Análisis comparativo de algoritmos de aprendizaje para predecir la evolución de pacientes con Daño Cerebral Adquirido

ste trabajo presenta un análisis comparativo entre tres algoritmos de aprendizaje diferentes basados en Árboles de Decisión (C4.5) y Redes Neuronales Artificiales (Perceptrón Multicapa MLP y Red Neuronal de Regresión General GRNN) que han sido implementados con el objetivo de predecir los resultados de la rehabilitación cognitiva de personas con daño cerebral adquirido. En el análisis se han incluido datos demográficos del paciente, el perfil de afectación y los resultados provenientes de las tareas de rehabilitación ejecutadas por los pacientes. Los modelos han sido evaluados utilizando la base de datos del Institut Guttmann. El rendimiento de los algoritmos se midió a través del análisis de la especificidad, sensibilidad y exactitud en la precisión y el análisis de la matriz de confusión. Los resultados muestran que la implementación del C4.5 alcanzó una especificidad, sensibilidad y exactitud en la precisión del 98.43%, 83.77% y 89.42% respectivamente. El rendimiento del C4.5 fue significativamente superior al obtenido por el Perceptrón Multicapa y la Red de Regresión General.

Minería de Datos usando Metaplasticidad Artificial en la Rehabilitación Cognitiva de Pacientes con Daño Cerebral

El propósito principal de esta investigación es la aplicación de la Metaplasticidad Artificial en un Perceptrón Multicapa (AMMLP) como una herramienta de minería de datos para la predicción y extracción explícita de conocimiento del proceso de rehabilitación cognitiva en pacientes con daño cerebral adquirido. Los resultados obtenidos por el AMMLP junto con el posterior análisis de la base de datos ayudarían a los terapeutas a conocer las características de los pacientes que mejoran y los programas de rehabilitación que han seguido. Esto incrementaría el conocimiento del proceso de rehabilitación y facilitaría la elaboración de hipótesis terapéuticas permitiendo la optimización y personalización de las terapias. La evaluación del AMMLP se ha realizado con datos proporcionados por el Institut Guttmann. Los resultados del AMMLP fueron comparados con los obtenidos con una red neuronal de retropropagación y con árboles de decisión. La exactitud en la predicción obtenida por el AMMLP en la subfunción cognitiva memoria verbal-visual fue de 90.71 %, resultado muy superior a los obtenidos por los demás algoritmos.

Improvement for detection of microcalcifications through clustering algorithms and artificial neural networks

A new method for detecting microcalcifications in regions of interest (ROIs) extracted from digitized mammograms is proposed. The top-hat transform is a technique based on mathematical morphology operations and, in this paper, is used to perform contrast enhancement of the mi-crocalcifications. To improve microcalcification detection, a novel image sub-segmentation approach based on the possibilistic fuzzy c-means algorithm is used. From the original ROIs, window-based features, such as the mean and standard deviation, were extracted; these features were used as an input vector in a classifier. The classifier is based on an artificial neural network to identify patterns belonging to microcalcifications and healthy tissue. Our results show that the proposed method is a good alternative for automatically detecting microcalcifications, because this stage is an important part of early breast cancer detection

A Prediction Model to Diabetes using Artificial Metaplasticity

Diabetes is the most common disease nowadays in all populations and in all age groups. Different techniques of artificial intelligence has been applied to diabetes problem. This research proposed the artificial metaplasticity on multilayer perceptron (AMMLP) as prediction model for prediction of diabetes. The Pima Indians diabetes was used to test the proposed model AMMLP. The results obtained by AMMLP were compared with other algorithms, recently proposed by other researchers, that were applied to the same database. The best result obtained so far with the AMMLP algorithm is 89.93%

Microcalcifications Detection using PFCM and ANN

This work presents a method to detect Microcalcifications in Regions of Interest from digitized mammograms. The method is based mainly on the combination of Image Processing, Pattern Recognition and Artificial Intelligence. The Top-Hat transform is a technique based on mathematical morphology operations that, in this work is used to perform contrast enhancement of microcalcifications in the region of interest. In order to find more or less homogeneous regions in the image, we apply a novel image sub-segmentation technique based on Possibilistic Fuzzy c-Means clustering algorithm. From the original region of interest we extract two window-based features, Mean and Deviation Standard, which will be used in a classifier based on a Artificial Neural Network in order to identify microcalcifications. Our results show that the proposed method is a good alternative in the stage of microcalcifications detection, because this stage is an important part of the early Breast Cancer detection

