Search

Breast cancer, a prevalent and life-threatening disease, necessitates early detection for the effective intervention and the improved patient health outcomes. This paper focuses on the critical problem of identifying breast cancer using a model called Attention U-Net. The model is utilized on the Breast Ultrasound Image Dataset (BUSI), comprising 780 breast images. The images are categorized into three distinct groups: 437 cases classified as benign, 210 cases classified as malignant, and 133 cases classified as normal. The proposed model leverages the attention-driven U-Net's encoder blocks to capture hierarchical features effectively. The model comprises four decoder blocks which is a pivotal component in the U-Net architecture, responsible for expanding the encoded feature representation obtained from the encoder block and for reconstructing spatial information. Four attention gates are incorporated strategically to enhance feature localization during decoding, showcasing a sophisticated design that facilitates accurate segmentation of breast tumors in ultrasound images. It displays its efficacy in accurately delineating and segregating tumor borders. The experimental findings demonstrate outstanding performance, achieving an overall accuracy of 0.98, precision of 0.97, recall of 0.90, and a dice score of 0.92. It demonstrates its effectiveness in precisely defining and separating tumor boundaries. This research aims to make automated breast cancer segmentation algorithms by emphasizing the importance of early detection in boosting diagnostic capabilities and enabling prompt and targeted medical interventions. [ABSTRACT FROM AUTHOR]

The methodologies based on neural networks are substantial to accomplish sentiment analysis in the Social Internet of Things (SIoT). With social media sentiment analysis, significant insights can produce efficient and intelligent applications. Neural networks such as recurrent neural networks (RNNs) and convolution neural networks (CNNs) have been considered widely in many text classification tasks. However, RNNs are computationally expensive and require complex training to capture contextual information and long-term dependencies. Similarly, traditional CNNs must stack multiple convolutional layers, requiring massive computations and additional parameters. To address these problems, this work initialized the novel architecture, in which contextual representations (CRs) based on the textual framework are proposed at the initial step. In CRs, state-of-the-art word representation models, such as GloVe (global vectors) and FastText (subword information), collectively produce word representations upon the input sequence using a weight mechanism. Secondly, a unique way is introduced: a three-parallel layered dilated convolutional network with global mean pooling. The experimental results show that the proposed methods when compared with baseline methods, the dilation in CNNs following CRs significantly increases the accuracy from 72.45 to 98.98% and reduces computational resources. [ABSTRACT FROM AUTHOR]

Software vulnerabilities pose a significant threat to system security, necessitating effective automatic detection methods. Current techniques face challenges such as dependency issues, language bias, and coarse detection granularity. This study presents a novel deep learning-based vulnerability detection system for Java code. Leveraging hybrid feature extraction through graph and sequence-based techniques enhances semantic and syntactic understanding. The system utilizes control flow graphs (CFG), abstract syntax trees (AST), program dependencies (PD), and greedy longest-match first vectorization for graph representation. A hybrid neural network (GCN-RFEMLP) and the pre-trained CodeBERT model extract features, feeding them into a quantum convolutional neural network with self-attentive pooling. The system addresses issues like long-term information dependency and coarse detection granularity, employing intermediate code representation and inter-procedural slice code. To mitigate language bias, a benchmark software assurance reference dataset is employed. Evaluations demonstrate the system's superiority, achieving 99.2% accuracy in detecting vulnerabilities, outperforming benchmark methods. The proposed approach comprehensively addresses vulnerabilities, including improper input validation, missing authorizations, buffer overflow, cross-site scripting, and SQL injection attacks listed by common weakness enumeration (CWE). [ABSTRACT FROM AUTHOR]

