Improving accuracy and speed of optimum-path forest classifier using combination of disjoint training subsets

The Optimum-Path Forest (OPF) classifier is a recent and promising method for pattern recognition, with a fast training algorithm and good accuracy results. Therefore, the investigation of a combining method for this kind of classifier can be important for many applications. In this paper we report a fast method to combine OPF-based classifiers trained with disjoint training subsets. Given a fixed number of subsets, the algorithm chooses random samples, without replacement, from the original training set. Each subset accuracy is improved by a learning procedure. The final decision is given by majority vote. Experiments with simulated and real data sets showed that the proposed combining method is more efficient and effective than naive approach provided some conditions. It was also showed that OPF training step runs faster for a series of small subsets than for the whole training set. The combining scheme was also designed to support parallel or distributed processing, speeding up the procedure even more. © 2011 Springer-Verlag.

Optimization of Radial Basis Function neural network employed for prediction of surface roughness in hard turning process using Taguchi's orthogonal arrays

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Improving the accuracy of the optimum-path forest supervised classifier for large datasets

In this work, a new approach for supervised pattern recognition is presented which improves the learning algorithm of the Optimum-Path Forest classifier (OPF), centered on detection and elimination of outliers in the training set. Identification of outliers is based on a penalty computed for each sample in the training set from the corresponding number of imputable false positive and false negative classification of samples. This approach enhances the accuracy of OPF while still gaining in classification time, at the expense of a slight increase in training time. © 2010 Springer-Verlag.

Is it possible to make pixel-based radar image classification user-friendly?

In this paper we would like to shed light the problem of efficiency and effectiveness of image classification in large datasets. As the amount of data to be processed and further classified has increased in the last years, there is a need for faster and more precise pattern recognition algorithms in order to perform online and offline training and classification procedures. We deal here with the problem of moist area classification in radar image in a fast manner. Experimental results using Optimum-Path Forest and its training set pruning algorithm also provided and discussed. © 2011 IEEE.

Optimum-Path Forest pruning parameter estimation through Harmony Search

Pattern recognition in large amount of data has been paramount in the last decade, since that is not straightforward to design interactive and real time classification systems. Very recently, the Optimum-Path Forest classifier was proposed to overcome such limitations, together with its training set pruning algorithm, which requires a parameter that has been empirically set up to date. In this paper, we propose a Harmony Search-based algorithm that can find near optimal values for that. The experimental results have showed that our algorithm is able to find proper values for the OPF pruning algorithm parameter. © 2011 IEEE.

Determinação de genes mórbidos e drogáveis a partir da construção e análise da rede integrada de interações moleculares entre genes humanos

The discovering process of new morbid genes and new target proteins for drugs have been shown to be very costly and laborious. Having in view cutting costs and speeding up this process, we propose, in this work, a new method to determine the gene druggability score and morbidity score, the probabilities of the protein encoded by the gene have the characteristics that make it a new target for drugs and in case of an alteration in that gene, we observed a phenotype that characterizes a genetic based illness. To determine these characteristics, we built, analyzed and determined the characteristics of the topology of the integrated molecular interactions network among human genes containing physical interactions between proteins, metabolic interactions and interactions of transcriptional regulation, and included other data such as level of gene transcription and cellular localization of the protein encoded by the gene. We tested our model in training sets and achieved results equal or better than the ones achieved by similar methods in the literature. Finally, with the purpose of investigating whether the assigned scores resembles the potential druggabilities and morbities of the previously unclassi ed genes, we looked for evidences in biomedical literature supporting the potential druggability and morbidity status of genes with the 10 highest scores. We found clear evidences for 73% and 90% of potential druggable and morbid genes respectively

Effect of Strength Training on Rate of Force Development in Older Women

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Effects of physical training with different intensities of effort on lipid metabolism in rats submitted to the neonatal application of alloxan

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

L-carnitine as an ergogenic aid for patients with chronic obstructive pulmonary disease submitted to whole-body and respiratory muscle training programs

The effects of adding L-carnitine to a whole-body and respiratory training program were determined in moderate-to-severe chronic obstructive pulmonary disease (COPD) patients. Sixteen COPD patients (66 ± 7 years) were randomly assigned to L-carnitine (CG) or placebo group (PG) that received either L-carnitine or saline solution (2 g/day, orally) for 6 weeks (forced expiratory volume on first second was 38 ± 16 and 36 ± 12%, respectively). Both groups participated in three weekly 30-min treadmill and threshold inspiratory muscle training sessions, with 3 sets of 10 loaded inspirations (40%) at maximal inspiratory pressure. Nutritional status, exercise tolerance on a treadmill and six-minute walking test, blood lactate, heart rate, blood pressure, and respiratory muscle strength were determined as baseline and on day 42. Maximal capacity in the incremental exercise test was significantly improved in both groups (P < 0.05). Blood lactate, blood pressure, oxygen saturation, and heart rate at identical exercise levels were lower in CG after training (P < 0.05). Inspiratory muscle strength and walking test tolerance were significantly improved in both groups, but the gains of CG were significantly higher than those of PG (40 ± 14 vs 14 ± 5 cmH2O, and 87 ± 30 vs 34 ± 29 m, respectively; P < 0.05). Blood lactate concentration was significantly lower in CG than in PG (1.6 ± 0.7 vs 2.3 ± 0.7 mM, P < 0.05). The present data suggest that carnitine can improve exercise tolerance and inspiratory muscle strength in COPD patients, as well as reduce lactate production.

