Papers by Antonio Peñalver Benavent
In this work, we address the problem of estimating the parameters of a Gaussian mixture model. Al... more In this work, we address the problem of estimating the parameters of a Gaussian mixture model. Although the standard Expectation-Maximization (EM) algorithm yields the maximum-likelihood solution, it is well-known that it is prone to the selection of the starting parameters of the model and it may converge to the boundary of the parameter space. Usually, some approaches like k-means are used to set the starting values of the model; however, only the convergence speed of the algorithm to a local maxima is increased because these approaches are local too, and a global maximum is not ensured in any case. Furthermore, the resulting mixture depends on the number of selected components, but the optimal number of kernels in the mixture may be unknown beforehand. Bibliografía Capítulo 1 Introducción y objetivos En este capítulo realizaremos una visión general de la tesis. Comenzaremos presentando los modelos de mezclas finitas y en particular los modelos de mezclas que emplean núcleos gausianos, las diferentes áreas en las que su aplicación es de especial interés y los distintos métodos existentes para ajustar adecuadamente el modelo a la resolución de un problema particular. Finalizaremos el capítulo con una descripción de los objetivos del trabajo así como con un esquema general de funcionamiento de la técnica propuesta.
Development of a XML-based Ubiquitous System for Testing Using Smartphones
In this paper, authors introduce a novelty XML-based m-learning author tool running on Windows mo... more In this paper, authors introduce a novelty XML-based m-learning author tool running on Windows mobile-based smartphones which allows students to take different stored test. This mobile-based application runs off-line and only connects through Internet whenever should be required. Initial performance was evaluated requesting user feed-back through informal interview, as a result a summarize evaluation of usability, user performance and behaviour
The measurement or evaluation and clinical significance of human sperm morphology has always been... more The measurement or evaluation and clinical significance of human sperm morphology has always been and still is a controversial aspect of the semen analysis for the determination of a male's fertility potential. The evaluation of sperm size, shape and morphological smear characteristics should be assesed by carefully observing a stained sperm sample under a microscope. In order to avoid subjectivity, numerous studies that incorporate image analysis techniques in the assessment of sperm morphology have been proposed. The primary step of all these methods is segmentation of sperm's parts. In this paper, we have proposed a new method for segmentation of sperm's Acrosome, Nucleus and Mid-piece. Sperm's Acrosome, Nucleus and Midpiece are segmented through a method based on a Bayesian classifier which utilizes the adaptive mixtures method (AMM) and Markov random field (MRF) model to obtain and upgrade the class conditional probability density function (CCPDF) and the probability of each class. To compare the performance of the proposed approach with those of previous approaches including manual segmentation, the Accuracy, Sensitivity and Specificity were calculated.
Learning Gaussian Mixture Models With Entropy-Based Criteria
IEEE Transactions on Neural Networks, Nov 1, 2009
In this paper, we address the problem of estimating the parameters of Gaussian mixture models. Al... more In this paper, we address the problem of estimating the parameters of Gaussian mixture models. Although the expectation-maximization (EM) algorithm yields the maximum-likelihood (ML) solution, its sensitivity to the selection of the starting parameters is well-known and it may converge to the boundary of the parameter space. Furthermore, the resulting mixture depends on the number of selected components, but the optimal number of kernels may be unknown beforehand. We introduce the use of the entropy of the probability density function (pdf) associated to each kernel to measure the quality of a given mixture model with a fixed number of kernels. We propose two methods to approximate the entropy of each kernel and a modification of the classical EM algorithm in order to find the optimum number of components of the mixture. Moreover, we use two stopping criteria: a novel global mixture entropy-based criterion called Gaussianity deficiency (GD) and a minimum description length (MDL) principle-based one. Our algorithm, called entropy-based EM (EBEM), starts with a unique kernel and performs only splitting by selecting the worst kernel attending to GD. We have successfully tested it in probability density estimation, pattern classification, and color image segmentation. Experimental results improve the ones of other state-of-the-art model order selection methods.
