Network


Latest external collaboration on country level. Dive into details by clicking on the dots.

Hotspot


Dive into the research topics where Marcelino Martínez-Sober is active.

Publication


Featured researches published by Marcelino Martínez-Sober.


Artificial Intelligence in Medicine | 2004

Foetal ECG recovery using dynamic neural networks

Gustavo Camps-Valls; Marcelino Martínez-Sober; Emilio Soria-Olivas; Rafael Magdalena-Benedito; Javier Calpe-Maravilla; Juan Guerrero-Martínez

Non-invasive electrocardiography has proven to be a very interesting method for obtaining information about the foetus state and thus to assure its well-being during pregnancy. One of the main applications in this field is foetal electrocardiogram (ECG) recovery by means of automatic methods. Evident problems found in the literature are the limited number of available registers, the lack of performance indicators, and the limited use of non-linear adaptive methods. In order to circumvent these problems, we first introduce the generation of synthetic registers and discuss the influence of different kinds of noise to the modelling. Second, a method which is based on numerical (correlation coefficient) and statistical (analysis of variance, ANOVA) measures allows us to select the best recovery model. Finally, finite impulse response (FIR) and gamma neural networks are included in the adaptive noise cancellation (ANC) scheme in order to provide highly non-linear, dynamic capabilities to the recovery model. Neural networks are benchmarked with classical adaptive methods such as the least mean squares (LMS) and the normalized LMS (NLMS) algorithms in simulated and real registers and some conclusions are drawn. For synthetic registers, the most determinant factor in the identification of the models is the foetal-maternal signal-to-noise ratio (SNR). In addition, as the electromyogram contribution becomes more relevant, neural networks clearly outperform the LMS-based algorithm. From the ANOVA test, we found statistical differences between LMS-based models and neural models when complex situations (high foetal-maternal and foetal-noise SNRs) were present. These conclusions were confirmed after doing robustness tests on synthetic registers, visual inspection of the recovered signals and calculation of the recognition rates of foetal R-peaks for real situations. Finally, the best compromise between model complexity and outcomes was provided by the FIR neural network. Both the methodology for selecting a model and the introduction of advanced neural models are the main contributions of this paper.


Expert Systems With Applications | 2012

Detecting rottenness caused by Penicillium genus fungi in citrus fruits using machine learning techniques

Juan Gómez-Sanchis; José David Martín-Guerrero; Emilio Soria-Olivas; Marcelino Martínez-Sober; Rafael Magdalena-Benedito; José Blasco

Penicillium fungi are among the main defects that may affect the commercialization of citrus fruits. Economic losses in fruit production may become enormous if an early detection of that kind of fungi is not carried out. That early detection is usually based either on UltraViolet light carried out manually. This work presents a new approach based on hyperspectral imagery for defect segmentation. Both the physical device and the data processing (geometric corrections and band selection) are presented. Achieved results using classifiers based on Artificial Neural Networks and Decision Trees show an accuracy around 98%; it shows up the suitability of the proposed approach.


Pacing and Clinical Electrophysiology | 1998

Quantification of the Modifications in the Dominant Frequency of Ventricular Fibrillation under Conditions of Ischemia and Reperfusion: An Experimental Study

Francisco J. Chorro; Juan Guerrero; Joaquín Cánoves; Marcelino Martínez-Sober; Luis Mainar; Juan Sanchis; J. Calpe; Esteban Llavador; José M. Espí; Vicente López-Merino

