Search

In the current supplement, we are proud to present ten relevant contributions from the 3rd International Work-Conference on Bioinformatics and Biomedical Engineering (IWBBIO 2015), which was held during April 15–17, 2015 in Granada (Spain) [1, 2]. These contributions have been chosen because of their quality and the importance of their findings. IWBBIO 2015 aimed to follow the same friendly environment as previous editions, allowing researchers, scientists and students to show their latest ideas, discoveries and outcomes. The conference sought to focus on diverse fields to create multidisciplinary researches integrating areas like biomedical engineering, computer science, mathematics, artificial intelligence, bioinformatics, statistics or biomedicine. IWBBIO 2015 also promoted the meeting, cooperation and collaboration of scientists with the presentation and discussion of their latest relevant ideas and work-in-progress. These ideas provided important advances to the scientific community in fields like genomics, next-generation sequencing, drug design and advanced pharmacology, biomedical modelling and e-health, among other. Additionally, IWBBIO 2015 was honored with the presence of three invited plenary speakers: Prof. Alfonso Valencia, Prof. Patrick Aloy and Prof. Xavier Estivill. These plenary lectures strengthened the aim of this conference for the diffusion and the discussion of high quality researches from some of the most recognized scientists in these fields. The IWBBIO 2015 has continued as a two-track conference, increasing the number of sessions to a total of 24 oral and 2 poster sessions. It received more than 210 contributions which were reviewed by at least two referees from our estimated program and steering committees. The conference continues accepting both full and abstract submissions for presentations. However, it still maintained a high rate of full contributions against abstracts. IWBBIO 2015 received more than 200 attendees from diverse European nationalities (Spain, United Kingdom, France, Italy, Poland, etc) but also overseas countries like United Stated, Korea, China or India. Those contributions which were considered more relevant taking into account the evaluation and opinion of reviewers and chairmen were then invited to participate in this supplement for the BioMedical Engineering OnLine journal. In this regard, this supplement is specially based on particular technological implementations and developments oriented to facilitate and improve healthcare and wellness. All the novel researches showed in this supplement have a massively interdisciplinary background, dealing with hot topics in the area of biomedicine and biomedical engineering like e-health, diagnosis or biology systems but also statistics and computer science like cloud computing, big data or mathematical modelling. Thus, the first paper authored by Gimenez et al. [3] describes a mathematical implementation of non-uniform flow models to simulate tilted holes and conical holes in the obstruction of ventricular catheters. Although these simulations have been developed in an ideal scenario, they provide promising results about the consequences of variations in diameters as well as tilt angles of the holes. This novel design improves the disadvantage of having too small holes which appear in other commercially available implementations. The article by Gamberger el al. [4] proposes a novel clustering algorithm which is able to determine small and heterogeneous subpopulations in Alzheimer’s disease (AD) patients. This method, called multilayer clustering, defines each clinical or biological descriptor as a “layer”, building clusters based on the co-existence of properties. This methodology was evaluated with a dataset with 33 properties and 916 patients with significant problems with dementia. Even though the methodology should be validated in other domains, it has shown the successful identification of subpopulations in the AD domain. Abbasi et al. [5] address in their article one traditional but still challenging problem in bioinformatics, the multiple sequence alignment (MSA). In this case, authors introduce a novel heuristic solution based on local search and an optimization based on the number of indels and a substitution score (multiobjective approach). From these bases, several implementations and configurations are proposed and compared including perturbation techniques to avoid suboptimal solutions due to local maximums. Taking into consideration the outcomes, this study has shown a significant improvement in alignments in a reasonable time. Following, the paper by Molina-Recio et al. [6] analyzes the effect of multidisciplinary