a. a. taheri

Finding an effective way to combine the base learners is an essential part of constructing a heterogeneous ensemble of classifiers. In this paper, we propose a framework for heterogeneous ensembles, which investigates using an artificial neural network to learn a nonlinear combination of the base classifiers. In the proposed framework, a set of heterogeneous classifiers are stacked to produce the first-level outputs. Then these outputs are augmented using several combination functions to construct the inputs of the second-level classifier. We conduct a set of extensive experiments on 121 datasets and compare the proposed method with other established and state-of-the-art heterogeneous methods. The results demonstrate that the proposed scheme outperforms many heterogeneous ensembles, and is superior compared to singly tuned classifiers. The proposed method is also compared to several homogeneous ensembles and performs notably better. Our findings suggest that the improvements are even more significant on larger datasets.

In thermal protection of healthy tissues during hyperthermia with the phase-change micro/nano-materials, the impossibility of performing a similar experiment with the theoretical parameters is inevitable because of different errors such as modeling, measuring, particle deposition area, etc. These errors may affect the practical thermal protection from damaging the healthy tissue or not destroying the tumor tissue. To perform a numerical procedure, the electrical potential is obtained solving the Laplace equation and the Pennes Biothermal equation is used to find the temperature distribution in the tissue using the finite difference method. The Pennes equation is transiently resolved by considering intracellular conductance, blood perfusion, and metabolic heating. Consequently, the deviation and the uncertainty of each parameters in the thermal protection including the concentration of the phase change material, the radius of microcapsules, the latent heat, the melting point, the temperature range of phase change of micro/nanoparticles, and the concentration and the radius of the superparamagnetic materials are investigated. According to the results of the uncertainty analysis, the radius of the superparamagnetic materials is the most important parameter so that a 20% deviation from the numerical value changes the temperature of the tissue up to 4 °C.

Grape powdery mildew (PM) caused by plant pathogenic fungus Erysiphe necator, is the most important disease on grapevine in world and Iran. In order to study the spread of this disease in vineyards of Sistan region, 30 vineyards in three county including Zabol, Zahak and Hamoon were selected during years 2017 and 2018. The vineyards were visited on a weekly basis to record PM incidence (I) and disease severity (S). Based on disease incidence, there were no significant differences between counties and vineyards in first, second and combined two years, and in combined two years Zabol and Zahak (57.6%, 57.2%) had the highest and lowest percent of disease incidence.Based on leaf and fruit disease severity, there were significant differences between counties and two years (P <0.001) and in combined two years Hamoon and Zahak (23.1%, 20.6%) had the highest and lowest percent of infection in leaves. Based on the fruit disease severity, Zabol and Zahak (25.4%, 21.8%) had the highest and lowest percent of infection in fruits respectively. Based on area under disease progress curves, there were no significant differences between years, counties and vineyards in disease incidence but in leaf and fruit disease severity, there were significant differences between years and counties and in combined two years Hamoon and Zabol (696.9 and 680.3) had the highest of leaf and fruit infections respectively.

Objectives:

The level of arterial pressure is one of the most important determinants of cardiac adaptation to hypertension and also one of the most important predictors of cardiovascular and cerebrovascular morbidity, including strokes. Multiple studies were performed on the association between left ventricular hypertrophy and preclinical brain damage in essential hypertensive patients.

Methods:

In order to identify the relation between hypertension level, duration, age, sex and left ventricular geometry to asymptomatic brain damage, we categorized 50 essential hypertensive patients (M/20-%40; F/30-60%) using M-mode echocardiography. All the patients had been admitted in the cardiac emergency room for hypertension control According to the value of end diastolic relative wall thickness (RWT) and left ventricular mass index (LVMI), the patients were categorized into four groups: 5 patients had normal LVMI and normal RWT (normal geometry) (group 1); 7 patients had increased (RWT) and normal LVMI (concentric remodeling) (group 2); 12 patients had increased LVMI and normal RWT (eccentric hypertrophy) (group3); and 26 patients had increased LVMI and RWT (concentric hypertrophy) (group 4). Afterwards, brain MRI was performed, followed by an evaluation of lacunar lesions and leukoaraiosis in the four groups.

Results:

The severity of leukoaraiosis was significantly greater in patients with concentric hypertrophy than in patients with normal left ventricular geometry (Chi-square 24.5, P=0.002). The number of lacunae was also significantly higher in patients with concentric left ventricular hypertrophy than in patients with normal left ventricular, geometry.(Chi -Square 17.25 P=0.000).

Conclusion:

Stepwise regression analysis confirmed that LVMI and RWT in addition to age and sytolic blood pressure were independent predictors for asymptomatic cerebrovascular damage.