Prévision du NO2 en utilisant la méthode du réseau de neurones

Cet article décrit une procédure de modélisation du phénomène de dispersion de la concentration du dioxyde de nitrogène (NO2) en utilisant la technique du perceptron multicouche (MLP). Notre objectif est de prouver que la concentration du NO2 est une variable autorégressive et expliquée par des variables météorologiques. Pour ce faire, nous avons procédé par trois étapes : dans la première étape nous avons utilisé la variable concentration NO2 uniquement, dans la seconde étape nous avons utilisé les variables météorologiques uniquement et dans la troisième étape nous avons utilisé la concentration du NO2 combinée avec les variables météorologiques. Les résultats ont montré que le troisième cas est plus performant que les deux autres ce qui prouve notre hypothèse.

Les immunoglobulines intraveineuses et la réponse spécifique des cellules T dans la prévention de la maladie lymphoproliférative post-greffe associée au virus Epstein-Barr chez les enfants greffés de cellules souches hématopoïétiques.

La maladie lymphoproliférative post-greffe (MLP) est une complication grave chez les greffés (d’organes solides ou de cellules souches hématopoïétiques) immunosupprimés suite à l'infection par le virus Epstein-Barr (VEB). En l’absence d'une réponse efficace des lymphocytes T cytotoxiques, les cellules B infectées par le VEB peuvent proliférer et donner lieu à la MLP. Dans le cas des receveurs de greffe immunosupprimés, les cellules B infectées par le VEB de façon lytique, produisent activement de nouveaux virions. Ces derniers infectent les cellules B voisines, entraînant leur expansion polyclonale. La gp350, une protéine du cycle lytique située dans l'enveloppe virale, joue un rôle important dans l'infection par le VEB. Elle interagit avec le récepteur CD21 exprimée à la surface des cellules B pour permettre l’entrée du virus. Ainsi, des anticorps neutralisants anti-gp350 sont considérés être des acteurs clés dans le blocage de l'infection, empêchant ainsi le développement de la MLP. L'effet protecteur des immunoglobulines intraveineuses (IgIV) à titre prophylactique contre le VEB et la MLP chez les greffés de cellules souches hématopoïétiques n’est pas clairement démontré. Par conséquent, le premier objectif de cette thèse a proposé d'évaluer l'efficacité des IgIV contre l'infection par le VEB et la MLP chez les receveurs de cellules souches hématopoïétiques. Le deuxième objectif a proposé de déterminer, en utilisant la technique ELISpot, si la présence d'une réponse forte des lymphocytes T contre l'antigène précoce BMLF1 du cycle lytique du VEB pourrait constituer un marqueur de protection contre la MLP chez les greffés de cellules souches hématopoïétiques. Les résultats ont montré d'une part que, si les IgIV peuvent neutraliser efficacement l'infection par le VEB in vitro, ils ne protègent pas efficacement les patients greffés contre l'infection par le VEB in vivo. D'autre part, l'étude de la réponse des lymphocytes T contre des antigènes du VEB a démontré que les cellules T de certains patients sont capables de reconnaître l'antigène lytique BMLF1. Cette réponse spécifique des lymphocytes T peut s’avérer un bon marqueur de la protection contre la MLP. Les résultats de cette thèse démontrent que l’infection lytique au VEB joue un rôle fondamental dans le développement de la MLP. Les données indiquent également que la présence d'une réponse spécifique des lymphocytes T contre un antigène du cycle lytique du VEB peut constituer un bon marqueur de la protection contre la MLP. Cependant, le traitement des patients recevant des greffes de cellules souches hématopoïétiques avec les IgIV n’apparaît pas efficace dans la prévention de la MLP.