research when developing mobile applications based on eHealth (also called mHealth). This systematic review highlights the exponential increase of research publications in this area and the importance of interdisciplinary teams involving more health professionals, whose contribution is currently lacking from. Nevertheless, authors of this study also conclude that the impact in the analyzed publications could be not necessarily correlated with the presence of multidisciplinary teams. In the next article, Macdonald et al. [7] presents an adaptive gradient matching approach based on Gaussian processes to estimate parameters in biological pathways, which are habitually expressed as pairs of ordinary differential equations (ODEs). This contribution is mainly based on a mathematical and numerical modelling but with a clearly promising application in the inference of system biology parameters. In fact, according to the provided results, the proposed approach outperforms other similar solutions when estimating parameters in two benchmarks to describe the voltage potential across the cell membrane and the protein signaling transduction pathway, respectively. The paper by Cisar et al. [8] describes BioWes, a complete and well-designed repository to store and to manage experimental descriptors and data. This tool is specifically designed taking into account the importance of “Big Data” and how dealing with it. With this purpose, it includes implementations for desktop application, web interface and web-based interface mobile devices. BioWes has been validated in the context of the AQUAEXCEL international infrastructure project in order to organize and store data from aquaculture experiments. The next contribution by Ortega et al. [9] analyzes different feature selection techniques in multi-resolution analysis (MRA) to classify electroencephalographic (EEG) images for brain-computer interfaces (BCIs). Several feature selection approaches are compared based on evolutionary multi-objective techniques and different structures of classifiers. Authors statistically probe that the proposed feature selection algorithms outperform the baseline MRA in terms of computational costs given the reduction in the number of features, providing similar or better classification performances. Subsequently, Banos et al. [10] present in their publication a system for personalized medicine, healthcare and wellness based on a new digital framework called Mining Minds. The proposed system exploits the advantages of cloud computing storage and high performance computing but with the novelty of incorporating wearable technology and Big Data. This framework addresses an efficient, flexible and robust design which makes it a promising and potential tool for other clinical developments and applications. The article from Larriba et al. [11] proposes a novel baby dinosaur robotic pet, called Pleo, to reduce pain and anxiety in hospitalized children. Simultaneously, this therapeutic pet is wireless connected to assist clinical personal to monitor and understand the child behavior through an Android app. This article describes the whole architecture of Pleo robot in detail but it also shows the successful results obtained from preliminary tests with children in hospital. Finally, Soria-Morillo et al. [12] describe a novel discrete classification algorithm to determine human reactions on TV advertisements from low-cost EEG signals. This algorithm is presented by integrating signal collection/transformation stages with different supervised machine learning approaches like C4.5, ANN or an own-designed discretization procedure. After validating and comparing these classifier’s performances, authors showed their approach clearly outperforms other tools in terms of accuracy and power consumption. As Guest editors, we would like to express our thankfulness to all the authors contributing with their high quality researches to the achievement of this supplement. Also, we are very grateful to expert scientists that have actively collaborated with their recommendations and suggestions to review and improve these contributions. We would like to sincerely acknowledge Profs. Kenneth R. Foster and Fong-Chin Su, Editors-in-Chief of BioMedical Engineering Online, for giving again the opportunity to publish this supplement in this relevant journal. We specially thank to Mr. Omar El Bakry for his excellent and constant support with the publication and edition of this supplement. It has been an honor for us to participate in it. We finally invite authors and readers of this supplement to submit their recent works to future editions of IWBBIO, which will be announced at http://iwbbio.ugr.es.