EMI filter design of a three-phase buck-type rectifier for aircraft applications

An EMI filter for a three-phase buck-type medium power pulse-width modulation rectifier is designed. This filter considers differential mode noise and complies with MIL-STD- 461E for the frequency range of 10kHz to 10MHz. In industrial applications, the frequency range of the standard starts at 150kHz and the designer typically uses a switching frequency of 28kHz because the fifth harmonic is out of the range. This approach is not valid for aircraft applications. In order to design the switching frequency in aircraft applications, the power losses in the semiconductors and the weight of the reactive components should be considered. The proposed design is based on a harmonic analysis of the rectifier input current and an analytical study of the input filter. The classical industrial design does not consider the inductive effect in the filter design because the grid frequency is 50/60Hz. However, in the aircraft applications, the grid frequency is 400Hz and the inductance cannot be neglected. The proposed design considers the inductance and the capacitance effect of the filter in order to obtain unitary power factor at full power. In the optimization process, several filters are designed for different switching frequencies of the converter. In addition, designs from single to five stages are considered. The power losses of the converter plus the EMI filter are estimated at these switching frequencies. Considering overall losses and minimal filter volume, the optimal switching frequency is selected

New Considerations in the Input Filter Design of a Three-Phase Buck-Type PWM Rectifier for Aircraft Applications

An EMI filter for a three-phase buck-type medium power pulse-width modulation rectifier is designed. This filter considers differential mode noise and complies with MIL-STD-461E for the frequency range of 10kHz to 10MHz. In industrial applications, the frequency range of the standard starts at 150kHz and the designer typically uses a switching frequency of 28kHz because the fifth harmonic is out of the range. This approach is not valid for aircraft applications. In order to design the switching frequency in aircraft applications, the power losses in the semiconductors and the weight of the reactive components should be considered. The proposed design is based on a harmonic analysis of the rectifier input current and an analytical study of the input filter. The classical industrial design does not consider the inductive effect in the filter design because the grid frequency is 50/60Hz. However, in the aircraft applications, the grid frequency is 400Hz and the inductance cannot be neglected. The proposed design considers the inductance and the capacitance effect of the filter in order to obtain unitary power factor at full power. In the optimization process, several filters are designed for different switching frequencies of the converter. In addition, designs from single to five stages are considered. The power losses of the converter plus the EMI filter are estimated at these switching frequencies. Considering overall losses and minimal filter volume, the optimal switching frequency is selected.

Manejo Integral de Agua y Suelo en Centroamérica. Bases científicas para el desarrollo rural comunitario.

Este libro recoge los frutos de la colaboración y trabajo conjunto de un grupo de Universidades Iberoamericanas entre 2007 y 2012 el marco de las actividades del Programa de Cooperación Comunidad, Agua y Bosque en Centroamérica (CAB Centroamérica, http://www2.caminos.upm.es/Departamentos/imt/Topografia/Cab/cab.html ). Las actividades se han realizado con el apoyo del Programa de Cooperación Universitaria PCI-AECID IBEROAMÉRICA, de la Dirección de Cooperación para el Desarrollo de la Universidad Politécnica de Madrid y de los fondos propios de las Universidades latinoamericanas, con especial mención a la Universidad de Costa Rica, coordinadora de los trabajos en Centroamérica. El inicio de esta colaboración se produjo en 2007 a partir de la identificación de un objetivo común: profundizar la investigación sobre la dinámica agua-suelo-planta para mejorar la producción y la calidad del agua de los sistemas de abastecimiento comunitarios en Centroamérica.

Inverse kinematics of a 6 DoF human upper limb using ANFIS and ANN for anticipatory actuation in ADL-based physical Neurorehabilitation

Objective: This research is focused in the creation and validation of a solution to the inverse kinematics problem for a 6 degrees of freedom human upper limb. This system is intended to work within a realtime dysfunctional motion prediction system that allows anticipatory actuation in physical Neurorehabilitation under the assisted-as-needed paradigm. For this purpose, a multilayer perceptron-based and an ANFIS-based solution to the inverse kinematics problem are evaluated. Materials and methods: Both the multilayer perceptron-based and the ANFIS-based inverse kinematics methods have been trained with three-dimensional Cartesian positions corresponding to the end-effector of healthy human upper limbs that execute two different activities of the daily life: "serving water from a jar" and "picking up a bottle". Validation of the proposed methodologies has been performed by a 10 fold cross-validation procedure. Results: Once trained, the systems are able to map 3D positions of the end-effector to the corresponding healthy biomechanical configurations. A high mean correlation coefficient and a low root mean squared error have been found for both the multilayer perceptron and ANFIS-based methods. Conclusions: The obtained results indicate that both systems effectively solve the inverse kinematics problem, but, due to its low computational load, crucial in real-time applications, along with its high performance, a multilayer perceptron-based solution, consisting in 3 input neurons, 1 hidden layer with 3 neurons and 6 output neurons has been considered the most appropriated for the target application.

