Sharif Digital Repository

We describe our work on incorporating probabilities of phone durations, learned by a neural net, into an ASR system. Phone durations are incorporated via lattice rescoring. The input features are derived from the phone identities of a context window of phones, plus the durations of preceding phones within that window. Unlike some previous work, our network outputs the probability of different durations (in frames) directly, up to a fixed limit. We evaluate this method on several large vocabulary tasks, and while we consistently see improvements inWord Error Rates, the improvements are smaller when the lattices are generated with neural net based acoustic models. Copyright © 2017 ISCA  

The deposition of tetrakis (4-sulonatophenyl) porphyrin (TPPS) thin film on optical fibers presents many possibilities for sensing applications. The J-form aggregation with a narrow and sharp spectral feature at about 490 nm and its sensitivity to humidity have been discussed; a fast change of wavelength occurs according with variation in the humidity level. The reproducibility and high sensitivity of TPPS-coated fibers, along with the capabilities of optical fibers, suggest the device as a good candidate for humidity sensing in harsh environments. © 2018, The Author(s)  

Surface-enhanced Raman spectroscopy is a highly sensitive phenomenon and a powerful fingerprint detection tool that reflects the small changes in polarizability on the pattern and intensity of Raman signals. The SERS enhancement signals elucidate with the surface-selection rules. In this regard, molecular configuration and adsorption orientation on the surface, in addition to the direction of external electric field, can lead to different patterns of SERS spectra. To evaluate how the variation of these features influences the pattern and reproducibility of the spectra, the chemical charge-transfer SERS spectra for pyridine on silver clusters are calculated for different field directions,... 

Recent progress in developing artificial neural network (ANN) metamodels has paved the way for reliable use of these models in the prediction of air pollutant concentrations in urban atmosphere. However, improvement of prediction performance, proper selection of input parameters and model architecture, and quantification of model uncertainties remain key challenges to their practical use. This study has three main objectives: to select an ensemble of input parameters for ANN metamodels consisting of meteorological variables that are predictable by conventional weather forecast models and variables that properly describe the complex nature of pollutant source conditions in a major city, to... 

Instrumental responses are hypothesized to be of two kinds: habitual and goal-directed, mediated by the sensorimotor and the associative cortico-basal ganglia circuits, respectively. The existence of the two heterogeneous associative learning mechanisms can be hypothesized to arise from the comparative advantages that they have at different stages of learning. In this paper, we assume that the goal-directed system is behaviourally flexible, but slow in choice selection. The habitual system, in contrast, is fast in responding, but inflexible in adapting its behavioural strategy to new conditions. Based on these assumptions and using the computational theory of reinforcement learning, we... 

In this study, a three-factor, three-level Box-Behnken design with response surface methodology were used to maximize the TOC removal and minimize the H2O2 residual in the effluent of the combined UV-C/H2O2-VUV system for the treatment of an actual slaughterhouse wastewater (SWW) collected from one of the meat processing plants in Ontario, Canada. The irradiation time and the initial concentrations of total organic carbon (TOCo) and hydrogen peroxide (H2O2o) were the three predictors, as independent variables, studied in the design of experiments. The multiple response approach was used to obtain desirability response surfaces at the optimum factor settings. Subsequently, the optimum... 

The photocatalytic degradation of Reactive Blue 19 (RB19) dye has been studied using TiO2 nanofibre as the photocatalyst in aqueous solution under UV irradiation. Titanium dioxide nanofibre was prepared using a templating method with tetraisopropylorthotitanate as a precursor. The experiments were carried out in the presence of the TiO2 nanofibre, and the effects of pH and electron acceptors on the degradation process were investigated. In order to observe the quality of the aqueous solution, chemical oxygen demand (COD) measurements were also carried out before and after the treatments. The photocatalytic decomposition of RB19 was most efficient in acidic solution. With the addition of... 

In this paper, the robust bounded-input bounded-output stability of a large class of linear time invariant fractional order families of systems with real parametric uncertainties is analyzed. The transfer functions contain polynomials in fractional powers of the Laplace variable s, possibly in combination with exponentials of fractional powers of s. Using the concept of the value set and a generalization of the zero exclusion condition theorem, a theorem to check the robust bounded-input bounded-output stability of these families of systems is presented. An upper cutoff frequency for drawing the value sets is provided as well. Finally, two numerical examples are given to illustrate results... 

Persian Gulf is a semi-enclosed highly saline reverse estuary that is exposed to the risk of oil spills in offshore oil and gas activities. In the early 2000s, a specific version of NOAA's Trajectory Analysis Planner (TAP II) was developed for Persian Gulf to assist regional organizations in preparing oil spill contingency plans. In this research, a new stochastic model is developed to cover the limitations of TAP II. The new model is based on an advanced trajectory model, which is now linked with high resolution spatiotemporal data of the wind and sea current. In a case study, the developed model is compared with TAP II, and evaluated by multiple tests designed for analysis of uncertainty,... 

Traditional load-control musculoskeletal and finite element (FE) models of the spine fail to accurately predict in vivo intervertebral joint loads due mainly to the simplifications and assumptions when estimating redundant trunk muscle forces. An alternative powerful protocol that bypasses the calculation of muscle forces is to drive the detailed FE models by image-based in vivo displacements. Development of subject-specific models, however, both involves the risk of extensive radiation exposures while imaging in supine and upright postures and is time consuming in terms of the reconstruction of the vertebrae, discs, ligaments, and facets geometries. This study therefore aimed to introduce a... 

