Shantia Yarahmadian

On the Use of the Genetic Algorithm Filter-Based Feature Selection Technique for Satellite Precipitation Estimation

A feature selection technique is used to enhance the precipitation estimation from remotely sensed imagery using an artificial neural network (PERSIANN) and cloud classification system (CCS) method (PERSIANN-CCS) enriched by wavelet features. The feature selection technique includes a feature similarity selection method and a filter-based feature selection using genetic algorithm (FFSGA). It is employed in this study to find an optimal set of features where redundant and irrelevant features are removed. The entropy index fitness function is used to evaluate the feature subsets. The results show that using the feature selection technique not only improves the equitable threat score by almost 7% at some threshold values for the winter season, but also it extremely decreases the dimensionality. The bias also decreases in both the winter (January and February) and summer (June, July, and August) seasons.

On the Use of a Cluster Ensemble Cloud Classification Technique in Satellite Precipitation Estimation

In this paper, the link-based cluster ensemble (LCE) method is utilized to improve cloud classification and satellite precipitation estimation. High resolution Satellite Precipitation Estimation (SPE) is based on the Precipitation Estimation from Remotely Sensed Imagery using an Artificial Neural Network Cloud Classification (PERSIANN-CCS) algorithm. This modified SPE with the incorporation of LCE involves the following four steps: 1) segmentation of infrared cloud images into patches; 2) cloud patch feature extraction; 3) clustering cloud patches using LCE; and 4) dynamic application of brightness temperature (Tb) and rain-rate relationships, derived using satellite observations. In order to cluster the cloud patches, the LCE method combines multiple data partitions from different clustering methods. The results show that using the cluster ensemble increases the performance of rainfall estimates compared to the SPE algorithm using a Self Organizing Map (SOM) neural network. The false alarm ratio (FAR), probabilities of detection (POD), equitable threat score (ETS), and bias are used as quantitative measures to assess the performance of the algorithm. It is shown that both the ETS and bias provide improvement in the summer and winter seasons. Almost 5% ETS improvement is obtained at some threshold values for the winter season using the cluster ensemble.

The dichotomous Markov process with nonparametric test application; a decision support method in long-term river behavioural analysis : The Zayandeh Rud River; a case study from central Iran

We use the Dichotomous Markov Noise model with constant transition rates to describe the dynamics of fluctuations in the water level as a stochastic process, which is imposed on river discharge changes. By applying this model, two different regimes are determined for the long-term behaviour of the river. Based on these regimes, we define two nonparametric classes of the overall increasing/decreasing nature of the water level in the long-term behaviour, which are separated by an exponential steady state regime. In this paper, we develop a nonparametric testing procedure to test exponentially (steady state regime) against an alternative overall decreasing level distribution. The proposed test predicts the long-term regime behaviour of the river. The mathematical tools introduced to handle the problem should be of general use and the testing procedure can be considered as a new mathematical tool in the study of water level dynamics. Under conditions of data austerity and as a case of study, we examine the stochastic characteristics of the Zayandeh Rud (Zāyandé-Rūd or Zāyanderūd, also spelled as Zayandeh-Rood or Zayanderood) River (Isfahan, Iran) water level.

Phase transition analysis of the dynamic instability of microtubules

This paper provides the phase transition analysis of a reaction diffusion equations system modelling the dynamic instability of microtubules (MTs). For this purpose, we have generalized the macroscopic model studied by Mour?o et al (2011 Comput. Biol. Chem. 35 269?81). This model investigates the interaction between the MT nucleation, the essential dynamics parameters and extinction, and their impact on the stability of the system. The considered framework encompasses a system of partial differential equations for the elongation and shortening of MTs, where the rates of elongation as well as the lifetimes of the elongating shortening phases are linear functions of GTP-tubulin concentration. In a novel way, this paper investigates the stability analysis and provides a bifurcation analysis for the dynamic instability of MTs in the presence of diffusion and all of the fundamental dynamics parameters. Our stability analysis introduces the phase transition method as a new mathematical tool in the study of MT dynamics. The mathematical tools introduced to handle the problem should be of general use.

Accuracy of shell and solid elements in vibration analyses of thin- and thick-walled isotropic cylinders

Three-dimensional (3D) theory of elasticity solutions for vibration analysis of simply supported cylinders are obtained and used to find the natural frequencies. The results are compared with various finite element models using both shell and three-dimensional solid elements to verify the accuracy of these finite element models. In addition, results are obtained for various hollow cylinders for possible verification of shell theories and shell elements based on such theories. A point is made here that although 3D elements can provide accurate results, they are not practical in industrial applications of thin- and thick-walled structures. Instead, shell elements are needed for such applications. However, further improvements to the shell theory used in the finite element formulation are needed to improve the accuracy of these shell elements.

The use of compressive sensing and peak detection in the reconstruction of microtubules length time series in the process of dynamic instability

Microtubules (MTs) are intra-cellular cylindrical protein filaments. They exhibit a unique phenomenon of stochastic growth and shrinkage, called dynamic instability. In this paper, we introduce a theoretical framework for applying Compressive Sensing (CS) to the sampled data of the microtubule length in the process of dynamic instability. To reduce data density and reconstruct the original signal with relatively low sampling rates, we have applied CS to experimental MT lament length time series modeled as a Dichotomous Markov Noise (DMN). The results show that using CS along with the wavelet transform significantly reduces the recovery errors comparing in the absence of wavelet transform, especially in the low and the medium sampling rates. In a sampling rate ranging from 0.2 to 0.5, the Root-Mean-Squared Error (RMSE) decreases by approximately 3 times and between 0.5 and 1, RMSE is small. We also apply a peak detection technique to the wavelet coefficients to detect and closely approximate the growth and shrinkage of MTs for computing the essential dynamic instability parameters, i.e., transition frequencies and specially growth and shrinkage rates. The results show that using compressed sensing along with the peak detection technique and wavelet transform in sampling rates reduces the recovery errors for the parameters.