Segmentation is an important step for the diagnosis of multiple sclerosis. This paper presents a ... more Segmentation is an important step for the diagnosis of multiple sclerosis. This paper presents a new approach for fully automatic segmentation of MS lesions in fluid attenuated inversion recovery (FLAIR) Magnetic Resonance (MR) images. At first, Brain image is considered to be three parts, namely the dark, the gray, and the white part. Then, the fuzzy regions of their member functions are determined by maximizing fuzzy entropy through the Genetic algorithm. Finally, MS lesions and CSF areas are determined by applying a localized weighted filter to the Bright and Dark membership images. To evaluate the result of the proposed method, similarity criteria of different slices related to 20 MS patients are calculated and compared with other methods which include manual segmentation. Also, volume of segmented lesions are computed and compared with Gold standard using correlation coefficient. The proposed method has better performance in comparison with previous works which are reported here.
Pedagogical Use of Tablet PC for Active and Collaborative Learning
In this paper we present our experience using tablet PC to support the pedagogical needs of the e... more In this paper we present our experience using tablet PC to support the pedagogical needs of the engineering classroom as well as typical engineering group collaborative environments. We applied this novel active and collaborative learning framework to undergraduate students from a fundamental database course in order to analyze the role of pedagogical use of tablet PC. Then, this educational test
Journal of Biomedical Science and Engineering, 2011
This paper presents an approach for fully automatic segmentation of MS lesions in fluid attenuate... more This paper presents an approach for fully automatic segmentation of MS lesions in fluid attenuated inversion recovery (FLAIR) Magnetic Resonance (MR) images. The proposed method estimates a gaussian mixture model with three kernels as cerebrospinal fluid (CSF), normal tissue and Multiple Sclerosis lesions. To estimate this model, an automatic Entropy based EM algorithm is used to find the best estimated Model. Then, Markov random field (MRF) model and EM algorithm are utilized to obtain and upgrade the class conditional probability density function and the apriori probability of each class. After estimation of Model parameters and apriori probability, brain tissues are classified using bayesian classification. To evaluate the result of the proposed method, similarity criteria of different slices related to 20 MS patients are calculated and compared with other methods which include manual segmentation. Also, volume of segmented lesions are computed and compared with gold standard using correlation coefficient. The proposed method has better performance in comparison with previous works which are reported here.
The use of real time multimedia in education has been greatly improved thanks to the advance of I... more The use of real time multimedia in education has been greatly improved thanks to the advance of Internet communication, e.g., wired or even also wireless bandwidth links. This constant progress enhances the implementation of many real time applications like collaborative and tele-teaching applications. In this work, authors present design and implementation issues for Tele-ed, i.e., an author's tool for video content delivery which concerns an Internet based distant learning application. Video content delivery focuses on multicast transmission which has been used to develop successfully Tele-ed.
PLOS ONE, Jun 17, 2013
Segmentation is an important step for the diagnosis of multiple sclerosis (MS). This paper presen... more Segmentation is an important step for the diagnosis of multiple sclerosis (MS). This paper presents a new approach to the fully automatic segmentation of MS lesions in Fluid Attenuated Inversion Recovery (FLAIR) Magnetic Resonance (MR) images. With the aim of increasing the contrast of the FLAIR MR images with respect to the MS lesions, the proposed method first estimates the fuzzy memberships of brain tissues (i.e., the cerebrospinal fluid (CSF), the normal-appearing brain tissue (NABT), and the lesion). The procedure for determining the fuzzy regions of their member functions is performed by maximizing fuzzy entropy through Genetic Algorithm. Research shows that the intersection points of the obtained membership functions are not accurate enough to segment brain tissues. Then, by extracting the structural similarity (SSIM) indices between the FLAIR MR image and its lesions membership image, a new contrast-enhanced image is created in which MS lesions have high contrast against other tissues. Finally, the new contrast-enhanced image is used to segment MS lesions. To evaluate the result of the proposed method, similarity criteria from all slices from 20 MS patients are calculated and compared with other methods, which include manual segmentation. The volume of segmented lesions is also computed and compared with Gold standard using the Intraclass Correlation Coefficient (ICC) and paired samples t test. Similarity index for the patients with small lesion load, moderate lesion load and large lesion load was 0.7261, 0.7745 and 0.8231, respectively. The average overall similarity index for all patients is 0.7649. The t test result indicates that there is no statistically significant difference between the automatic and manual segmentation. The validated results show that this approach is very promising.