Data mining applied to the cognitive rehabilitation of patients with acquired brain injury

Acquired brain injury (ABI) is one of the leading causes of death and disability in the world and is associated with high health care costs as a result of the acute treatment and long term rehabilitation involved. Different algorithms and methods have been proposed to predict the effectiveness of rehabilitation programs. In general, research has focused on predicting the overall improvement of patients with ABI. The purpose of this study is the novel application of data mining (DM) techniques to predict the outcomes of cognitive rehabilitation in patients with ABI. We generate three predictive models that allow us to obtain new knowledge to evaluate and improve the effectiveness of the cognitive rehabilitation process. Decision tree (DT), multilayer perceptron (MLP) and general regression neural network (GRNN) have been used to construct the prediction models. 10-fold cross validation was carried out in order to test the algorithms, using the Institut Guttmann Neurorehabilitation Hospital (IG) patients database. Performance of the models was tested through specificity, sensitivity and accuracy analysis and confusion matrix analysis. The experimental results obtained by DT are clearly superior with a prediction average accuracy of 90.38%, while MLP and GRRN obtained a 78.7% and 75.96%, respectively. This study allows to increase the knowledge about the contributing factors of an ABI patient recovery and to estimate treatment efficacy in individual patients.

Maintenance management model for industrial Smes

This paper presents the design and results of the implementation of a model for the evaluation and improvement of maintenance management in industrial SMEs. A thorough review of the state of the art on maintenance management was conducted to determine the model variables; to characterize industrial SMEs, a questionnaire was developed with Likert variables collected in the previous step. Once validated the questionnaire, we applied the same to a group of seventy-five (75) SMEs in the industrial sector, located in Bolivar State, Venezuela. To identify the most relevant variables maintenance management, we used exploratory factor analysis technique applied to the data collected. The score obtained for all the companies evaluated (57% compliance), highlights the weakness of maintenance management in industrial SMEs, particularly in the areas of planning and continuous improvement; most SMEs are evaluated in corrective maintenance stage, and its performance standard only response to the occurrence of faults.

Modelo de gestión logística para pymes industriales

This paper presents the design and results of applying a model for logistics management in industrial SMEs. To identify the variables in the model, we conducted a thorough review of the state of the art logistics management; to characterize SMEs, developed a Likert questionnaire with the variables collected in the previous step. Once validated the questionnaire, was applied the same to a group of seventy-five (75) SMEs in the industrial sector, located in Bolivar State, Venezuela. To determine statistically the most relevant variables of management was used exploratory factor analysis technique applied to the data collected. The qualification obtained for all companies evaluated (47% compliance), highlights the weakness of logistics management in industrial SME.

Atlas-free Brain Tissue Segmentation Using a Single T1-weighted MRI Acquisition

Many studies investigating the aging brain or disease-induced brain alterations rely on accurate and reproducible brain tissue segmentation. Being a preliminary processing step prior to the segmentation, reliableskull-stripping the removal ofnon-brain tissue is also crucial for all later image assessment. Typically, segmentation algorithms rely on an atlas i.e. pre-segmented template data. Brain morphology, however, differs considerably depending on age, sex and race. In addition, diseased brains may deviate significantly from the atlas information typically gained from healthy volunteers. The imposed prior atlas information can thus lead to degradation of segmentation results. The recently introduced MP2RAGE sequence provides a bias-free T1 contrast with heavily reduced T2*- and PD-weighting compared to the standard MP-RAGE [1]. To this end, it acquires two image volumes at different inversion times in one acquisition, combining them to a uniform, i.e. homogenous image. In this work, we exploit the advantageous contrast properties of the MP2RAGE and combine it with a Dixon (i.e. fat-water separation) approach. The information gained by the additional fat image of the head considerably improves the skull-stripping outcome [2]. In conjunction with the pure T1 contrast of the MP2RAGE uniform image, we achieve robust skull-stripping and brain tissue segmentation without the use of an atlas

Data mining for the diagnosis of type 2 diabetes

Diabetes is the most common disease nowadays in all populations and in all age groups. diabetes contributing to heart disease, increases the risks of developing kidney disease, blindness, nerve damage, and blood vessel damage. Diabetes disease diagnosis via proper interpretation of the diabetes data is an important classification problem. Different techniques of artificial intelligence has been applied to diabetes problem. The purpose of this study is apply the artificial metaplasticity on multilayer perceptron (AMMLP) as a data mining (DM) technique for the diabetes disease diagnosis. The Pima Indians diabetes was used to test the proposed model AMMLP. The results obtained by AMMLP were compared with decision tree (DT), Bayesian classifier (BC) and other algorithms, recently proposed by other researchers, that were applied to the same database. The robustness of the algorithms are examined using classification accuracy, analysis of sensitivity and specificity, confusion matrix. The results obtained by AMMLP are superior to obtained by DT and BC.

Gestión de seguridad y salud en el trabajo: aplicación a las pymes industriales

En este artículo se presentan el diseño y los resultados de la aplicación de un modelo para la evaluación y mejora de la gestión de la seguridad y salud en el trabajo; esta investigación forma parte de un proyecto más amplio dirigido al diseño de un modelo de sistema inteligente para la evaluación y mejora de la gestión empresarial, que soporte la toma de decisión en las Pymes industriales.