Background: Colloid cysts arise from congenital abnormalities of the endodermis and are characteristically comprised of an external fibrous layer, with a lining of cuboidal and columnar epithelial cells and endodermal ciliated cells. They contain a gelatinous matrix, produced by mucus-producing epithelial cells. Case presentation: An 8-year-old Saudi girl died in Dammam Central Hospital, to which she was admitted by her parents following her complaints of intense headaches. When the body was examined externally, it was noted that she had bilateral eye congestion, evidence of needle marks on all four limbs and in the region of the femoral triangle and bruising in the central chest and left shoulder areas. No other evidence of injury was noted. A cyst in the third ventricle of the brain was identified at post mortem. Investigations for alcohol, toxins and illicit substances were negative. Conclusion: It was surmised that the child had died as a result of a cerebral colloid cyst and its clinical sequelae. [ABSTRACT FROM AUTHOR]

This work is devoted to study a new class of static neural network. We propose a static multidirectional associative memory neural network with time-varying coefficients and delays. We assess the existence and global exponential stability of weighted pseudo almost-periodic solutions for the systems through exponential dichotomy theory, contraction mapping fixed point theorem, differential inequality techniques and weighted pseudo almost-periodic functions. The theoretical results obtained are supported with numerical simulations. To our knowledge, no paper in the literature has investigated the existence and the delay-independent stability of weighted pseudo almost-periodic solution for static delayed MAM neural network. The results obtained are new and complements that found in the literature. [ABSTRACT FROM AUTHOR]

This paper introduces a novel Energy Efficient Mobility-Based Watchman Algorithm (E2-MBWA) to intensification packet delivery ratio of mitigating the Hotspot issue in Wireless Sensor and Actor Networks (WSAN). Hotspot issue mostly causes of network breakdown and decrease of data packet delivery. Therefore, it is required to design a new technique for data packet forwarding that can resolve these issues in the network. In this study, E2-MBWA has introduced, that cope with the layer-by-layer mechanism for data packet forwarding. The proposed algorithm works with the help of the Data Packet Forwarding Algorithm (DFPA) and Watchman Layer Update Mechanism (WLUM). Furthermore, it also rescues the data storage issues, for this, used secondary nods as substitutes. Moreover, proposed technique is compared with some latest baseline's approaches, for example, Efficient Traffic Load Reduction Algorithm (ETLRA). The analytical energy model is also described for the best health of the network to measure the accuracy level of the Hotspot issue. [ABSTRACT FROM AUTHOR]

AA7050 is an aluminum alloy with superior mechanical properties; however, it is prone to hot tearing (HT) during its production via direct-chill casting. This study focuses on extracting constitutive parameters of the alloy thermomechanical behavior in semi-solid state as well as gaining insight in its failure behavior. Tensile tests were performed using an Instron 5944 at solid fractions between 0.85 (550 °C) and 1.0 (465 °C), at deformation rates of 0.2 and 2 mm/min. The results showed that there are three mechanical behavior regimes in this solid fraction range: ductile at 1.0 (T = 465 °C) ≤ fs < 0.97 (T = 473 °C), brittle at 0.97 (T = 473 °C) ≤ fs ≤ 0.9 (T = 485 °C) and then ductile again (at 0.9 (T = 485 °C) < fs ≤ 0.85 (T = 550 °C)). Fracture surface analysis revealed that the fracture mode was mostly intergranular with fracture propagating through solid bridges as well. Semi-solid constitutive parameters were obtained by making a simple thermal model and numerical tensile tests in ALSIM software package and comparing the simulation results with experimental mechanical tests. The extracted constitutive parameters and available information from the literature support the fact that AA7050 is more susceptible to HT than AA5182 and Al-2 wt pct Cu alloys. The obtained parameters can further enhance the predictive capability of computer simulations of direct-chill casting. [ABSTRACT FROM AUTHOR]

This paper deals with a generalized multidirectional associative memory neural network with time-varying coefficients and mixed delays. Some criteria are established for the existence and the global exponential stability of weighted pseudo almost-periodic solutions for the studied model. To do so, the exponential dichotomy theory, the contraction mapping fixed point theorem and the differential inequality techniques are used. Moreover, an example and its numerical simulation are employed to illustrate the main results. [ABSTRACT FROM AUTHOR]