Journal of Biomedical Science and Engineering, 2012
In the last years, digital image processing and analysis are used for computer assisted evaluatio... more In the last years, digital image processing and analysis are used for computer assisted evaluation of semen quality with therapeutic goals or to estimate its fertility by means of spermatozoid motility and morphology. Sperm morphology is assessed routinely as part of standard laboratory analysis in the diagnosis of human male infertility. Nowadays assessments of sperm morphology are mostly done based on subjective criteria. In order to avoid subjectivity, numerous studies that incorporate image analysis techniques in the assessment of sperm morphology have been proposed. The primary step of all these methods is segmentation of sperm's parts. In this paper, we have proposed a new method for segmentation of sperm's Acrosome, Nucleus, Mid-piece and identification of sperm's tail through some points which are placed on the sperm's tail, accurately. These estimated points could be used to verify the morphological characteristics of sperm's tail such as length, shape and etc. At first, sperm's Acrosome, Nucleus and Mid-piece are segmented through a method based on a Bayesian classifier which utilizes the entropy based expectationmaximization (EM) algorithm and Markov random field (MRF) model to obtain and upgrade the class conditional probability density function (CCPDF) and the apriori probability of each class. Then, a pixel at the end of sperm's Mid-piece, is selected as an initial point. To find other pixels which are placed on the sperm's tail, structural similarity index (SSIM) is used in an iterative scheme. In order to stop the algorithm automatically at the end of sperm's tail, local entropy is estimated and used as a feature to determine if a point is located on the sperm's tail or not. To compare the performance of the proposed approach with those of previous approaches including manual segmentation, the Accuracy, Sensitivity and Specificity were calculated.
In this paper we address the problem of estimating the parameters of a Gaussian mixture model. Al... more In this paper we address the problem of estimating the parameters of a Gaussian mixture model. Although the EM algorithm yields the maximum-likelihood solution it requires a careful initialization of the parameters and the optimal number of kernels in the mixture may be unknown beforehand. We propose a criterion based on the entropy of the pdf (probability density function) associated to each kernel to measure the quality of a given mixture model. A novel method for estimating Shannon entropy based on Entropic Spanning Graphs is developed and a modification of the classical EM algorithm to find the optimal number of kernels in the mixture is presented. We test our algorithm in probability density estimation, pattern recognition and color image segmentation.
Work in Progress - Pedagogical use of Video Streaming
New real-time multimedia technologies have brought on many changes in activities, content distrib... more New real-time multimedia technologies have brought on many changes in activities, content distribution, strategies, and attitudes in the field of continuing education. In this Work in Progress (WIP), we introduce our technology innovation based on a distance learning framework. Our aim is to provide a way of understating the role of pedagogical use of video streaming and as it changes
Abstract- Over last two decades, new real-time multimedia technologies have brought on many chang... more Abstract- Over last two decades, new real-time multimedia technologies have brought on many changes in pedagogical strategies through the field of e-learning applications. Today, it’s well-known the emergence of devices like Tablet PC which support collaborative learning in ways never envisioned for many decades. In this paper we present our experience using Classroom Presenter as available module inside ConferenceXP research platform. Main goal dealt with exploring how to make collaborative learning a compelling and rich experience. Thus, we applied this collaborative learning framework to undergraduate students from an engineering course in order to analyze the role of pedagogical use of this module inside ConferenceXP platform. Afterwards, this educational experiment was favorably received by students and conclusions deal with (i) getting student assessment about their experiences in this experimental course, (ii) promoting use of next-generation pedagogical collaborative learnin...
