most machine learning approaches to the problem have been relatively ad hoc; for example, common approaches for handling observations in which the event status is unknown include (1) discarding those observations, (2) treating them as non-events, (3) splitting those observations into two observations: one where the event occurs and one where the event does not. In this paper, we present a general-purpose approach to account for right-censored outcomes using inverse probability of censoring weighting (IPCW). We illustrate how IPCW can easily be incorporated into a number of existing machine learning algorithms used to mine big health care data including Bayesian networks, k-nearest neighbors, decision trees, and generalized additive models. We then show that our approach leads to better calibrated predictions than the three ad hoc approaches when applied to predicting the 5-year risk of experiencing a cardiovascular adverse event, using EHD from a large U.S. Midwestern healthcare system.

A Naive Bayes machine learning approach to risk prediction using censored, time-to-event data

Predicting an individual's risk of experiencing a future clinical outcome is a statistical task with important consequences for both practicing clinicians and public health experts. Modern observational databases such as electronic health records provide an alternative to the longitudinal cohort studies traditionally used to construct risk models, bringing with them both opportunities and challenges. Large sample sizes and detailed covariate histories enable the use of sophisticated machine learning techniques to uncover complex associations and interactions, but observational databases are often 'messy', with high levels of missing data and incomplete patient follow-up. In this paper, we propose an adaptation of the well-known Naive Bayes machine learning approach to time-to-event outcomes subject to censoring. We compare the predictive performance of our method with the Cox proportional hazards model which is commonly used for risk prediction in healthcare populations, and illustrate its application to prediction of cardiovascular risk using an electronic health record dataset from a large Midwest integrated healthcare system.

Elucidation of pathways driving asthma pathogenesis: development of a systems-level analytic strategy

Asthma is a genetically complex, chronic lung disease defined clinically as episodic airflow limitation and breathlessness that is at least partially reversible, either spontaneously or in response to therapy. Whereas asthma was rare in the late 1800s and early 1900s, the marked increase in its incidence and prevalence since the 1960s points to substantial gene × environment interactions occurring over a period of years, but these interactions are very poorly understood (1-6). It is widely believed that the majority of asthma begins during childhood and manifests first as intermittent wheeze. However, wheeze is also very common in infancy and only a subset of wheezy children progress to persistent asthma for reasons that are largely obscure. Here, we review the current literature regarding causal pathways leading to early asthma development and chronicity. Given the complex interactions of many risk factors over time eventually leading to apparently multiple asthma phenotypes, we suggest that deeply phenotyped cohort studies combined with sophisticated network models will be required to derive the next generation of biological and clinical insights in asthma pathogenesis.

Bayesian neural network approach for determining the risk of re-intervention after endovascular aortic aneurysm repair

This article proposes a Bayesian neural network approach to determine the risk of re-intervention after endovascular aortic aneurysm repair surgery. The target of proposed technique is to determine which patients have high chance to re-intervention (high-risk patients) and which are not (low-risk patients) after 5 years of the surgery. Two censored datasets relating to the clinical conditions of aortic aneurysms have been collected from two different vascular centers in the United Kingdom. A Bayesian network was first employed to solve the censoring issue in the datasets. Then, a back propagation neural network model was built using the uncensored data of the first center to predict re-intervention on the second center and classify the patients into high-risk and low-risk groups. Kaplan-Meier curves were plotted for each group of patients separately to show whether there is a significant difference between the two risk groups. Finally, the logrank test was applied to determine whether the neural network model was capable of predicting and distinguishing between the two risk groups. The results show that the Bayesian network used for uncensoring the data has improved the performance of the neural networks that were built for the two centers separately. More importantly, the neural network that was trained with uncensored data of the first center was able to predict and discriminate between groups of low risk and high risk of re-intervention after 5 years of endovascular aortic aneurysm surgery at center 2 (p = 0.0037 in the logrank test).