An Îto stochastic differential equations model for the dynamics of the MCF-7 breast cancer cell line treated by radiotherapy

In this paper, a new mathematical model is proposed for studying the population dynamics of breast cancer cells treated by radiotherapy by using a system of stochastic differential equations. The novelty of the model is essentially in capturing the concept of the cell cycle in the modeling to be able to evaluate the tumor lifespan. According to the cell cycle, each cell belongs to one of three subpopulations G, S, or M, representing gap, synthesis and mitosis subpopulations. Cells in the M subpopulation are highly radio-sensitive, whereas cells in the S subpopulation are highly radio-resistant. Therefore, in the process of radiotherapy, cell death rates of different subpopulations are not equal. In addition, since flow cytometry is unable to detect apoptotic cells accurately, the small changes in cell death rate in each subpopulation during treatment are considered. Subsequently, the proposed model is calibrated using experimental data from previous experiments involving the MCF-7 breast cancer cell line. Consequently, the proposed model is able to predict tumor lifespan based on the number of initial carcinoma cells. The results show the effectiveness of the radiation under the condition of stability, which describes the decreasing trend of the tumor cells population.

Integration of Thiele Continued Fractions and the method of fundamental solutions for solving unconfined seepage problems

Unconfined seepage through an earth dam or a levee is recognized as a challenging problem. This complexity is mainly due to the fact that determination of the phreatic line through the dam/levee body is not straightforward. For simulation purposes, mesh generation as well as accurate and smooth alteration of the phreatic line at the junction with the downstream slope (referred to as exit point) for an unconfined seepage problem with complex geometry generally makes cumbersome numerical solutions. This study presents an innovative boundary-type meshfree method to determine the phreatic line location in unconfined seepage problems. The current study explicitly addresses the problem that alternative methods commonly face to deal with the exit point. The method is developed based upon integrating the Method of Fundamental Solutions (MFS), Particle Swarm Optimization (PSO) algorithm, and Thiele Continued Fractions (TCF). To accurately estimate the phreatic line location, the proposed framework uses MFS to solve the flow continuity equation, TCF to generate the phreatic line and PSO to optimize the phreatic line location generated by TCF. As a boundary method, MFS only deals with the boundaries of the domain and consequently, it only takes the exact position of phreatic line as a variable boundary. The proposed approach employs TCF to guarantee that the phreatic line is tangent to the downstream slope at the exit point, a characteristics which is important especially for the cases where abrupt changes occur in the phreatic line near the exit point. For comparison and validation purposes, the phreatic lines determined by the proposed approach for two unconfined seepage problems are compared and verified against those obtained from alternative analytical and numerical methods as well as a set of experimental results. An excellent agreement is demonstrated upon comparison of the proposed method to the results attained from the analytical solutions and experimental tests.

An improvement on the prediction power of the 3D-QSAR CoMFA models using a hybrid of statistical and machine learning methods: a case study on γ‑secretase modulators of Alzheimer’s disease

A comparative molecular field analysis has been developed to study the three-dimensional quantitative structure–activity relationship of a series of triterpene-based γ-secretase modulators. We have performed the genetic algorithm on a large set of comparative molecular field analysis fields to select the most responsible fields contributing to inhibitory activities of these compounds against Alzheimer’s disease. The genetic algorithm-selected comparative molecular field analysis fields were introduced into the partial least squares and principal component analysis to reduce the dimensionality of the input features. The extracted partial least squares components were used as inputs to build partial least squares regression (genetic algorithm-partial least squares regression), and the extracted principal components were used as inputs for principal component regression (genetic algorithm-principal component regression) and support vector regression (genetic algorithm-principal component analysis-support vector regression). The classic three-dimensional quantitative structure–activity relationship comparative molecular field analysis analysis (partial least squares regression) is also carried out for the sake of comparison. The results show that among the constructed models, in terms of root mean squares and leave-one-out cross-validated R2(q2), the combination of principal component analysis and support vector machine can effectively improve the prediction performance (RMSEtrain = 0.231, RMSEtest = 0.360, and q2 = 0.638) compared with PLSR (RMSEtrain = 0.415, RMSEtest = 0.680, and q2 = 0.311). The performances of the genetic algorithm-principal component regression and genetic algorithm-partial least squares regression were also comparable but less powerful than genetic algorithm-principal component analysis-support vector regression. Finally, based on the information derived from the comparative molecular field analysis contour map, some key features for increasing the activity of γ-secretase modulators have been identified to design new triterpene-based Alzheimer’s disease drugs.

A load balancing system for autonomous connection management in heterogeneous wireless networks

In modern heterogeneous wireless networks (HWN), multi-mode devices perform autonomous connection management (ACM) to select the best connections. This selection process causes the challenge of providing global objectives such as load balancing, which have a significant impact on utilization of network resources. In this paper, the proposed connection management system considers the load balancing in HWNs through the trade-off between the individual connection quality and global network objectives. First, a centralized entity calculates the load state of the HWNs by predicting stochastic connection interests of the mobile hosts, then the calculated state is used by ACM system of mobile hosts to improve the network global objectives as well as their own connection quality. The system performance is studied through simulation and modeling in various scenarios. The overall system throughput, load distribution in the network, fairness in access to resources and user satisfaction are evaluated. The results show the effectiveness of the proposed system.