Search

Systematic reviews (SR) are vital to health care, but have become complicated and time-consuming, due to the rapid expansion of evidence to be synthesised. Fortunately, many tasks of systematic reviews have the potential to be automated or may be assisted by automation. Recent advances in natural language processing, text mining and machine learning have produced new algorithms that can accurately mimic human endeavour in systematic review activity, faster and more cheaply. Automation tools need to be able to work together, to exchange data and results. Therefore, we initiated the International Collaboration for the Automation of Systematic Reviews (ICASR), to successfully put all the parts of automation of systematic review production together. The first meeting was held in Vienna in October 2015. We established a set of principles to enable tools to be developed and integrated into toolkits. This paper sets out the principles devised at that meeting, which cover the need for improvement in efficiency of SR tasks, automation across the spectrum of SR tasks, continuous improvement, adherence to high quality standards, flexibility of use and combining components, the need for a collaboration and varied skills, the desire for open source, shared code and evaluation, and a requirement for replicability through rigorous and open evaluation. Automation has a great potential to improve the speed of systematic reviews. Considerable work is already being done on many of the steps involved in a review. The ‘Vienna Principles’ set out in this paper aim to guide a more coordinated effort which will allow the integration of work by separate teams and build on the experience, code and evaluations done by the many teams working across the globe.

Background With the invention of fitness trackers, it has been possible to continuously monitor a user’s biometric data such as heart rates, number of footsteps taken, and amount of calories burned. This paper names the time series of these three types of biometric data, the user’s “activeness”, and investigates the feasibility in modeling and predicting the long-term activeness of the user. Methods The dataset used in this study consisted of several months of biometric time-series data gathered by seven users independently. Four recurrent neural network (RNN) architectures–as well as a deep neural network and a simple regression model–were proposed to investigate the performance on predicting the activeness of the user under various length-related hyper-parameter settings. In addition, the learned model was tested to predict the time period when the user’s activeness falls below a certain threshold. Results A preliminary experimental result shows that each type of activeness data exhibited a short-term autocorrelation; and among the three types of data, the consumed calories and the number of footsteps were positively correlated, while the heart rate data showed almost no correlation with neither of them. It is probably due to this characteristic of the dataset that although the RNN models produced the best results on modeling the user’s activeness, the difference was marginal; and other baseline models, especially the linear regression model, performed quite admirably as well. Further experimental results show that it is feasible to predict a user’s future activeness with precision, for example, a trained RNN model could predict–with the precision of 84%–when the user would be less active within the next hour given the latest 15 min of his activeness data. Conclusions This paper defines and investigates the notion of a user’s “activeness”, and shows that forecasting the long-term activeness of the user is indeed possible. Such information can be utilized by a health-related application to proactively recommend suitable events or services to the user.

Background Single-nucleotide polymorphisms (SNPs) selection and identification are the most important tasks in Genome-wide association data analysis. The problem is difficult because genome-wide association data is very high dimensional and a large portion of SNPs in the data is irrelevant to the disease. Advanced machine learning methods have been successfully used in Genome-wide association studies (GWAS) for identification of genetic variants that have relatively big effects in some common, complex diseases. Among them, the most successful one is Random Forests (RF). Despite of performing well in terms of prediction accuracy in some data sets with moderate size, RF still suffers from working in GWAS for selecting informative SNPs and building accurate prediction models. In this paper, we propose to use a new two-stage quality-based sampling method in random forests, named ts-RF, for SNP subspace selection for GWAS. The method first applies p-value assessment to find a cut-off point that separates informative and irrelevant SNPs in two groups. The informative SNPs group is further divided into two sub-groups: highly informative and weak informative SNPs. When sampling the SNP subspace for building trees for the forest, only those SNPs from the two sub-groups are taken into account. The feature subspaces always contain highly informative SNPs when used to split a node at a tree. Results This approach enables one to generate more accurate trees with a lower prediction error, meanwhile possibly avoiding overfitting. It allows one to detect interactions of multiple SNPs with the diseases, and to reduce the dimensionality and the amount of Genome-wide association data needed for learning the RF model. Extensive experiments on two genome-wide SNP data sets (Parkinson case-control data comprised of 408,803 SNPs and Alzheimer case-control data comprised of 380,157 SNPs) and 10 gene data sets have demonstrated that the proposed model significantly reduced prediction errors and outperformed most existing the-state-of-the-art random forests. The top 25 SNPs in Parkinson data set were identified by the proposed model including four interesting genes associated with neurological disorders. Conclusion The presented approach has shown to be effective in selecting informative sub-groups of SNPs potentially associated with diseases that traditional statistical approaches might fail. The new RF works well for the data where the number of case-control objects is much smaller than the number of SNPs, which is a typical problem in gene data and GWAS. Experiment results demonstrated the effectiveness of the proposed RF model that outperformed the state-of-the-art RFs, including Breiman's RF, GRRF and wsRF methods.