In the paper, under some sufficient conditions, the existence and the global exponential stability of a pseudo almost automorphic solution to multidirectional associative memory (MAM) neural network with time-varying coefficients and mixed delays is obtained by combining both differential inequality techniques and the classical Banach's fixed-point principle. As you will see, our results improve and generalize many previous known results in MAM. Finally, an example together with its numeric simulations show the feasibility and effectiveness of our main results. [ABSTRACT FROM AUTHOR]

A simplified three-phase, multiscale macrosegregation model which describes the growth kinetics of equiaxed grains and the coupling between microstructure morphology and the macroscopic transport has been proposed previously. In this paper, the model is validated by comparing the numerical model predictions to the experimental data from DC casting of an AA7050 alloy billet. The morphology of the equiaxed grains has an important influence on the macrosegregation, and we show that the model predictions are accurate when the grain morphology is described correctly. [ABSTRACT FROM AUTHOR]

Direct chill (DC) casting of aluminum involves alloys employing different solute elements. In this article, a qualitative analysis and comparison of macrosegregation formation is presented for three different alloy systems: Al-Mg, Al-Zn and Al-Cu. For this purpose, a multiphase, multiscale solidification model based on a volume-averaging method accounting for shrinkage-induced flow, thermal-solutal convection and grain motion is used and applied to an industrial-scale DC-cast ingot. The primary difference between these alloys is the thermal-solutal convection with Al-Mg having a competing thermal and solutal convection, whereas the other two systems have a cooperating thermal and solutal convection. In the study, the combined effect of the macrosegregation mechanisms is analyzed for each alloy to assess the role of the alloy system on the final macrosegregation. [ABSTRACT FROM AUTHOR]

Macrosegregation is a result of the interplay of various transport mechanisms, including natural convection, solidification shrinkage, and grain motion. Experimental observations also indicate the impact of grain morphology, ranging from dendritic to globular, on macrosegregation formation. To avoid the complexity arising due to modeling of an equiaxed dendritic grain, we present the development of a simplified three-phase, multiscale equiaxed dendritic solidification model based on the volume-averaging method, which accounts for the above-mentioned transport phenomena. The validity of the model is assessed by comparing it with the full three-phase model without simplifications. It is then applied to qualitatively analyze the impact of grain morphology on macrosegregation formation in an industrial scale direct chill cast aluminum alloy ingot. [ABSTRACT FROM AUTHOR]

In this paper, we shall explain a new result concerning piecewise weighted pseudo almost-periodic solution of impulsive recurrent neural networks with time-varying coefficients and mixed delays. Precisely, some sufficient conditions are given to prove the existence and the exponential stability of piecewise weighted pseudo almost-periodic solution by employing fixed point theorem, generalized Gronwall-Bellman inequality and differential inequality techniques. Finally, an illustrative example is given to demonstrate the effectiveness of our results. [ABSTRACT FROM AUTHOR]

Existence of piecewise differentiable pseudo almost-periodic solutions for a class of impulsive high-order Hopfield neural networks with leakage delays are established by employing the fixed point theorem, differential inequality and Lyapunov functionals. The results of this paper are new and they supplement previously known works. Numerical example with graphical illustration is given to illuminate our main results. [ABSTRACT FROM AUTHOR]

A two-phase mathematical model for the study of hot tearing formation is presented. The model accounts for the main phenomena associated with the formation of hot tears, i.e., the lack of feeding at the late stages of solidification and the localization of viscoplastic deformation. The model incorporates an advanced viscoplastic constitutive model for the coherent part of the mushy zone, allowing for the possibility of dilatation/densification of the semisolid skeleton under applied deformation. Based on quantities computed by the model, a hot tearing criterion is proposed where liquid feeding difficulties and viscoplastic deformation at the late stages of solidification are taken into account. The model is applied to study hot tearing formation during the start-up phase for direct-chill (DC) casting of extrusion ingots, and to discuss the effect of different phenomena and process parameters. The modeling results are also compared to experimentally measured hot tearing susceptibilities, and the model is able to reproduce known experimental trends such as the effect of the casting speed and the importance of the design of the starting block. [ABSTRACT FROM AUTHOR]