e-dap: An e-learning tool for Managing, Distributing and Capturing Knowledge
Learning Gaussian Mixture Models With
Ieee Transactions on Neural Networks, 2009
Journal of Biomedical Science and Engineering, 2021
Recovering from multiple traumatic brain injury (TBI) is a very difficult task, depending on the ... more Recovering from multiple traumatic brain injury (TBI) is a very difficult task, depending on the severity of the lesions, the affected parts of the brain and the level of damage (locomotor, cognitive or sensory). Although there are some software platforms to help these patients to recover part of the lost capacity, the variety of existing lesions and the different degree to which they affect the patient, do not allow the generalization of the appropriate treatments and tools in each case. The aim of this work is to design and evaluate a machine vision-based UI (User Interface) allowing patients with a high level of injury to interact with a computer. This UI will be a tool for the therapy they follow and a way to communicate with their environment. The interface provides a set of specific activities, developed in collaboration with the multidisciplinary team that is currently evaluating each patient, to be used as a part of the therapy they receive. The system has been successfully tested with two patients whose degree of disability prevents them from using other types of platforms.
Structural Change and Economic Dynamics, 2020
We assess the effects of monetary policy shocks on income and wealth inequality through direct in... more We assess the effects of monetary policy shocks on income and wealth inequality through direct inequality measures and by analyzing several transmission channels explored in recent literature. Furthermore, we analyze two additional channels: the Housing and the Fiscal channels. The methodology adopted is a Bayesian proxy SVAR using a high-frequency identification based on the external instruments approach. Our own policy shocks are constructed for this purpose. The results show that an expansionary monetary policy shock does not have a significant effect on income inequality due to the existence of opposite channels, whereas it increases wealth inequality mainly through the portfolio channel.
Proceedings of the 13th International Conference on Interaccion Persona-Ordenador (Interaccion'12), 2012
Objective: To report a rare complication of homocystinuria in a child and highlight the associati... more Objective: To report a rare complication of homocystinuria in a child and highlight the association of homocystinuria with lower gastrointestinal bleeding and intestinal thrombosis. Clinical Presentation and Intervention: A 7-year-old boy with homocystinuria and poor compliance with treatment presented with abdominal pain and bloody stools. Doppler ultrasound showed superior mesenteric and middle colic vein thrombosis. Laparotomy demonstrated ischemic small bowel necessitating resection. The patient improved clinically following resection and the initiation of the anticoagulation with homocystinuria treatment and was discharged home. Conclusion: This case showed the need to sustain a high index of suspicion of thromboembolism in a patient with homocystinuria and lower gastrointestinal bleeding and the importance of compliance with treatment to avoid vascular complications.
Human-Computer Interaction: Interaction Technologies, 2015
Nowadays, the advent of mobile technologies with increasing functionality and computing power is ... more Nowadays, the advent of mobile technologies with increasing functionality and computing power is changing the way people interact with their applications in more and more different contexts of use. This way, many traditional user interfaces are evolving towards "distributed" user ones, allowing that interaction elements can now be distributed among heterogeneous devices from different platforms. In this paper we present an HTTP-Based framework for generating and distributing UIs (User Interfaces) of custom applications, allowing device change with state preservation. We use a schema-based definition of DUIs (Distributed User Interfaces), allowing the specification of the elements to be distributed. The framework is based on open standards and supports any markup-based web language. We provide a graphic case of use implemented in HTML5.
Estimating Simultaneous Equation Models through an Entropy-Based Incremental Variational Bayes Learning Algorithm
Entropy, 2021
The presence of unaccounted heterogeneity in simultaneous equation models (SEMs) is frequently pr... more The presence of unaccounted heterogeneity in simultaneous equation models (SEMs) is frequently problematic in many real-life applications. Under the usual assumption of homogeneity, the model can be seriously misspecified, and it can potentially induce an important bias in the parameter estimates. This paper focuses on SEMs in which data are heterogeneous and tend to form clustering structures in the endogenous-variable dataset. Because the identification of different clusters is not straightforward, a two-step strategy that first forms groups among the endogenous observations and then uses the standard simultaneous equation scheme is provided. Methodologically, the proposed approach is based on a variational Bayes learning algorithm and does not need to be executed for varying numbers of groups in order to identify the one that adequately fits the data. We describe the statistical theory, evaluate the performance of the suggested algorithm by using simulated data, and apply the two...
Uploads
Papers by Antonio Peñalver Benavent