Efecto de bajos niveles de control de la comprensión en el arraigo de ideas lamarckianas

La investigación didáctica ha revelado las destrezas cognitivas y metacognitivas como excelentes predictores del éxito académico frente a otros factores cognitivos e instruccionales (Campanario y Otero, 2000; Baker, 1994). Cuando un sujeto resuelve preguntas o lee con otro objetivo, debe ser capaz de comprender, utilizar información y reflexionar a partir de la información obtenida. Es en este último aspecto, cuando entra en juego el módulo de control, para revisar la representación y evaluar la ausencia de incoherencias temáticas y de contradicciones con su conocimiento previo. En síntesis, la comprensión del tema de Evolución Biológica y por ende cualquier texto científico, requiere tanto de habilidades de Comprensión Lectora para poder construir una buena representación semántica del texto (la Base del Texto), como de estrategias que Controlen el proceso de Comprensión por el que se vincula esta representación textual con el conocimiento previo para formar la representación Modelo de la Situación. Esta construcción requiere la vinculación vía inferencia de la información de la base del texto con la información de la MLP. Si se tiene un nivel alto de Control de la Comprensión, el número y fuerza de las conexiones realizadas vía inferencia, en esa construcción se optimiza. Asimismo se detectan inconsistencias (incoherencias) y se intentan superar. Dada la importancia del control de la comprensión, resulta interesante examinar la situación actual de dichas estrategias en temáticas particulares, tales como la evolución biológica y comprobar hasta que punto se replican los resultados obtenidos en investigaciones anteriores que indican que los niveles bajos en control de la comprensión se asocian en mayor grado con persistencia en ideas lamarckianas tras un tratamiento instruccional

Comprensión de textos : una constatación crucial entre el modelo proposicional de Kintsch y colaboradores y el modelo de memoria de Sanford y Garrod.

Dentro de un marco psicolingüístico, se plantea una contrastación empírica que permita determinar si en la comprensión de textos el factor decisivo es la estructura proposicional del texto o la estructura de memoria del sujeto. N = 108 estudiantes de Psicología de todos los niveles, curso de adaptación y licenciados (18 y 30 años). Asignación aleatoria a los grupos experimentales. Dos diseños: a) Experimental: intrasujeto 2x3 (presencia-ausencia de argumentos comunes en dos segmentos consecutivos de un texto y presencia-ausencia de indicadores de que liciten un escenario); tres valores: ausencia de indicadores, indicadores de escenario facilitador e indicadores de escenario obstaculizador. Variables dependientes: tiempo de lectura. b) Correlacional: intersujeto. Variables predictoras: representación proposicional, recuperación MLP, representación sobre contenidos mnésicos y recuperación de escenarios de la MLP. Variable criterio: tiempo de lectura. Distribución en 6 grupos (n=18) y aplicación de la prueba (lectura de 30 textos, 12 textos intergrupo constantes (correlacional), 12 intergrupo variables (experimental) y 6 intergrupo constantes (mixto) que actúan como control). Pruebas experimentales ad hoc: selección de 30 textos diseñados según cruce de variables utilizadas. Diseño experimental: ANOVA de dos factores. Correlacional: correlación múltiple. Verificación de los textos de control: línea base. W de Kendall. Correlación de Pearson. Diferencias de medias. Ecuaciones de transformación de puntuaciones. Normalidad: Lilliefords. Homogeneidad de varianzas: Bartlett. Intercomparaciones: Scheffe. En el ANOVA resulta un efecto significativo del factor de coherencia postulado por Sanford y Garrod. Por otro lado, ambos modelos obtienen valores predictivos muy similares. Aunque otros resultados indican una superioridad del segundo modelo. Se observa que el efecto obstaculizador de un escenario es cuatro veces superior a su efecto facilitador (relación asimétrica). Se observan diferencias entre sujetos lentos y rápidos debidas a la facilidad para recuperar información de la memoria. Se destaca la adecuada validez convergente de los experimentos, aunque ciertos defectos metodológicos limitan la generalización de resultados y la variabilidad de éstos sea grande. Cabe concluir que los procesos cognitivos implicados en la comprensión de textos operan sobre contenidos de la memoria (modelo de Stanford y Garrod) y no sobre la representación proposicional de los textos (modelo de Kintsch).

