A new deep convolutional neural network model for classifying breast cancer histopathological images and the hyperparameter optimisation of the proposed model

The Journal of Supercomputing - Deep learning algorithms have yielded remarkable results in medical diagnosis and image analysis, besides their contribution to improvements in a number of fields...     

A Stream X-Machine Tool for Modelling and Generating Test Cases for Chronic Diseases Based on State-Counting Approach

Programming and Computer Software - In the biomedical domain, diagrammatical models have been extensively used to describe and understand the behaviour of biological organisms (biological agents)...     

Large-scale SRAM-based IP lookup architectures using compact trie search structures

SRAM-based pipelined architectures for high-speed IP lookup using Field Programmable Gate Arrays (FPGAs) has recently attracted a great deal of attention from researchers. Due to the limited amount of on-chip memory and the number of I/O pins of FPGAs, compact data structures providing high memory efficiency are in great demand.     

Viscoelastic and Processing Effects on the Fiber-Matrix Interphase Strength. Part I. The Effects of Loading Rate, Test Temperature, Fiber Sizing and Global Strain Level

The effects of loading rate, fiber sizing, test temperature and global strain level on the adhesion strength between carbon fibers and a thermosetting epoxy (Epon 815) are studied using the single fiber fragmentation test procedure. Analytical methodology describing the viscoelastic behavior observed is also presented. The possibility of rate-temperature-interphase thickness superposition for the interfacial strength function is illustrated based on the analytical models discussed. Experimental data are discussed using Weibull statistics and also presented in the form of percent relative frequency histograms for the fiber fragments in a collective fashion. The use of histograms allows for interpretation of the skewness in the data population.     

Preparation Polyelectrolyte Complexes of Chitosan-Polyacrylic Acid-Modified Iron Sand Leachate for Proton Exchange Membranes

ABSTRACT

Polyelectrolyte complex (PEC) of chitosan (Chi) and poly (acrylic acid) (PAA)-modified iron sand leachate were prepared and considered for applicability as a proton exchange membrane in fuel cells. Chi-PAA-hematite blended in different weight ratios and the resulting membranes were treated to enable the formation of the polyelectrolyte. The membranes of Chi-PAA polyblend were treated using iron sand leachate and reveal high ion exchange capacity (IEC), proton conductivity, water uptake, and good mechanical stability. The result of research indicated that the membrane with 40 wt% of Chi and 60 wt% of PAA blend which its conductivity of 6.10 × 10^?2 S cm^?1 was potentially for a proton exchange membrane in fuel cell applications.     

Fabrication of chitosan/polyvinyl alcohol/amine modified carbon nanotube composite films for rapid chromate removal

Composite adsorbent films with amine and hydroxyl functionalities were synthesized from chitosan (CS), polyvinyl alcohol (PVA), and amine-modified carbon nanotubes (a-MWCNT) by solvent casting method. Weight proportions of CS to PVA and weight percent of a-MWCNT were optimized to achieve highest chromate removal capacity. Structural characteristics of the composites were investigated using scanning electron microscopy, Fourier transform infrared spectroscopy, Raman spectroscopy, and thermal gravimetric analysis. Accordingly, incorporation of a-MWCNT to CS/PVA structure resulted in the generation of nanochannels, which enhanced adsorption capacity. Moreover, the composite comprising 0.4% wt. a-MWCNT provided over 99% of Cr (VI) removal from 50 mg L⁻¹ Cr (VI) solution within five minutes of contact time. Redlich–Peterson and Radke–Prausnitz isotherm models provided the highest conformity to adsorption data. Maximum chromate sorption capacity of CS/PVA/a-MWCNT composite film was determined as 134.2 mg g⁻¹ being 172% higher than that of CS/PVA. Regeneration was best achieved in 1.0 M NaOH and the composite was shown to retain at least 70% of its original capacity after five consecutive adsorption–desorption cycles.     

Signal-adaptive discrete evolutionary transform as a sparse time–frequency representation

Discrete evolutionary transform (DET) has usually been applied to signals in a blind-way without using any parameters to characterize the signal. For this reason, it is not optimal and needs improvement by using some information about the signal. In this paper, we propose an improvement for the discrete evolutionary transform in order to obtain a sparse representation and redefine the generalized time-bandwidth product optimal short-time Fourier transform as a special case of it. In case of linear FM-type signals, the optimized kernel function of the transform is determined according to signal parameters including the instantaneous frequency. The performance of the adaptive-DET is illustrated on three distinct cases. In case of multi-component LFM signals, when the concentration of the proposed distribution is compared to the ordinary sinusoidal-DET, the improvement is computed as 28% in terms of the ratio of norms. Furthermore we define a new and a general class of distribution functions named as the short-time generalized discrete Fourier transform (ST-GDFT) which is a larger set of signal representations including the adaptive-DET.     

Using an artificial bee colony algorithm for the optimal placement of viscous dampers in planar building frames

In this study, an Artificial Bee Colony Algorithm (ABCA) is used to obtain the optimal size and location of viscous dampers in planar buildings to reduce the damage to the frame systems during an earthquake. The transfer function amplitude of the top displacement and the elastic base shear force evaluated at the first natural circular frequency of structures are chosen as objective functions. The damper coefficients of the added viscous dampers are taken into consideration as design variables in a planar building frame. Transfer function amplitude of the top displacement and the amplitude of the elastic base shear force at the fundamental natural frequency are minimized under an active constraint on sum of the damper coefficients of the added dampers. According to two specified objective functions, an optimization algorithm based on the ABCA is proposed. The proposed method is verified by a gradient-based algorithm; steepest direction search algorithm (SDSA). The proposed ABCA and the SDSA are applied to find the optimal damper distribution for a nine-storey planar building then the optimal damper allocation obtained from the ABCA is investigated to rehabilitate models of irregular planar buildings. The validity of the proposed method was demonstrated through a time history analysis of the optimal damper designs, which were determined based on the frequency domain using the ABCA. The numerical results of the proposed optimal damper design method show that the use of the ABCA can be a practical and powerful tool to determine the optimal damper allocation in planar building structures.