A volume-averaged two-phase model addressing the main transport phenomena associated with hot tearing in an isotropic mushy zone during solidification of metallic alloys has recently been presented elsewhere along with a new hot tearing criterion addressing both inadequate melt feeding and excessive deformation at relatively high solid fractions. The viscoplastic deformation in the mushy zone is addressed by a model in which the coherent mush is considered as a porous medium saturated with liquid. The thermal straining of the mush is accounted for by a recently developed model taking into account that there is no thermal strain in the mushy zone at low solid fractions because the dendrites then are free to move in the liquid, and that the thermal strain in the mushy zone tends toward the thermal strain in the fully solidified material when the solid fraction tends toward one. In the present work, the authors determined how variations in the parameters of the constitutive equation for thermal strain influence the hot tearing susceptibility calculated by the criterion. It turns out that varying the parameters in this equation has a significant effect on both liquid pressure drop and viscoplastic strain, which are key parameters in the hot tearing criterion. However, changing the parameters in this constitutive equation will result in changes in the viscoplastic strain and the liquid pressure drop that have opposite effects on the hot tearing susceptibility. The net effect on the hot tearing susceptibility is thus small. [ABSTRACT FROM AUTHOR]

A two-phase volume-averaged continuum model is presented that quantifies macrosegregation formation during solidification of metallic alloys caused by deformation of the dendritic network and associated melt flow in the coherent part of the mushy zone. Also, the macrosegregation formation associated with the solidification shrinkage (inverse segregation) is taken into account. Based on experimental evidence established elsewhere, volumetric viscoplastic deformation (densification/dilatation) of the coherent dendritic network is included in the model. While the thermomechanical model previously outlined (M. M'Hamdi, A. Mo, and C.L. Martin: Metall. Mater. Trans. A, 2002, vol. 33A, pp. 2081-93) has been used to calculate the temperature and velocity fields associated with the thermally induced deformations and shrinkage driven melt flow, the solute conservation equation including both the liquid and a solid volume-averaged velocity is solved in the present study. In modeling examples, the macrosegregation formation caused by mechanically imposed as well as by thermally induced deformations has been calculated. The modeling results for an A1-4 wt pct Cu alloy indicate that even quite small volumetric strains (≈2 pct), which can be associated with thermally induced deformations, can lead to a macroscopic composition variation in the final casting comparable to that resulting from the solidification shrinkage induced melt flow. These results can be explained by the relatively large volumetric viscoplastic deformation in the coherent mush resulting from the applied constitutive model, as well as the relatively large difference in composition for the studied Al-Cu alloy in the solid and liquid phases at high solid fractions at which the deformation takes place. [ABSTRACT FROM AUTHOR]

A constitutive equation for thermal strain in the mushy zone during solidification of aluminum alloys has been determined based on a two-phase volume-averaged description of the mushy zone. The change of linear dimensions in the horizontal plane of cast samples during solidification is investigated experimentally. The temperature when the solid part of the mushy zone starts to contract, and the total contraction of the solid part of the mushy zone, are determined experimentally. The constitutive model for thermal strain in the mushy zone reflects that there is no thermal strain in the solid part of the mushy zone at a low solid fraction and that the thermal strain in the mushy zone approaches thermal contraction in fully solid as the solid fraction increases towards one. The parameters in the constitutive model for thermal strain in the mushy zone are determined by combining experimental results with thermomechanical simulations of the experiments. The heat-transfer coefficients and the parameters in the model are tuned to make the simulations reproduce the experimentally determined temperature field and contraction. Al-Cu alloys with Cu concentrations from 0.3 to 9 wt pct Cu and an Al-7 wt pct Si-0.3 wt pct Mg alloy are tested both with and without grain refinement and at various cooling rates. The solid fractions when the alloys start to contract (gsth) are in the range of 0.48 to 0.97. A lower solute concentration, grain refinement, and higher cooling rate increase gsth. [ABSTRACT FROM AUTHOR]