Nuevas orientaciones para la didáctica de la lectura a partir de los procesos psicológicos implicados en su aprendizaje.

Corroborar la siguiente hipótesis: los problemas lectores son consecuencia de una deficiencia en memoria que se manifiesta en tres aspectos: deficiencia en MCP y MLP y diferencias en estrategias de recuerdo. Marco teórico en torno a la psicología cognitiva y el procesamiento de la información. Enfoque estructural de la memoria (MCP y MLP como niveles de procesamiento). 66 sujetos buenos lectores y 66 malos lectores, según rendimiento en prueba de lectura, emparejados entre sí en inteligencia, medida con un test, pertenecientes a los cursos tercero, cuarto y quinto de EGB. Diseño de dos grupos apareados en inteligencia: buenos y malos lectores y se les aplicó una batería de memoria. También se controlaron las variables: escolarización regular y problema grave personal o familiar. Posteriormente se realizó un análisis factorial. Test de Raven, prueba de habilidad en lectura y escritura no especificada, para la selección de la muestra. Batería de memoria: prueba de dígitos, prueba de Wickens-Brown-Peterson, prueba de reconocimiento. Paradigma A-B, A-BR. Estadística descriptiva. Prueba de significación estadística. Análisis de varianza. Análisis factorial, rotación Varimax. Los buenos lectores procesan mayor número de unidades de información por unidad de tiempo y mayor cantidad de información en cada unidad, por lo tanto su memoria de trabajo es más amplia. Buenos y malos lectores difieren en las estrategias utilizadas para procesar la información recibida. El buen lector en función de su superior bagaje cognitivo y de su superior memoria semántica, desarrolla unas estrategias de recuedo más eficaces, en consecuencia obtiene mejores resultados en memoria a corto y a largo plazo. A la vista de los resultados obtenidos queda corroborada la hipótesis: entre buenos y malos lectores existen diferencias significativas en reconocimiento de imágenes y en reconocimiento de palabras, diferencias siempre a favor de los buenos lectores.

Myocyte remodeling in response to hypertrophic stimuli requires nucleocytoplasmic shuttling of muscle LIM protein

CSRP3 or muscle LIM protein (MLP) is a nucleocytoplasmic shuttling protein and a mechanosensor in cardiac myocytes. MLP regulation and function was studied in cultured neonatal rat myocytes treated with pharmacological or mechanical stimuli. Either verapamil or BDM decreased nuclear MLP while phenylephrine and cyclic strain increased it. These results suggest that myocyte contractility regulates MLP subcellular localization. When RNA polymerase II was inhibited with alpha-amanitin, nuclear MLP was reduced by 30%. However, when both RNA polymerase I and II were inhibited with actinomycin D, there was a 90% decrease in nuclear MLP suggesting that its nuclear translocation is regulated by both nuclear and nucleolar transcriptional activity. Using cell permeable synthetic peptides containing the putative nuclear localization signal (NLS) of MLP, nuclear import of the protein in cultured rat neonatal myocytes was inhibited. The NLS of MLP also localizes to the nucleolus. Inhibition of nuclear translocation prevented the increased protein accumulation in response to phenylephrine. Furthermore, cyclic strain of myocytes after prior NLS treatment to remove nuclear MLP resulted in disarrayed sarcomeres. Increased protein synthesis and brain natriuretic peptide expression were also prevented suggesting that MLP is required for remodeling of the myo filaments and gene expression. These findings suggest that nucleocytoplasmic shuttling MLP plays an important role in the regulation of the myocyte remodeling and hypertrophy and is required for adaptation to hypertrophic stimuli. (C) 2009 Elsevier Inc. All rights reserved.

On simple adaptive momentum

Simple Adaptive Momentum [1] was introduced as a simple means of speeding the training of multi-layer perceptrons (MLPs) by changing the momentum term depending on the angle between the current and previous changes in the weights of the MLP. In the original paper. the weight changes of the whole network are used in determining this angle. This paper considers adapting the momentum term using certain subsets of these weights. This idea was inspired by the author's object oriented approach to programming MLPs. successfully used in teaching students: this approach is also described. It is concluded that the angle is best determined using the weight changes in each layer separately.

