New morphological features based on the Sholl analysis for automatic classification of traced neurons.

BACKGROUND: This article addresses the automatic classification of reconstructed neurons through their morphological features. The purpose was to extend the capabilities of the L-Measure software. METHODS: New morphological features were developed, based on modifications of the conventional Sholl analysis. The lengths of the compartments, as well as their volumes, were added to the features used in the classical analysis in order to improve the results during automatic neuron classification. FSM were used to obtain subsets of lower cardinality from the full feature sets and the usefulness of these subsets was tested through their use in supervised classification tasks. The study was based on two types of neurons belonging to mice: pyramidal and GABAergic interneurons. Furthermore, a set of pyramidal neurons belonging to Later 4 and Layer 5 was analyzed. RESULTS: RF classifier shown the best performance combined with a Wrapper method.U-WNAD set allowed to obtain higher values than WN, A and D in all cases and better results than LM for the filters and wrappers FSM. U-LM-WNAD set, led to the highest AUC values for all the FSM studied. Similar results for different regions of cortex were obtained. Comparison with Existing Methods The new features exhibited high discriminatory power with which the values of AUC and Acc obtained in the experiments exceeded those obtained using only the features provided by L-Measure. CONCLUSIONS: The highest values of AUC and Acc were obtained from the sets U-WNAD and U-LM-WNAD, evidencing the discriminatory power of the new proposed features.

New Features for Neuron Classification.

This paper addresses the problem of obtaining new neuron features capable of improving results of neuron classification. Most studies on neuron classification using morphological features have been based on Euclidean geometry. Here three one-dimensional (1D) time series are derived from the three-dimensional (3D) structure of neuron instead, and afterwards a spatial time series is finally constructed from which the features are calculated. Digitally reconstructed neurons were separated into control and pathological sets, which are related to three categories of alterations caused by epilepsy, Alzheimer's disease (long and local projections), and ischemia. These neuron sets were then subjected to supervised classification and the results were compared considering three sets of features: morphological, features obtained from the time series and a combination of both. The best results were obtained using features from the time series, which outperformed the classification using only morphological features, showing higher correct classification rates with differences of 5.15, 3.75, 5.33% for epilepsy and Alzheimer's disease (long and local projections) respectively. The morphological features were better for the ischemia set with a difference of 3.05%. Features like variance, Spearman auto-correlation, partial auto-correlation, mutual information, local minima and maxima, all related to the time series, exhibited the best performance. Also we compared different evaluators, among which ReliefF was the best ranked.

Prevalence and Characteristics of Exercise-Induced Bronchoconstriction in High School and College Athletes at 2,240 m Altitude.

BACKGROUND: Athletes practicing strenuous physical activities may develop exercise-induced bronchoconstriction (EIB). We aimed to determine the prevalence and features of this condition in Mexico City (altitude, 2,240 m). METHODS: In the present study, 208 high school and college athletes performed a standardized EIB test on a treadmill. RESULTS: Responses to exercise had large between-subject variability in all physiological parameters (forced expiratory volume in one second [FEV1], heart rate, blood oxygen saturation level [SpO2], blood pressure), with nearly similar proportions of subjects in whom FEV1 increased or decreased. According to the recommended cut-off value of 10% FEV1 decrease, only 15 (7.2%) athletes had a positive EIB test. Weight lifters were more prone to develop EIB (three out of seven athletes; p = 0.01). Subjects with a positive EIB test already had a lower baseline forced expiratory volume in one second/forced vital capacity (FEV1/FVC) ratio (96.4 vs. 103.2% of predicted, respectively; p = 0.047), and developed more respiratory symptoms after exercise than subjects with a negative test. There were no differences with respect to age, gender, body mass index, history of asthma or atopic diseases, smoking habit, and exposure to potential indoor allergens. CONCLUSIONS: The relatively low prevalence of EIB in athletes from Mexico City raises the possibility that high altitude constitutes a protective factor for EIB. In contrast, weight lifters were especially prone to develop EIB, which suggests that repetitive Valsalva maneuvers could be a novel risk factor for EIB. There was a large between-subject variability of all physiological responses to exercise.