Artifact removal is an essential part in electroencephalogram (EEG) recording and the raw EEG signals require preprocessing before feature extraction. In this work, we implemented three filtering methods and demonstrated their effects on the performance of different classifiers. Bandpass digital filtering, median filtering and facet method are three preprocessing approaches investigated in this paper. We used data set lib from the BCI competition 2003 for training and testing phase. Our accuracy varied between 80% and 96%. In our work, we demonstrated that the problems of choosing the classifier and preprocessing methods are not independent of each other. Two of our approaches could achieve... 

The biological identity of nanoparticles (NPs) is established by their interactions with a wide range of biomolecules around their surfaces after exposure to biological media. Understanding the true nature of the biomolecular corona (BC) in its native state is, therefore, essential for its safe and efficient application in clinical settings. The fundamental challenge is to visualize the biomolecules within the corona and their relationship/association to the surface of the NPs. Using a synergistic application of cryo-electron microscopy, cryo-electron tomography, and three-dimensional reconstruction, we revealed the unique morphological details of the biomolecules and their... 

Polymeric microspheres which can load biomolecules, cells and active agents play an important role in tissue engineering and drug delivery systems. The conventional double emulsion method has been frequently used to fabricate polymeric microspheres. However, this method has two major shortcomings: the complicated fabrication process which makes it difficult to predict the characteristics of the final microspheres while the size distribution of the microspheres has a wide range. In this study, we eliminate the shortcomings of the conventional double emulsion method and increase its performance without decreasing its high production rate. This can make the proposed modified method a promising... 

A method based on solid-phase microextraction (SPME) followed by on-fiber derivatization and gas chromatography-mass spectrometry detection (GC-MS) for determination of phenol in air was developed. Three different types of SPME fibers, polar and non-polar poly(dimethylsiloxane) (PDMS) and polyethylene glycol (PEG) were synthesized using sol-gel technology and their feasibility to the sampling of phenol were investigated. Different derivatization reagents for post on-fiber derivatization of phenol were studied. Important parameters influencing the extraction and derivatization process such as type of fiber coating, type and volume of derivatizing reagent, derivatization time and temperature,... 

In this paper the field-effect transistor DNA biosensor for detecting hepatitis B virus (HBV) based on indium tin oxide nanowires (ITO NWs) in label free approach has been fabricated. Because of ITO nanowires intensive conductance and functional modified surface, the probe immobilization and target hybridization were increased strongly. The high resolution transmission electron microscopy (HRTEM) measurement showed that ITO nanowires were crystalline and less than 50 nm in diameter. The single-stranded hepatitis B virus DNA (SS-DNA) was immobilized as probe on the Au-modified nanowires. The DNA targets were measured in a linear concentration range from 1fM to 10 µM. The detection limit of... 

Multi-walled carbon nanotubes decorated with Fe 3O 4 nanoparticles (Fe 3O 4NPs/MWCNT) were prepared and used to construct a novel biosensor for the simultaneous detection of adenine and guanine. The direct electro-oxidation of adenine and guanine on the modified electrode were investigated by linear sweep voltammetry. The results indicate a remarkable increase in the oxidation peak currents together with negative shift in the oxidation peak potentials for both adenine and guanine, in comparison to the bare glassy carbon electrode (GCE). The surface morphology and nature of the composite film deposited on GCE were characterized by transmission electron microscopy, atomic force microscopy,... 

A headspace adsorptive microextraction technique was developed using a novel polyaniline-nylon-6 (PANI-N6) nanofiber sheet, fabricated by electrospinning. The homogeneity and the porosity of the prepared PANI-N6 sheet were studied using the scanning electron microscopy (SEM) and nanofibers diameters were found to be around 200nm. The novel nanofiber sheet was examined as an extracting medium to isolate some selected chlorobenzenes (CBs), as model compounds, from aquatic media. The extracted analytes were desorbed using μL-amounts of solvent and eventually an aliquot of extractant was injected into gas chromatography-mass spectrometry (GC-MS). Various parameters affecting the extraction and... 

A novel voltammetric sensor based on glassy carbon electrode (GCE) modified with a thin film of multi-walled carbon nanotubes (MWCNTs) coated with an electropolymerized layer of tiron-doped polypyrrole was developed and the resulting electrode was applied for the determination of acyclovir (ACV). The surface morphology and property of the modified electrode were characterized by field emission scanning electron microscopy and electrochemical impedance spectroscopy techniques. The electrochemical performance of the modified electrode was investigated by means of linear sweep voltammetry (LSV). The effect of several experimental variables, such as pH of the supporting electrolyte, drop size of... 

The feasibility of a novel system to reliably estimate the normalized central blood pressure (CBPN) from the radial photoplethysmogram (PPG) is investigated. Right-wrist radial blood pressure and left-wrist PPG were simultaneously recorded in five different days. An industry-standard applanation tonometer was employed for recording radial blood pressure. The CBP waveform was amplitude-normalized to determine CBPN. A total of fifteen second-order autoregressive models with exogenous input were investigated using system identification techniques. Among these 15 models, the model producing the lowest coefficient of variation (CV) of the fitness during the five days was... 

Goal: Continuous blood pressure (BP) monitoring can provide invaluable information about individuals' health conditions. However, BP is conventionally measured using inconvenient cuff-based instruments, which prevents continuous BP monitoring. This paper presents an efficient algorithm, based on the pulse arrival time (PAT), for the continuous and cuffless estimation of the systolic BP, diastolic blood pressure (DBP), and mean arterial pressure (MAP) values. Methods: The proposed framework estimates the BP values through processing vital signals and extracting two types of features, which are based on either physiological parameters or whole-based representation of vital signals. Finally,...