Procesamiento de imágenes biológicas

Este proyecto se ha desarrollado a petición del CSIC. El proyecto consiste en crear una aplicación que automatice la localización y obtención de la intensidad de unos puntos luminosos en las células que aparecen en el vídeo. También se desea identificar en qué estadio celular se encuentran las células y si los puntos hallados están en la zona del septim ring. Para ello, se ha hecho un estudio y probado dos métodos para la localización de los puntos y se ha realizado un filtro mediante Adaboost para la localización de las células. También se ha realizado una interfaz gráfica para el usuario final.

Regiones salientes complejas para aplicaciones de seguimiento facial

Este trabajo presenta una metodología para detectar y realizar el seguimiento de características faciales. En el primer paso del procedimiento se detectan caras mediante Adaboost con cascadas de clasificadores débiles. El segundo paso busca las características internas de la cara mediante el CSR, detectando zonas de interés. Una vez que estas características se capturan, un proceso de tracking basado en el descriptor SIFT, que hemos llamado pseudo-SIFT, es capaz de guardar información sobre la evolución de movimiento en las regiones detectadas. Además, un conjunto de datos públicos ha sido desarrollado con el propósito de compartirlo con otras investigaciones sobre detección, clasificación y tracking. Experimentos reales muestran la robustez de este trabajo y su adaptabilidad para trabajos futuros.

Reconocimiento de objetos multi-clase basado en descriptores de forma

Este trabajo presenta un sistema para detectar y clasificar objetos binarios según la forma de éstos. En el primer paso del procedimiento, se aplica un filtrado para extraer el contorno del objeto. Con la información de los puntos de forma se obtiene un descriptor BSM con características altamente descriptivas, universales e invariantes. En la segunda fase del sistema se aprende y se clasifica la información del descriptor mediante Adaboost y Códigos Correctores de Errores. Se han usado bases de datos públicas, tanto en escala de grises como en color, para validar la implementación del sistema diseñado. Además, el sistema emplea una interfaz interactiva en la que diferentes métodos de procesamiento de imágenes pueden ser aplicados.

A Robust Multiple Feature Approach To Endpoint Detection In Car Environment Based On Advanced Classifiers

In this paper we propose an endpoint detection system based on the use of several features extracted from each speech frame, followed by a robust classifier (i.e Adaboost and Bagging of decision trees, and a multilayer perceptron) and a finite state automata (FSA). We present results for four different classifiers. The FSA module consisted of a 4-state decision logic that filtered false alarms and false positives. We compare the use of four different classifiers in this task. The look ahead of the method that we propose was of 7 frames, which are the number of frames that maximized the accuracy of the system. The system was tested with real signals recorded inside a car, with signal to noise ratio that ranged from 6 dB to 30dB. Finally we present experimental results demonstrating that the system yields robust endpoint detection.

A radiochemical method for carbamoyl-phosphate synthetase I: application to rats fed a hyperproteic diet

A method for the measurement of carbamoyl-phosphate synthetase I activity in animal tissues has been developed using the livers of rats under normal and hyperproteic diets. The method is based on the incorporation of 14C-ammonium bicarbonate to carbamoyl-phosphate in the presence of ATP-Mg and N-acetyl-glutamate. The reaction is stopped by chilling, lowering the pH and adding ethanol. Excess bicarbonate is flushed out under a gentle stream of cold CO2. The only label remaining in the medium was that incorporated into carbamoyl-phosphate, since all 14C-CO2 from bicarbonate was eliminated. The method is rapid and requires only a low pressure supply of CO2 to remove the excess substrate. The reaction is linear up to 10 min using homogenate dilutions of 1:20 to 1:200 (w/v). Rat liver activity was in the range of 89±8 nkat/g. Hyperproteic diet resulted in a significant 1.4-fold increase. The design of the method allows for the processing of multiple samples at the same time, and incubation medium manipulation is unnecessary, since the plastic incubation vial and its contents are finally counted together.