An introduction to radial basis functions for system identification: a comparison with other neural network methods

A look is taken at the use of radial basis functions (RBFs), for nonlinear system identification. RBFs are firstly considered in detail themselves and are subsequently compared with a multi-layered perceptron (MLP), in terms of performance and usage.

Parkinson’s Disease tremor classification – a comparison between Support Vector Machines and neural networks

Deep Brain Stimulation has been used in the study of and for treating Parkinson’s Disease (PD) tremor symptoms since the 1980s. In the research reported here we have carried out a comparative analysis to classify tremor onset based on intraoperative microelectrode recordings of a PD patient’s brain Local Field Potential (LFP) signals. In particular, we compared the performance of a Support Vector Machine (SVM) with two well known artificial neural network classifiers, namely a Multiple Layer Perceptron (MLP) and a Radial Basis Function Network (RBN). The results show that in this study, using specifically PD data, the SVM provided an overall better classification rate achieving an accuracy of 81% recognition.

Modulation of stretch-induced myocyte remodeling and gene expression by nitric oxide: a novel role for lipoma preferred partner in myofibrillogenesis

Prolonged hemodynamic load as a result of hypertension eventually leads to maladaptive cardiac adaptation and heart failure. The signalling pathways that underlie these changes are still poorly understood. The adaptive response to mechanical load is mediated by mechanosensors which convert the mechanical stimuli into a biological response. We examined the effect of cyclic mechanical stretch on myocyte adaptation using neonatal rat ventricular myocytes with 10% (adaptive) or 20% (maladaptive) maximum strain, 1Hz for 48 hours to mimic in vivo mechanical stress. Cells were also treated with and without L-NAME, a general nitric oxide synthase (NOS) inhibitor to suppress NO production. Maladaptive 20% mechanical stretch led to a significant loss of intact sarcomeres which was rescued by LNAME (P<0.05, n≥5 cultures). We hypothesized that the mechanism was through NOinduced alteration of myocyte gene expression. L-NAME up-regulated the mechanosensing proteins Muscle LIM protein (MLP (by 100%, p<0.05, n=4 cultures)) and lipoma preferred partner, a novel cardiac protein (LPP (by 80%, p<0.05, n=4 cultures)). L-NAME also significantly altered the subcellular localisation of LPP and MLP in a manner that favoured growth and adaptation. These findings suggest that NO participates in stretch-mediated adaptation. The use of isoform selective NOS inhibitors indicated a complex interaction between iNOS and nNOS isoforms regulate gene expression. LPP knockdown by siRNA led to formation of α-actinin aggregates and Z-bodies showing that myofibrillogenesis was impaired. There was an up-regulation of E3 ubiquitin ligase (MUL1) by 75% (P<0.05, n=5 cultures). This indicates that NO contributes to stretch-mediated adaptation via the upregulation of proteins associated mechansensing and myofibrillogenesis, thereby presenting potential therapeutic targets during the progression of heart failure. Keywords: Mechanotransduction, heart failure, stretch, heart, hypertrophy

Complex-valued b-spline neural networks for modelling and inverse of wiener systems

Communication signal processing applications often involve complex-valued (CV) functional representations for signals and systems. CV artificial neural networks have been studied theoretically and applied widely in nonlinear signal and data processing [1–11]. Note that most artificial neural networks cannot be automatically extended from the real-valued (RV) domain to the CV domain because the resulting model would in general violate Cauchy-Riemann conditions, and this means that the training algorithms become unusable. A number of analytic functions were introduced for the fully CV multilayer perceptrons (MLP) [4]. A fully CV radial basis function (RBF) nework was introduced in [8] for regression and classification applications. Alternatively, the problem can be avoided by using two RV artificial neural networks, one processing the real part and the other processing the imaginary part of the CV signal/system. A even more challenging problem is the inverse of a CV