The characteristics of ventricular fibrillatory signals vary as a function of the time elapsed from the onset of arrhythmia and the maneuvers used to maintain coronary perfusion. The dominant frequency (FrD) of the power spectrum of ventricular fibrillation (VF) is known to decrease after interrupting coronary perfusion, though the corresponding recovery process upon reestablishing coronary flow has not been quantified to date. With the aim of investigating the recovery of the FrD during reperfusion after a brief ischemic, period, 11 isolated and perfused rabbit heart preparations were used to analyze the signals obtained with three unipolar epicardial electrodes (E1‐E3) and a bipolar electrode immersed in the thermostatizfid organ bath (E4), following the electrical induction of VF. Recordings were made under conditions of maintained coronary perfusion (5 min), upon interrupting perfusion (15 mini, and after reperfusion (5 min), FrD was determined using Welchs method. The variations in FrD were quantified during both ischemia and reperfusion, based on an exponential model AFrD = A exp (‐t/C). During ischemia ΔFrD is the difference between FrD and the minimum value, while t is the time elapsed from the interruption of coronary perfusion. During reperfusion ΔFrD is the difference between the maximum value and FrD, while t is the time elapsed from the restoration of perfusion, A is one of the constants of the model, and C is the time constant. FrD exhibited respective initial values of 16.20 ± 1.67, 16.03 ± 1.38, and 16.03 ± 1.80 Hz in the epicardial leads, and 15.09 ±1.07 Hz in the bipolar lead within the bath. No significant variations were observed during maintained coronary perfusion. The fit of the FrD variations to the model during ischemia and reperfusion proved significant in nine experiments. The mean time constants C obtained on fitting to the model during ischemia were as follows: El =294.4 ± 75.6, E2 = 225.7 ± 48.5, E3 = 327.4 ± 79.7, and E4 = 298.7 ± 43.9 seconds. The mean values of C obtained during reperfusion, and the significance of the differences with respect to the ischemic period were: El = 57.5 ± 8.4 (P ± 0.01), E2 = 64.5 ± 11.2 (P0.01), E3 = 80.7 ± 13.3 (P < 0.01), and E4 = 74.9 ± 13.6 (P < 0.0001). The time course variations of the FrD of the VF power spectrum fit an exponential model during ischemia and reperfusion. The time constants of the model during reperfusion after a brief ischemic period are significantly shorter than those obtained during ischemia.


Expert Systems With Applications | 2013

Expert system for predicting unstable angina based on Bayesian networks

Joan Vila-Francés; Juan Sanchis; Emilio Soria-Olivas; Antonio J. Serrano; Marcelino Martínez-Sober; Clara Bonanad; Silvia Ventura

The use of computer-based clinical decision support (CDS) tools is growing significantly in recent years. These tools help reduce waiting lists, minimise patient risks and, at the same time, optimise the cost health resources. In this paper, we present a CDS application that predicts the probability of having unstable angina based on clinical data. Due to the characteristics of the variables (mostly binary) a Bayesian network model was chosen to support the system. Bayesian-network model was constructed using a population of 1164 patients, and subsequently was validated with a population of 103 patients. The validation results, with a negative predictive value (NPV) of 91%, demonstrate its applicability to help clinicians. The final model was implemented as a web application that is currently been validated by clinician specialists.


IEEE Transactions on Education | 1998

An easy demonstration of the optimum value of the adaptation constant in the LMS algorithm [FIR filter theory]

E. Soria-Olivas; Javier Calpe-Maravilla; Juan Guerrero-Martínez; Marcelino Martínez-Sober; J. Espi-Lopez

Since the introduction of the LMS algorithm, many variants have been proposed to improve its performance. Doubltless, the most popular is the Normalized LMS, which uses a value for the adaptation constant that assures the fastest convergence. This correspondence shows a new demonstration of the algorithm based on a mathematical approach easier than the usually proposed.


Computer Methods and Programs in Biomedicine | 2014

Prediction of the hemoglobin level in hemodialysis patients using machine learning techniques

José María Martínez-Martínez; Pablo Escandell-Montero; Carlo Barbieri; Emilio Soria-Olivas; Flavio Mari; Marcelino Martínez-Sober; Claudia Amato; Antonio López; Marcello Bassi; Rafael Magdalena-Benedito; Andrea Stopper; José David Martín-Guerrero; Emanuele Gatti

Patients who suffer from chronic renal failure (CRF) tend to suffer from an associated anemia as well. Therefore, it is essential to know the hemoglobin (Hb) levels in these patients. The aim of this paper is to predict the hemoglobin (Hb) value using a database of European hemodialysis patients provided by Fresenius Medical Care (FMC) for improving the treatment of this kind of patients. For the prediction of Hb, both analytical measurements and medication dosage of patients suffering from chronic renal failure (CRF) are used. Two kinds of models were trained, global and local models. In the case of local models, clustering techniques based on hierarchical approaches and the adaptive resonance theory (ART) were used as a first step, and then, a different predictor was used for each obtained cluster. Different global models have been applied to the dataset such as Linear Models, Artificial Neural Networks (ANNs), Support Vector Machines (SVM) and Regression Trees among others. Also a relevance analysis has been carried out for each predictor model, thus finding those features that are most relevant for the given prediction.