Biomarker selection and classification of "-omics" data using a two-step bayes classification framework

Identification of suitable biomarkers for accurate prediction of phenotypic outcomes is a goal for personalized medicine. However, current machine learning approaches are either too complex or perform poorly. Here, a novel two-step machine-learning framework is presented to address this need. First, a Naïve Bayes estimator is used to rank features from which the top-ranked will most likely contain the most informative features for prediction of the underlying biological classes. The top-ranked features are then used in a Hidden Naïve Bayes classifier to construct a classification prediction model from these filtered attributes. In order to obtain the minimum set of the most informative biomarkers, the bottom-ranked features are successively removed from the Naïve Bayes-filtered feature list one at a time, and the classification accuracy of the Hidden Naïve Bayes classifier is checked for each pruned feature set. The performance of the proposed two-step Bayes classification framework was tested on different types of -omics datasets including gene expression microarray, single nucleotide polymorphism microarray (SNParray), and surface-enhanced laser desorption/ionization time-of-flight (SELDI-TOF) proteomic data. The proposed two-step Bayes classification framework was equal to and, in some cases, outperformed other classification methods in terms of prediction accuracy, minimum number of classification markers, and computational time.

Predicting dementia development in Parkinson's disease using Bayesian network classifiers

Parkinson's disease (PD) has broadly been associated with mild cognitive impairment (PDMCI) and dementia (PDD). Researchers have studied surrogate, neuroanatomic biomarkers provided by magnetic resonance imaging (MRI) that may help in the early diagnosis of this condition. In this article, four classification models (naïve Bayes, multivariate filter-based naïve Bayes, filter selective naïve Bayes and support vector machines, SVM) have been applied to evaluate their capacity to discriminate between cognitively intact patients with Parkinson's disease (PDCI), PDMCI and PDD. For this purpose, the MRI studies of 45 subjects (16 PDCI, 15 PDMCI and 14 PDD) were acquired and post-processed with Freesurfer, obtaining 112 variables (volumes of subcortical structures and thickness of cortical parcels) per subject. A multivariate filter-based naïve Bayes model was found to be the best classifier, having the highest cross-validated sensitivity, specificity and accuracy. Additionally, the most relevant variables related to dementia in PD, as predicted by our classifiers, were cerebral white matter, and volumes of the lateral ventricles and hippocampi.

AN L1-REGULARIZED NAÏVE BAYES-INSPIRED CLASSIFIER FOR DISCARDING REDUNDANT AND IRRELEVANT PREDICTORS

The naïve Bayes model is a simple but often satisfactory supervised classification method. The original naïve Bayes scheme, does, however, have a serious weakness, namely, the harmful effect of redundant predictors. In this paper, we study how to apply a regularization technique to learn a computationally efficient classifier that is inspired by naïve Bayes. The proposed formulation, combined with an L1-penalty, is capable of discarding harmful, redundant predictors. A modification of the LARS algorithm is devised to solve this problem. We tackle both real-valued and discrete predictors, assuring that our method is applicable to a wide range of data. In the experimental section, we empirically study the effect of redundant and irrelevant predictors. We also test the method on a high dimensional data set from the neuroscience field, where there are many more predictors than data cases. Finally, we run the method on a real data set than combines categorical with numeric predictors. Our approach is compared with several naïve Bayes variants and other classification algorithms (SVM and kNN), and is shown to be competitive.

A new algorithm for reducing the workload of experts in performing systematic reviews

OBJECTIVE

To determine whether a factorized version of the complement naïve Bayes (FCNB) classifier can reduce the time spent by experts reviewing journal articles for inclusion in systematic reviews of drug class efficacy for disease treatment.

DESIGN

The proposed classifier was evaluated on a test collection built from 15 systematic drug class reviews used in previous work. The FCNB classifier was constructed to classify each article as containing high-quality, drug class-specific evidence or not. Weight engineering (WE) techniques were added to reduce underestimation for Medical Subject Headings (MeSH)-based and Publication Type (PubType)-based features. Cross-validation experiments were performed to evaluate the classifier's parameters and performance.

MEASUREMENTS

Work saved over sampling (WSS) at no less than a 95% recall was used as the main measure of performance.

RESULTS

The minimum workload reduction for a systematic review for one topic, achieved with a FCNB/WE classifier, was 8.5%; the maximum was 62.2% and the average over the 15 topics was 33.5%. This is 15.0% higher than the average workload reduction obtained using a voting perceptron-based automated citation classification system.

CONCLUSION

The FCNB/WE classifier is simple, easy to implement, and produces significantly better results in reducing the workload than previously achieved. The results support it being a useful algorithm for machine-learning-based automation of systematic reviews of drug class efficacy for disease treatment.