The analysis of daily living human behavior has proven to be of key importance to prevent unhealthy habits. The diversity of activities and the individuals' particular execution style determine that several sources of information are normally required. One of the main issues is to optimally combine them to guarantee performance, scalability and robustness. In this paper we present a fusion classification methodology which takes into account the potential of the individual decisions yielded at both activity and sensor classification levels. Particularly tested on a wearable sensors based system, the method reinforces the idea that some parts of the body (i.e., sensors) may be specially informative for the recognition of each particular activity, thus supporting the ranking of the decisions provided by each associated sensor decision entity. Our method systematically outperforms the results obtained by traditional multiclass models which otherwise may require a high-dimensional feature space to acquire a similar performance. The comparison with other activity-recognition fusion approaches also demonstrates our model scales significantly better for small sensor networks.

The visible/near-infrared spectrum consists of overtones and combination bands of the fundamental molecular absorptions found in the visible and near-infrared region. The analysis of the spectrum might be difficult because overlapping vibrational bands may appear nonspecific and poorly resolved. Nevertheless, the information it could be retrieved from the analysis of the spectrum might be very useful for the food industry producers, consumers, and food distributors because the meat could be classified based on the spectrum in several aspects such as the quality, tenderness, and kind of meat. This paper applies Mutual Information theory and several classification models (Radial Basis Function Neural Networks and Support Vector Machines) in order to determine the breed of pork meat (Iberian or White) using only as input the infrared spectrum. First, the more relevant wavelengths from the spectrum will be chosen, then, those wavelengths will be the input data to design the classifiers. As the experiments will show, the proposed techniques, when applied with a correct design methodology are capable of obtaining quality results for this specific problem.

The function approximation problem has been tackled many times in the literature by using radial basis function neural networks (RBFNNs). In the design of these neural networks there are several stages where, the most critical stage is the initialization of the centers of each RBF since the rest of the steps to design the RBFNN strongly depend on it. The improved clustering for function approximation (ICFA) algorithm was recently introduced and proved successful for the function approximation problem. In the ICFA algorithm, a fuzzy partition of the input data is performed but, a fuzzy partition can behave inadequately in noise conditions. Possibilistic and mixed approaches, combining fuzzy and possibilistic partitions, were developed in order to improve the performance of a fuzzy partition. In this paper, a study of the influence of replacing the fuzzy partition used in the ICFA algorithm with the possibilistic and the fuzzy-possibilistic partitions will be done. A comparative analysis of each kind of partition will be performed in order to see if the possibilistic approach can improve the performance of the ICFA algorithm both in normal and in noise conditions. The results will show how the employment of a mixed approach combining fuzzy and possibilistic approach can lead to improve the results when designing RBFNNs.

The use of Radial Basis Function Neural Networks (RBFNNs) to solve functional approximation problems has been addressed many times in the literature. When designing an RBFNN to approximate a function, the first step consists of the initialization of the centers of the RBFs. This initialization task is very important because the rest of the steps are based on the positions of the centers. Many clustering techniques have been applied for this purpose achieving good results although they were constrained to the clustering problem. The next step of the design of an RBFNN, which is also very important, is the initialization of the radii for each RBF. There are few heuristics that are used for this problem and none of them use the information provided by the output of the function, but only the centers or the input vectors positions are considered. In this paper, a new algorithm to initialize the centers and the radii of an RBFNN is proposed. This algorithm uses the perspective of activation grades for each neuron, placing the centers according to the output of the target function. The radii are initialized using the center's positions and their activation grades so the calculation of the radii also uses the information provided by the output of the target function. As the experiments show, the performance of the new algorithm outperforms other algorithms previously used for this problem.

In this paper, a temperature control in real time control process was presented using several control algorithms. A quantitative comparison based on the real power consumption and (the precision and the robustness) of these controllers during the same control process and under the same conditions will be done. The proposed Adaptive and Self Organizing Fuzzy policy has been able to prove its superiority against the remaining controllers. The new Adaptive and Self Organizing Fuzzy Logic Controller starts the control with a very limited information about the controlled process (delay and the monotonicity sign) and without any kind of offline pre-training, the adaptive controller acts online to collect the necessary background to adapt their rules consequents and to self organize their membership functions from the real behavior of the controlled process. During 200 minutes and under the same conditions all the performed controllers have been used to control the room temperature, each simulation has been repeated five times with two different sets of set points. These amounts of results was used as a set of sampling for the statistical tool ANOVA (Analysis of Variance) that can prove and illustrate the validity and the extrapolability of the conclusions extracted from several stages of this work.