Archive | 2012

Intelligent Data Analysis for Real-Life Applications: Theory and Practice

Rafael Magdalena-Benedito; Marcelino Martínez-Sober; José María Martínez-Martínez; Joan Vila-Francés; Pablo Escandell-Montero

Prompted by crater counts as the only available tool for measuring remotely the relative ages of geologic formations on planets, advances in remote sensing have produced a very large database of high resolution planetary images, opening up an opportunity to survey much more numerous small craters improving the spatial and temporal resolution of stratigraphy. Automating the process of crater detection is key to generate comprehensive surveys of smaller craters. Here, the authors discuss two supervised machine learning techniques for crater detection algorithms (CDA): identification of craters from digital elevation models (also known as range images), and identification of craters from panchromatic images. They present applications of both techniques and demonstrate how such automated analysis has produced new knowledge about planet Mars. DOI: 10.4018/978-1-4666-1806-0.ch008


Computer Applications in Engineering Education | 2010

Description and evaluation of an introductory course to Matlab for a heterogeneous group of university students

Emilio Soria-Olivas; José David Martín-Guerrero; Marcelino Martínez-Sober; Amparo Ayuso-Moya

This paper presents the experience of the authors in teaching the Matlab™ software to students with different levels of academic training, even though all of them belong to the field of Science. Matlab™ is a worldwide standard in technical programming and computing, and its use is still growing; therefore, it is taught in many different university degree programs (Physics, Mathematics, Engineering, etc.). The course analyzed in this paper contains examples of different fields of knowledge to show the students the versatility and power of this tool. At the end of each course, which consists of 30 h, the students expressed their opinions about the content in a private opinion poll (questionnaire). The results of these opinion polls show that the methodology followed is considered appropriate and useful by the students who attended the course.


Computer Methods and Programs in Biomedicine | 2013

Matlab-based interface for the simultaneous acquisition of force measures and Doppler ultrasound muscular images

José Ferrer-Buedo; Marcelino Martínez-Sober; Yasser Alakhdar-Mohmara; Emilio Soria-Olivas; Josep Carles Benítez-Martínez; José María Martínez-Martínez

This paper tackles the design of a graphical user interface (GUI) based on Matlab (MathWorks Inc., MA), a worldwide standard in the processing of biosignals, which allows the acquisition of muscular force signals and images from a ultrasound scanner simultaneously. Thus, it is possible to unify two key magnitudes for analyzing the evolution of muscular injuries: the force exerted by the muscle and section/length of the muscle when such force is exerted. This paper describes the modules developed to finally show its applicability with a case study to analyze the functioning capacity of the shoulder rotator cuff.


international conference on data mining | 2016

Machine Learning for Modeling the Biomechanical Behavior of Human Soft Tissue

José David Martín-Guerrero; María J. Rupérez-Moreno; Francisco Martínez-Martínez; Delia Lorente-Garrido; Antonio J. Serrano-López; C. Monserrat; S. Martínez-Sanchis; Marcelino Martínez-Sober

An accurate modeling of the biomechanical properties of human soft tissue is crucial in many clinical applications, such as, radiotherapy administration or surgery. The finite element method (FEM) is the usual choice to carry out such modeling due to its high accuracy. However, FEM is computationally very costly, and hence, its application in real-time or even off-line with short delays are still challenges to overcome. This paper proposes a framework based on Machine Learning to learn FEM modeling, thus having a tool able to yield results that may be sufficiently fast for clinical applications. In particular, the use of ensembles of Decision Trees has shown its suitability in modeling the behavior of the liver and the breast.

Collaboration


Dive into the Marcelino Martínez-Sober's collaboration.

Top Co-Authors

Avatar
Top Co-Authors

Avatar
Top Co-Authors

Avatar
Top Co-Authors

Avatar
Top Co-Authors

Avatar
Top Co-Authors

Avatar
Top Co-Authors

Avatar
Top Co-Authors

Avatar

C. Monserrat

Polytechnic University of Valencia

View shared research outputs
Top Co-Authors

Avatar

D. Lorente

University of Valencia

View shared research outputs
Top Co-Authors

Avatar
Researchain Logo
Decentralizing Knowledge