Prediction of cancer class with majority voting genetic programming classifier using gene expression data

In order to get a better understanding of different types of cancers and to find the possible biomarkers for diseases, recently, many researchers are analyzing the gene expression data using various machine learning techniques. However, due to a very small number of training samples compared to the huge number of genes and class imbalance, most of these methods suffer from overfitting. In this paper, we present a majority voting genetic programming classifier (MVGPC) for the classification of microarray data. Instead of a single rule or a single set of rules, we evolve multiple rules with genetic programming (GP) and then apply those rules to test samples to determine their labels with majority voting technique. By performing experiments on four different public cancer data sets, including multiclass data sets, we have found that the test accuracies of MVGPC are better than those of other methods, including AdaBoost with GP. Moreover, some of the more frequently occurring genes in the classification rules are known to be associated with the types of cancers being studied in this paper.

Detecting reliable gene interactions by a hierarchy of Bayesian network classifiers

The main purpose of a gene interaction network is to map the relationships of the genes that are out of sight when a genomic study is tackled. DNA microarrays allow the measure of gene expression of thousands of genes at the same time. These data constitute the numeric seed for the induction of the gene networks. In this paper, we propose a new approach to build gene networks by means of Bayesian classifiers, variable selection and bootstrap resampling. The interactions induced by the Bayesian classifiers are based both on the expression levels and on the phenotype information of the supervised variable. Feature selection and bootstrap resampling add reliability and robustness to the overall process removing the false positive findings. The consensus among all the induced models produces a hierarchy of dependences and, thus, of variables. Biologists can define the depth level of the model hierarchy so the set of interactions and genes involved can vary from a sparse to a dense set. Experimental results show how these networks perform well on classification tasks. The biological validation matches previous biological findings and opens new hypothesis for future studies.

Machine learning method for knowledge discovery experimented with otoneurological data

We have been interested in developing an otoneurological decision support system that supports diagnostics of vertigo diseases. In this study, we concentrate on testing its inference mechanism and knowledge discovery method. Knowledge is presented as patterns of classes. Each pattern includes attributes with weight and fitness values concerning the class. With the knowledge discovery method it is possible to form fitness values from data. Knowledge formation is based on frequency distributions of attributes. Knowledge formed by the knowledge discovery method is tested with two vertigo data sets and compared to experts' knowledge. The experts' and machine learnt knowledge are also combined in various ways in order to examine effects of weights on classification accuracy. The classification accuracy of knowledge discovery method is compared to 1- and 5-nearest neighbour method and Naive-Bayes classifier. The results showed that knowledge bases combining machine learnt knowledge with the experts' knowledge yielded the best classification accuracies. Further, attribute weighting had an important effect on the classification capability of the system. When considering different diseases in the used data sets, the performance of the knowledge discovery method and the inference method is comparable to other methods employed in this study.

What is behind a summary-evaluation decision?

Research in psychology has reported that, among the variety of possibilities for assessment methodologies, summary evaluation offers a particularly adequate context for inferring text comprehension and topic understanding. However, grades obtained in this methodology are hard to quantify objectively. Therefore, we carried out an empirical study to analyze the decisions underlying human summary-grading behavior. The task consisted of expert evaluation of summaries produced in critically relevant contexts of summarization development, and the resulting data were modeled by means of Bayesian networks using an application called Elvira, which allows for graphically observing the predictive power (if any) of the resultant variables. Thus, in this article, we analyzed summary-evaluation decision making in a computational framework.

Bayesian classification for the selection of in vitro human embryos using morphological and clinical data

In vitro fertilization (IVF) is a medically assisted reproduction technique that enables infertile couples to achieve successful pregnancy. Given the uncertainty of the treatment, we propose an intelligent decision support system based on supervised classification by Bayesian classifiers to aid to the selection of the most promising embryos that will form the batch to be transferred to the woman's uterus. The aim of the supervised classification system is to improve overall success rate of each IVF treatment in which a batch of embryos is transferred each time, where the success is achieved when implantation (i.e. pregnancy) is obtained. Due to ethical reasons, different legislative restrictions apply in every country on this technique. In Spain, legislation allows a maximum of three embryos to form each transfer batch. As a result, clinicians prefer to select the embryos by non-invasive embryo examination based on simple methods and observation focused on morphology and dynamics of embryo development after fertilization. This paper proposes the application of Bayesian classifiers to this embryo selection problem in order to provide a decision support system that allows a more accurate selection than with the actual procedures which fully rely on the expertise and experience of embryologists. For this, we propose to take into consideration a reduced subset of feature variables related to embryo morphology and clinical data of patients, and from this data to induce Bayesian classification models. Results obtained applying a filter technique to choose the subset of variables, and the performance of Bayesian classifiers using them, are presented.