Convolutional Neural Network for Histopathological Osteosarcoma Image Classification

Osteosarcoma is one of the most widespread causes of bone cancer globally and has a high mortality rate. Early diagnosis may increase the chances of treatment and survival however the process is time-consuming (reliability and complexity involved to extract the hand-crafted features) and largely depends on pathologists’ experience. Convolutional Neural Network (CNN—an end-to-end model) is known to be an alternative to overcome the aforesaid problems. Therefore, this work proposes a compact CNN architecture that has been rigorously explored on a Small Osteosarcoma histology Image Dataaseet (a high-class imbalanced dataset). Though, during training, class-imbalanced data can negatively affect the performance of CNN. Therefore, an oversampling technique has been proposed to overcome the aforesaid issue and improve generalization performance. In this process, a hierarchical CNN model is designed, in which the former model is non-regularized (due to dense architecture) and the later one is regularized, specifically designed for small histopathology images. Moreover, the regularized model is integrated with CNN’s basic architecture to reduce overfitting. Experimental results demonstrate that oversampling might be an effective way to address the imbalanced class problem during training. The training and testing accuracies of the non-regularized CNN model are 98% & 78% with an imbalanced dataset and 96% & 81% with a balanced dataset, respectively. The regularized CNN model training and testing accuracies are 84% & 75% for an imbalanced dataset and 87% & 86% for a balanced dataset.     

Multilayer heuristics based clustering framework (MHCF) for author name disambiguation

Scientometrics - Author name ambiguity is a nontrivial problem currently faced by digital libraries and scholarly data search engines affecting their findings related to the authorship data...     

Completing features for author name disambiguation (AND): an empirical analysis

Scientometrics - This study presents a feature enriched AND dataset to develop diverse and better performance achieving AND techniques, by utilizing AND features which have better discriminating...     

Convolutional neural networks for classification of music-listening EEG: comparing 1D convolutional kernels with 2D kernels and cerebral laterality of musical influence

Neural Computing and Applications - This paper highlights the ability of convolutional neural networks (CNNs) at classifying EEG data listening to different kinds of music without the requirement...     

An integrated fine-grain runtime system for MPI

Fine-grain MPI (FG-MPI) extends the execution model of MPI to allow for interleaved execution of multiple concurrent MPI processes inside an OS-process. It provides a runtime that is integrated into the MPICH2 middleware and uses light-weight coroutines to implement an MPI-aware scheduler. In this paper we describe the FG-MPI runtime system and discuss the main design issues in its implementation. FG-MPI enables expression of function-level parallelism, which along with a runtime scheduler, can be used to simplify MPI programming and achieve performance without adding complexity to the program. As an example, we use FG-MPI to re-structure a typical use of non-blocking communication and show that the integrated scheduler relieves the programmer from scheduling computation and communication inside the application and brings the performance part outside of the program specification into the runtime.     

A new technique for the construction of confusion component based on inverse LA-semigroups and its application in steganography

The present article introduced a novel idea for information hiding namely steganography. We have used new notions for the construction of the nonlinear component for block cipher based on inverse LA-semigroups. This nonlinear component fundamentally provides confidentiality in the proposed steganographic algorithm. The construction of the algorithm is fundamentally twofold. Firstly, we have constructed a novel scheme to design confusion component namely substitution box (S-box). Secondly, we have utilized the anticipated nonlinear component in digital steganography. The suggested algorithm is tested for different standard digital images. The authentication of the proposed algorithm is confirmed through statistical analysis.

     

Graphene-SnO(2) composites for highly efficient photocatalytic degradation of methylene blue under sunlight

Graphene sheets decorated with SnO(2) nanoparticles (RGO-SnO(2)) were prepared via a redox reaction between graphene oxide (GO) and SnCl(2). Graphene oxide (GO) was reduced to graphene (RGO) and Sn(2+) was oxidized to SnO(2) during the redox reaction, leading to a homogeneous distribution of SnO(2) nanoparticles on RGO sheets. The scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images show uniform distribution of the nanoparticles on the RGO surface and high-resolution transmission electron microscopy (HRTEM) shows an average particle size of 3-5?nm. The RGO-SnO(2) composite showed an enhanced photocatalytic degradation activity for the organic dye methylene blue under sunlight compared to bare SnO(2) nanoparticles. This result leads us to believe that the RGO-SnO(2) composite could be used in catalytic photodegradation of other organic dyes.     

The influence of temperature and interface strength on the microstructure and performance of sol–gel silica–epoxy nanocomposites

A series of silica–epoxy nanocomposites were prepared by hydrolysis of tetraethoxysilane within the organic matrix at different processing temperatures, i.e., 25 and 60 °C. Epoxy matrices reinforced with 2.0–10.0 wt% silica were subsequently crosslinked with an aliphatic diamine hardener to give optically transparent nanocomposite films. Interphase connections between silica networks and organic matrix were established by in situ functionalization of silica with 2.0 wt% γ-aminopropyltriethoxysilane (APTS). The microstructure of silica–epoxy nanocomposites as studied by transmission electron microscopy indicated the formation of very well-matched nanocomposites with homogeneous distribution of silica at relatively higher temperatures and in the presence of APTS. Thermogravimetric and static mechanical analyses confirmed considerable increase in thermal stability, stiffness, and toughness of the modified composite materials as compared to neat epoxy polymer and unmodified silica–epoxy nanocomposites. A slight improvement in the glass transition temperatures was also recorded by differential scanning calorimetry measurements. High temperature of hydrolysis during the in situ sol–gel process not only improved reaction kinetics but also promoted mutual solubility of the two phases, and consequently enhanced the interface strength. In addition, APTS influenced the size and distribution of the inorganic domain and resulted in better performance of the modified silica–epoxy nanocomposites.     

Seismic Behavior of Flexural Dominated Reinforced Concrete Bridge Columns at Low Temperatures

This paper presents the results from Phase II of an experimental study on the behavior of reinforced concrete bridge columns in cold seismicly active regions. Six half-scale circular reinforced concrete columns, designed to be flexural dominated, were tested under reversed cyclic loading while subjected to temperatures ranging from ?36°C (?33°F) to 22°C (72°F). Four of the units tested were reinforced concrete filled steel tube (RCFST) columns and the other two were ordinary reinforced concrete columns. Results obtained reiterated the observations made in Phase I, which is that low temperatures cause an increase in the flexural strength and initial stiffness as well as a reduction in the spread of plasticity and displacement capacity of the column. Another important observation made was that the plastic hinge length is drastically reduced in the RCFST units compromising the displacement capacity of this type of column even at room temperature conditions. Current predictive models were revised and modified to account for the low-temperature effect.