A radiochemical method for carbamoyl-phosphate synthetase I: application to rats fed a hyperproteic diet

A method for the measurement of carbamoyl-phosphate synthetase I activity in animal tissues has been developed using the livers of rats under normal and hyperproteic diets. The method is based on the incorporation of 14C-ammonium bicarbonate to carbamoyl-phosphate in the presence of ATP-Mg and N-acetyl-glutamate. The reaction is stopped by chilling, lowering the pH and adding ethanol. Excess bicarbonate is flushed out under a gentle stream of cold CO2. The only label remaining in the medium was that incorporated into carbamoyl-phosphate, since all 14C-CO2 from bicarbonate was eliminated. The method is rapid and requires only a low pressure supply of CO2 to remove the excess substrate. The reaction is linear up to 10 min using homogenate dilutions of 1:20 to 1:200 (w/v). Rat liver activity was in the range of 89±8 nkat/g. Hyperproteic diet resulted in a significant 1.4-fold increase. The design of the method allows for the processing of multiple samples at the same time, and incubation medium manipulation is unnecessary, since the plastic incubation vial and its contents are finally counted together.

Structural control on present-day topography of a basement massif: the Central and Eastern Anti-Atlas (Morocco)

The Anti-Atlas basement massif extends South of the High Atlas, and, despite a very mild Cenozoic deformation, its altitude exceeds 1500m in large areas, reaching 3305m in Jbel Sirwa. Structural contours of the present elevation of a polygenic planation surface (the High Erosional surface) and of the base of Cretaceous and Neogene inliers have been performed to characterize the major tectonic structures. Gentle Cenozoic WSW-ENE- and N-Strending folds, of 60 to100km wavelength, reactivate Variscan structures, being the major contributors to the local topography of the Anti-Atlas. Reactivated thrusts of decakilometric to kilometric-scale and E-W trend involving the Neogene rocks exhibit a steep attitude and a small displacement, but they also produce a marked topographic expression. The resulting Cenozoic horizontal shortening along N-S sections across the Anti-Atlas is about 1%. The position of the major anticlinal hinges determines the location of the fluvial divides of the Warzazat basin and the Anti-Atlas, and a structural depression on one of these hinges (Jbel Saghro anticline) allowed the formerly endorheic Warzazat basin to drain southwards. The first Cenozoic structures generating local topography are of pre-mid Miocene age (postdated by 6.7Ma volcanic rocks at the Jbel Saghro), whereas the youngest thrust movements postdate the Pliocene sedimentary and volcanic rocks (involving 2.1Ma volcanic rocks at Jbel Sirwa). In addition to these features, the mean elevation of the Anti-Atlas at the regional scale is also the result of a mantle thermal anomaly reported in previous works for the entire Atlas system.

Shared feature extraction for nearest neighbor face recognition

In this paper, we propose a new supervised linearfeature extraction technique for multiclass classification problemsthat is specially suited to the nearest neighbor classifier (NN).The problem of finding the optimal linear projection matrix isdefined as a classification problem and the Adaboost algorithmis used to compute it in an iterative way. This strategy allowsthe introduction of a multitask learning (MTL) criterion in themethod and results in a solution that makes no assumptions aboutthe data distribution and that is specially appropriated to solvethe small sample size problem. The performance of the methodis illustrated by an application to the face recognition problem.The experiments show that the representation obtained followingthe multitask approach improves the classic feature extractionalgorithms when using the NN classifier, especially when we havea few examples from each class

Detection of severe obstructive sleep apnea through voice analysis

tThis paper deals with the potential and limitations of using voice and speech processing to detect Obstruc-tive Sleep Apnea (OSA). An extensive body of voice features has been extracted from patients whopresent various degrees of OSA as well as healthy controls. We analyse the utility of a reduced set offeatures for detecting OSA. We apply various feature selection and reduction schemes (statistical rank-ing, Genetic Algorithms, PCA, LDA) and compare various classifiers (Bayesian Classifiers, kNN, SupportVector Machines, neural networks, Adaboost). S-fold crossvalidation performed on 248 subjects showsthat in the extreme cases (that is, 127 controls and 121 patients with severe OSA) voice alone is able todiscriminate quite well between the presence and absence of OSA. However, this is not the case withmild OSA and healthy snoring patients where voice seems to play a secondary role. We found that thebest classification schemes are achieved using a Genetic Algorithm for feature selection/reduction.