Background Current approaches of predicting protein functions from a protein-protein interaction (PPI) dataset are based on an assumption that the available functions of the proteins (a.k.a. annotated proteins) will determine the functions of the proteins whose functions are unknown yet at the moment (a.k.a. un-annotated proteins). Therefore, the protein function prediction is a mono-directed and one-off procedure, i.e. from annotated proteins to un-annotated proteins. However, the interactions between proteins are mutual rather than static and mono-directed, although functions of some proteins are unknown for some reasons at present. That means when we use the similarity-based approach to predict functions of un-annotated proteins, the un-annotated proteins, once their functions are predicted, will affect the similarities between proteins, which in turn will affect the prediction results. In other words, the function prediction is a dynamic and mutual procedure. This dynamic feature of protein interactions, however, was not considered in the existing prediction algorithms. Results In this paper, we propose a new prediction approach that predicts protein functions iteratively. This iterative approach incorporates the dynamic and mutual features of PPI interactions, as well as the local and global semantic influence of protein functions, into the prediction. To guarantee predicting functions iteratively, we propose a new protein similarity from protein functions. We adapt new evaluation metrics to evaluate the prediction quality of our algorithm and other similar algorithms. Experiments on real PPI datasets were conducted to evaluate the effectiveness of the proposed approach in predicting unknown protein functions. Conclusions The iterative approach is more likely to reflect the real biological nature between proteins when predicting functions. A proper definition of protein similarity from protein functions is the key to predicting functions iteratively. The evaluation results demonstrated that in most cases, the iterative approach outperformed non-iterative ones with higher prediction quality in terms of prediction precision, recall and F-value.

Background Designing appropriate machine learning methods for identifying genes that have a significant discriminating power for disease outcomes has become more and more important for our understanding of diseases at genomic level. Although many machine learning methods have been developed and applied to the area of microarray gene expression data analysis, the majority of them are based on linear models, which however are not necessarily appropriate for the underlying connection between the target disease and its associated explanatory genes. Linear model based methods usually also bring in false positive significant features more easily. Furthermore, linear model based algorithms often involve calculating the inverse of a matrix that is possibly singular when the number of potentially important genes is relatively large. This leads to problems of numerical instability. To overcome these limitations, a few non-linear methods have recently been introduced to the area. Many of the existing non-linear methods have a couple of critical problems, the model selection problem and the model parameter tuning problem, that remain unsolved or even untouched. In general, a unified framework that allows model parameters of both linear and non-linear models to be easily tuned is always preferred in real-world applications. Kernel-induced learning methods form a class of approaches that show promising potentials to achieve this goal. Results A hierarchical statistical model named kernel-imbedded Gaussian process (KIGP) is developed under a unified Bayesian framework for binary disease classification problems using microarray gene expression data. In particular, based on a probit regression setting, an adaptive algorithm with a cascading structure is designed to find the appropriate kernel, to discover the potentially significant genes, and to make the optimal class prediction accordingly. A Gibbs sampler is built as the core of the algorithm to make Bayesian inferences. Simulation studies showed that, even without any knowledge of the underlying generative model, the KIGP performed very close to the theoretical Bayesian bound not only in the case with a linear Bayesian classifier but also in the case with a very non-linear Bayesian classifier. This sheds light on its broader usability to microarray data analysis problems, especially to those that linear methods work awkwardly. The KIGP was also applied to four published microarray datasets, and the results showed that the KIGP performed better than or at least as well as any of the referred state-of-the-art methods did in all of these cases. Conclusion Mathematically built on the kernel-induced feature space concept under a Bayesian framework, the KIGP method presented in this paper provides a unified machine learning approach to explore both the linear and the possibly non-linear underlying relationship between the target features of a given binary disease classification problem and the related explanatory gene expression data. More importantly, it incorporates the model parameter tuning into the framework. The model selection problem is addressed in the form of selecting a proper kernel type. The KIGP method also gives Bayesian probabilistic predictions for disease classification. These properties and features are beneficial to most real-world applications. The algorithm is naturally robust in numerical computation. The simulation studies and the published data studies demonstrated that the proposed KIGP performs satisfactorily and consistently.

Background Multicategory Support Vector Machines (MC-SVM) are powerful classification systems with excellent performance in a variety of data classification problems. Since the process of generating models in traditional multicategory support vector machines for large datasets is very computationally intensive, there is a need to improve the performance using high performance computing techniques. Results In this paper, Parallel Multicategory Support Vector Machines (PMC-SVM) have been developed based on the sequential minimum optimization-type decomposition method for support vector machines (SMO-SVM). It was implemented in parallel using MPI and C++ libraries and executed on both shared memory supercomputer and Linux cluster for multicategory classification of microarray data. PMC-SVM has been analyzed and evaluated using four microarray datasets with multiple diagnostic categories, such as different cancer types and normal tissue types. Conclusion The experiments show that the PMC-SVM can significantly improve the performance of classification of microarray data without loss of accuracy, compared with previous work.