Nuclear accumulation of myocyte muscle LIM protein is regulated by heme oxygenase 1 and correlates with cardiac function in the transition to failure

Impaired mechanosensing leads to heart failure and we have previously shown that a decreased ratio of cytoplasmic to nuclear CSRP3/Muscle LIM protein (MLP ratio) is associated with a loss of mechanosensitivity. Here we tested whether passive or active stress/strain was important in modulating the MLP ratio and determined whether this correlated with heart function during the transition to failure. We exposed cultured neonatal rat myocytes to 10% cyclic mechanical stretch at 1 Hz, or electrically paced myocytes at 6.8 V (1 Hz) for 48 h. The MLP ratio decreased 50% (P < 0.05, n = 4) only in response to electrical pacing, suggesting impaired mechanosensitivity. Inhibition of contractility with 10 μM blebbistatin resulted in a ∼3 fold increase in the MLP ratio (n = 8, P < 0.05), indicating that myocyte contractility regulates nuclear MLP. Inhibition of histone deacetylase (HDAC) signaling with trichostatin A increased nuclear MLP following passive stretch, suggesting that HDACs block MLP nuclear accumulation. Inhibition of heme-oxygenase1 (HO-1) activity with PPZII blocked MLP nuclear accumulation. To examine how mechanosensitivity changes during the transition to heart failure, we studied a guinea pig model of angiotensin II infusion (400 ng/kg/min) over 12 weeks. Using subcellular fractionation we showed that the MLP ratio increased 88% (n = 4, P < 0.01) during compensated hypertrophy, but decreased significantly during heart failure (P < 0.001, n = 4). The MLP ratio correlated significantly with the E/A ratio (r = 0.71, P < 0.01 n = 12), a clinical measure of diastolic function. These data indicate for the first time that myocyte mechanosensitivity as indicated by the MLP ratio is regulated primarily by myocyte contractility via HO-1 and HDAC signaling.

A Road Following Approach Using Artificial Neural Networks Combinations

Navigation is a broad topic that has been receiving considerable attention from the mobile robotic community over the years. In order to execute autonomous driving in outdoor urban environments it is necessary to identify parts of the terrain that can be traversed and parts that should be avoided. This paper describes an analyses of terrain identification based on different visual information using a MLP artificial neural network and combining responses of many classifiers. Experimental tests using a vehicle and a video camera have been conducted in real scenarios to evaluate the proposed approach.

A neural networks study of quinone compounds with trypanocidal activity

This work investigates neural network models for predicting the trypanocidal activity of 28 quinone compounds. Artificial neural networks (ANN), such as multilayer perceptrons (MLP) and Kohonen models, were employed with the aim of modeling the nonlinear relationship between quantum and molecular descriptors and trypanocidal activity. The calculated descriptors and the principal components were used as input to train neural network models to verify the behavior of the nets. The best model for both network models (MLP and Kohonen) was obtained with four descriptors as input. The descriptors were T(5) (torsion angle), QTS1 (sum of absolute values of the atomic charges), VOLS2 (volume of the substituent at region B) and HOMO-1 (energy of the molecular orbital below HOMO). These descriptors provide information on the kind of interaction that occurs between the compounds and the biological receptor. Both neural network models used here can predict the trypanocidal activity of the quinone compounds with good agreement, with low errors in the testing set and a high correctness rate. Thanks to the nonlinear model obtained from the neural network models, we can conclude that electronic and structural properties are important factors in the interaction between quinone compounds that exhibit trypanocidal activity and their biological receptors. The final ANN models should be useful in the design of novel trypanocidal quinones having improved potency.