Algoritmos de aprendizaje automático para la clasificación de neuronas piramidales afectadas por el envejecimiento / Machine learning algorithms for classification of pyramidal neurons affected by aging

Una caracterización morfológica precisa de las múltiples clases neuronales del cerebro facilitaría la elucidación de la función cerebral y los cambios funcionales que subyacen a los trastornos neurológicos tales como enfermedades de Parkinson o la Esquizofrenia. El análisis morfológico manual es muy lento y sufre de falta de exactitud porque algunas características de las células no se cuantifican fácilmente. Este artículo presenta una investigación en la clasificación automática de un conjunto de neuronas piramidales de monos jóvenes y adultos, las cuales degradan su estructura morfológica con el envejecimiento. Un conjunto de 21 características se utilizaron para describir su morfología con el fin de identificar las diferencias entre las neuronas. En este trabajo se evalúa el desempeño de cuatro métodos de aprendizaje automático populares en la clasificación de árboles neuronales. Los métodos de aprendizaje de máquinas utilizadas son: máquinas de vectores soporte (SVM), k-vecinos más cercanos (KNN), regresión logística multinomial (MLR) y la red neuronal de propagación hacia atrás (BPNN). Los resultados mostraron las ventajas de MLR y BPNN con respecto a los demás para estos fines. Estos algoritmos de clasificación automáticaofrecen ventajas sobre la clasificación manualbasada en expertos. Mientras que la neurociencia está pasando rápidamente a datos digitales, los principios detrás de los algoritmos de clasificación automática permanecen a menudo inaccesibles para los neurocientíficos, lo que limita las posibilidades de avances(AU)

Accurate morphological characterization of the multiple neuronal classes of the brain would facilitate the elucidation of brain function and the functional changes that underlie neurological disorders such as Parkinson's diseases or Schizophrenia. Manual morphological analysis is very time-consuming and suffers from a lack of accuracy because some cell characteristics are not readily quantified. This paper presents an investigation in the automatic classification of a data set of pyramidal neurons of young and adult monkeys, which degrade his morphologic structure with the aging. A set of 21 features were used to describe their morphology in order to identify differences between neurons. Thispaper evaluates the performance of four popular machine learning methods, in the classification of neural trees. The machine learning methods used are: support vector machines (SVMs), k-nearest neighbors (KNN), multinomial logistic regression (MLR) and back propagation neural network (BPNN). The results showed the advantages of MLR and BPNN with respect to others for this purposes. These automatic classification algorithms offer advantages over manual expert based classification. While neuroscience is rapidly transitioning to digital data, the principles behind automatic classification algorithms remain often inaccessible to neuroscientists, limiting the potential for breakthroughs(AU)

Algoritmos de aprendizaje automático para la clasificación de neuronas piramidales afectadas por el envejecimiento / Machine learning algorithms for classification of pyramidal neurons affected by aging

Una caracterización morfológica precisa de las múltiples clases neuronales del cerebro facilitaría la elucidación de la función cerebral y los cambios funcionales que subyacen a los trastornos neurológicos tales como enfermedades de Parkinson o la Esquizofrenia. El análisis morfológico manual es muy lento y sufre de falta de exactitud porque algunas características de las células no se cuantifican fácilmente. Este artículo presenta una investigación en la clasificación automática de un conjunto de neuronas piramidales de monos jóvenes y adultos, las cuales degradan su estructura morfológica con el envejecimiento. Un conjunto de 21 características se utilizaron para describir su morfología con el fin de identificar las diferencias entre las neuronas. En este trabajo se evalúa el desempeño de cuatro métodos de aprendizaje automático populares en la clasificación de árboles neuronales. Los métodos de aprendizaje de máquinas utilizadas son: máquinas de vectores soporte (SVM), k-vecinos más cercanos (KNN), regresión logística multinomial (MLR) y la red neuronal de propagación hacia atrás (BPNN). Los resultados mostraron las ventajas de MLR y BPNN con respecto a los demás para estos fines. Estos algoritmos de clasificación automáticaofrecen ventajas sobre la clasificación manualbasada en expertos.Mientras que la neurociencia está pasando rápidamente a datos digitales, los principios detrás de los algoritmos de clasificación automática permanecen a menudo inaccesibles para los neurocientíficos, lo que limita las posibilidades de avances(AU)

Accurate morphological characterization of the multiple neuronal classes of the brain would facilitate the elucidation of brain function and the functional changes that underlie neurological disorders such as Parkinson's diseases or Schizophrenia. Manual morphological analysis is very time-consuming and suffers from a lack of accuracy because some cell characteristics are not readily quantified. This paper presents an investigation in the automatic classification of a data set of pyramidal neurons of young and adult monkeys, which degrade his morphologic structure with the aging. A set of 21 features were used to describe their morphology in order to identify differences between neurons. Thispaper evaluates the performance of four popular machine learning methods, in the classification of neural trees. The machine learning methods used are: support vector machines (SVMs), k-nearest neighbors (KNN), multinomial logistic regression (MLR) and back propagation neural network (BPNN). The results showed the advantages of MLR and BPNN with respect to others for this purposes. These automatic classification algorithms offer advantages over manual expert based classification. While neuroscience is rapidly transitioning to digital data, the principles behind automatic classification algorithms remain often inaccessible to neuroscientists, limiting the potential for breakthroughs(AU)