The effect of eccentric strength training on heart rate and on its variability during isometric exercise in healthy older men

The purpose of this study was to investigate if chronic eccentric strength training (ST) affects heart rate (HR) and heart rate variability (HRV) during sub-maximal isometric voluntary contractions (SIVC). The training group (TG) (9 men, 62 ± 2) was submitted to ST (12 weeks, 2 days/week, 2 - 4 sets of 8-12 repetitions at 75-80% peak torque (PT). The control group (CG) (8 men, 64 ± 4) did not perform ST. The HR and the HRV (RMSSD index) were evaluated during SIVC of the knee extension (15, 30 and 40% of PT). ST increased the eccentric torque only in TG, but did not change the isometric PT and the duration of SIVC. During SIVC, the HR response pattern and the RMSSD index were similar for both groups in pre- and post-training evaluations. Although ST increased the eccentric torque in the TG, it did not generate changes in HR or HRV. © Springer-Verlag 2008.

Modulation of insulin secretion by physical training during recovery from protein malnutrition in rats

Objective: The purpose of the present study was to examine insulin secretion in rats submitted to protein restriction and nutritional recovery associated or not to physical training. Methods: The experiment was designed in two sets of five weeks each. In the first set the rats were fed a nonnal-protein diet(17%-control group) or a low-protein diet (6%-malnourished group) for five weeks. After this, all animals were fed the 17% protein diet and separated into four groups: sedentary control(SC); trained eontrol(TC); sedentary recovered(SR) and trained recovered(TR). TC and TR rats performed swimming exercise. Results: The results indicated efficiency of the 6% protein diet in producing signs of malnutrition, as reduction in body weight gain and serum albumin levels, as well as liver fat. Serum insulin in the fed state and insulin secretion by isolated pancreatic islets in response to glucose were Keduced,but peripheral sensitivity to insulin was increased and glucose tolerance was not changed in the protein deficient rats, indicating adaptation to malnutrition. Diet protocol for nutritional recovery was efficient in repairing body weight gain, serum albumin and liver fat levels of the previously malnourished rats. Glucose induced insulin release by pancreatic islets remained low after nutritional recovery. Insulin secretion by the islets isolated from rats submitted to exercise training during nutritional recovery was improved when compared with the sedentary animals. Conclusion: This indicates that exercise training may be useful in the treatment of protein calorie malnutrition, concerning to glucose induced insulip secretion.

Myogenin, MyoD and IGF-I regulate muscle mass but not fiber-type conversion during resistance training in rats

The purpose of this study was to test the hypothesis that skeletal muscle adaptations induced by long-term resistance training (RT) are associated with increased myogenic regulatory factors (MRF) and insulin-like growth factor-I (IGF-I) mRNA expression in rats skeletal muscle. Male Wistar rats were divided into 4 groups: 8-week control (C8), 8-week trained (T8), 12-week control (C12) and 12-week trained (T12). Trained rats were submitted to a progressive RT program (4 sets of 10-12 repetitions at 65-75% of the 1RM, 3 day/week), using a squat-training apparatus with electric stimulation. Muscle hypertrophy was determined by measurement of muscle fiber cross-sectional area (CSA) of the muscle fibers, and myogenin, MyoD and IGF-I mRNA expression were measured by RT-qPCR. A hypertrophic stabilization occurred between 8 and 12 weeks of RT (control-relative % area increase, T8: 29% vs. T12: 35%; p>0.05) and was accompanied by the stabilization of myogenin (control-relative % increase, T8: 44.8% vs. T12: 37.7%, p>0.05) and MyoD (control-relative % increase, T8: 22.9% vs. T12: 22.3%, p>0.05) mRNA expression and the return of IGF-I mRNA levels to the baseline (control-relative % increase, T8: 30.1% vs. T12: 1.5%, p<0.05). Moreover, there were significant positive correlations between the muscle fiber CSA and mRNA expression for MyoD (r=0.85, p=0.0001), myogenin (r=0.87, p=0.0001), and IGF-I (r=0.88, p=0.0001). The significant (p<0.05) increase in myogenin, MyoD and IGF-I mRNA expression after 8 weeks was not associated with changes in the fiber-type frequency. In addition, there was a type IIX/D-to-IIA fiber conversion at 12 weeks, even with the stabilization of MyoD and myogenin expression and the return of IGF-I levels to baseline. These results indicate a possible interaction between MRFs and IGF-I in the control of muscle hypertrophy during long-term RT and suggest that these factors are involved more in the regulation of muscle mass than in fiber-type conversion. © Georg Thieme Verlag KG Stuttgart · New York.

Effect of different rest intervals, between sets, on muscle performance during leg press exercise, in trained older women

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Effects of resistance training on the performance of activities of daily living in patients with Alzheimer's disease

Aim: The objective of this study was to investigate the effects of resistance training in activities of daily living performance in Alzheimer's disease (AD) patients. Methods: An exploratory and longitudinal study, lasting for 16weeks, with the participation of 34 patients divided equally in: the training group (TG), who participated in a resistance training protocol (three sets of 20 repetitions in five exercises); and the social gathering group (SGG), who participated in a social interaction protocol (i.e. group dynamics, writing and reading activities). Results: We observed significant differences between the groups in moving around the house, climbing stairs, standing up from the floor and putting on socks tests. Conclusion: This study showed that resistance training improves agility, lower limb strength, balance and flexibility in AD patients, while SGG protocol is important to improve the agility. © 2012 Japan Geriatrics Society.