Automatic spiral analysis for objective assessment of motor symptoms in Parkinson's disease

Objective: To develop a method for objective quantification of PD motor symptoms related to Off episodes and peak dose dyskinesias, using spiral data gathered by using a touch screen telemetry device. The aim was to objectively characterize predominant motor phenotypes (bradykinesia and dyskinesia), to help in automating the process of visual interpretation of movement anomalies in spirals as rated by movement disorder specialists. Background: A retrospective analysis was conducted on recordings from 65 patients with advanced idiopathic PD from nine different clinics in Sweden, recruited from January 2006 until August 2010. In addition to the patient group, 10 healthy elderly subjects were recruited. Upper limb movement data were collected using a touch screen telemetry device from home environments of the subjects. Measurements with the device were performed four times per day during week-long test periods. On each test occasion, the subjects were asked to trace pre-drawn Archimedean spirals, using the dominant hand. The pre-drawn spiral was shown on the screen of the device. The spiral test was repeated three times per test occasion and they were instructed to complete it within 10 seconds. The device had a sampling rate of 10Hz and measured both position and time-stamps (in milliseconds) of the pen tip. Methods: Four independent raters (FB, DH, AJ and DN) used a web interface that animated the spiral drawings and allowed them to observe different kinematic features during the drawing process and to rate task performance. Initially, a number of kinematic features were assessed including ‘impairment’, ‘speed’, ‘irregularity’ and ‘hesitation’ followed by marking the predominant motor phenotype on a 3-category scale: tremor, bradykinesia and/or choreatic dyskinesia. There were only 2 test occasions for which all the four raters either classified them as tremor or could not identify the motor phenotype. Therefore, the two main motor phenotype categories were bradykinesia and dyskinesia. ‘Impairment’ was rated on a scale from 0 (no impairment) to 10 (extremely severe) whereas ‘speed’, ‘irregularity’ and ‘hesitation’ were rated on a scale from 0 (normal) to 4 (extremely severe). The proposed data-driven method consisted of the following steps. Initially, 28 spatiotemporal features were extracted from the time series signals before being presented to a Multilayer Perceptron (MLP) classifier. The features were based on different kinematic quantities of spirals including radius, angle, speed and velocity with the aim of measuring the severity of involuntary symptoms and discriminate between PD-specific (bradykinesia) and/or treatment-induced symptoms (dyskinesia). A Principal Component Analysis was applied on the features to reduce their dimensions where 4 relevant principal components (PCs) were retained and used as inputs to the MLP classifier. Finally, the MLP classifier mapped these components to the corresponding visually assessed motor phenotype scores for automating the process of scoring the bradykinesia and dyskinesia in PD patients whilst they draw spirals using the touch screen device. For motor phenotype (bradykinesia vs. dyskinesia) classification, the stratified 10-fold cross validation technique was employed. Results: There were good agreements between the four raters when rating the individual kinematic features with intra-class correlation coefficient (ICC) of 0.88 for ‘impairment’, 0.74 for ‘speed’, 0.70 for ‘irregularity’, and moderate agreements when rating ‘hesitation’ with an ICC of 0.49. When assessing the two main motor phenotype categories (bradykinesia or dyskinesia) in animated spirals the agreements between the four raters ranged from fair to moderate. There were good correlations between mean ratings of the four raters on individual kinematic features and computed scores. The MLP classifier classified the motor phenotype that is bradykinesia or dyskinesia with an accuracy of 85% in relation to visual classifications of the four movement disorder specialists. The test-retest reliability of the four PCs across the three spiral test trials was good with Cronbach’s Alpha coefficients of 0.80, 0.82, 0.54 and 0.49, respectively. These results indicate that the computed scores are stable and consistent over time. Significant differences were found between the two groups (patients and healthy elderly subjects) in all the PCs, except for the PC3. Conclusions: The proposed method automatically assessed the severity of unwanted symptoms and could reasonably well discriminate between PD-specific and/or treatment-induced motor symptoms, in relation to visual assessments of movement disorder specialists. The objective assessments could provide a time-effect summary score that could be useful for improving decision-making during symptom evaluation of individualized treatment when the goal is to maximize functional On time for patients while minimizing their Off episodes and troublesome dyskinesias.