Search

Simple Summary: The classification is performed later by an interactively learning Swin Transformer block, the core unit for feature representation and long-range semantic information. In particular, the proposed strategy improved significantly and was very resilient while dealing with small liver pieces, discontinuous liver regions, and fuzzy liver boundaries. The experimental results confirm that the proposed APESTNet is more effective in classifying liver tumours than the current state-of-the-art models. Without compromising accuracy, the proposed method conserved resources. However, the proposed method is prone to slight over-segmentation or under-segmentation errors when dealing with lesions or tumours at the liver boundary. Therefore, our future work will concentrate on completely utilizing the z-axis information in 3D to reduce errors. Diagnosis and treatment of hepatocellular carcinoma or metastases rely heavily on accurate segmentation and classification of liver tumours. However, due to the liver tumor's hazy borders and wide range of possible shapes, sizes, and positions, accurate and automatic tumour segmentation and classification remains a difficult challenge. With the advancement of computing, new models in artificial intelligence have evolved. Following its success in Natural language processing (NLP), the transformer paradigm has been adopted by the computer vision (CV) community of the NLP. While there are already accepted approaches to classifying the liver, especially in clinical settings, there is room for advancement in terms of their precision. This paper makes an effort to apply a novel model for segmenting and classifying liver tumours built on deep learning. In order to accomplish this, the created model follows a three-stage procedure consisting of (a) pre-processing, (b) liver segmentation, and (c) classification. In the first phase, the collected Computed Tomography (CT) images undergo three stages of pre-processing, including contrast improvement via histogram equalization and noise reduction via the median filter. Next, an enhanced mask region-based convolutional neural networks (Mask R-CNN) model is used to separate the liver from the CT abdominal image. To prevent overfitting, the segmented picture is fed onto an Enhanced Swin Transformer Network with Adversarial Propagation (APESTNet). The experimental results prove the superior performance of the proposed perfect on a wide variety of CT images, as well as its efficiency and low sensitivity to noise. [ABSTRACT FROM AUTHOR]

Abstract: Epitaxial growth of SiGe alloy films under certain kinetically limiting conditions has previously been shown to lead to the self-assembly of nanostructures called Quantum Dot Molecules, QDMs. These QDMs consist of an assembly of pyramidal pits and elongated pyramidal islands, which exist at the edges of the pits. We investigate the nano-scale chemistry of QDMs in Si0.7Ge0.3/Si (100) using Auger electron spectroscopy (AES). First, our AES analysis indicates that compressively strained QDM pit bases are most Ge rich regions in QDMs, consistent with previously reported observations. Second, our Auger analysis of the QDMs shows that Ge composition continuously increases from outside edges of the islands to the pit cusps. This results in asymmetric distribution of Ge about the apex of the islands. The segregation of Ge to the pit cusps is unexpected on the basis of strain energy minimization, and we propose that it is due to attachment of Ge to steps on the interiors of the QDMs. The continually varying Ge composition towards the pit cusps is also inconsistent with previous observations of islands in GeSi growth, where the composition is reported to be symmetric about the apex of islands. [Copyright &y& Elsevier]

Microchannels and minichannels are being considered for high heat flux applications under microgravity environment in space missions. An experimental study is undertaken to determine the effect of gravitational orientation on flow boiling characteristics of water in a set of six parallel minichannels, each 1054 µm wide by 197 µm deep and 63.5 mm long with a hydraulic diameter of 333 µm. Three orientations--horizontal, vertical downflow, and vertical upflow--are investigated under identical operating conditions of heat and mass fluxes. High-speed images are obtained to reveal the detailed two-phase flow structure and liquid-vapor interactions. The experimental data and high speed flow visualization indicate that compared to the horizontal case, the flow becomes less chaotic for the vertical upflow case, while the reversed flow becomes more pronounced in the vertical downflow case. The resulting increase in the backflow is responsible for channel-to-channel flow maldistribution and heat transfer degradation. From the heat transfer data it is concluded that the performance of the tested channels in a microgravity environment will be similar to the horizontal flow case. [ABSTRACT FROM AUTHOR]

Flow boiling heat transfer in parallel minichannels and microchannels is one of the solutions proposed for cooling high heat flux .systems. The associated increase in the pressure drop poses a problem that needs to be studied in detail before the small diameter channels can he implemented in practical systems. The pressure drop fluctuations and the flow instability in a network of parallel channels connected by a common header also need to be addressed for the stable operation of flow boiling systems. The current work focuses on studying the flow patterns, pressure drop fluctuations, and flow instabilities in a set of six parallel rectangular minichannels, each with 333 μm in hydraulic diameter. Deionized and degassed water was used for all the experiments. The pressure fluctuations are recorded and signal analysis is performed to find the dominant frequencies and their amplitudes. These pressure fluctuations are then mapped to their corresponding flow patterns observed using a high speed camera. The results help us to relate pressure fluctuations to different flow characteristics and their effect on flow instability. [ABSTRACT FROM AUTHOR]

Flow boiling in mini- and microchannels offer very high heat transfer capabilities and find applications in many emerging technologies, such as electronics cooling and fuel cells. The low flow rate employed in such geometries, coupled with the small flow channels, often results in a laminar flow with all flow as liquid. Since the single-phase flow with all liquid is in the laminar range, the flow boiling correlations developed for conventional tubes with an inner diameter larger than 3 mm and turbulent flow need to be carefully reviewed. In the present work, flow boiling correlation for large diameter tubes developed by Kandlikar [1, 2] is modified for flow boiling in minichannels by using the laminar single-phase heat transfer coefficient for all liquid flow. The correlation is also extended for flow boiling in microchannels using the nucleate boiling as the dominant part of the original correlation. The trends in heat transfer coefficient versus quality are compared in the laminar and deep laminar regions in minichannels and microchannels. Excellent agreement is obtained between predicted values and experimental data. [ABSTRACT FROM AUTHOR]

Unbalanced load condition is one of the major issues of all commercial, industrial and residential sectors. Unbalanced load means that, when different loads are distributed on a three-phase four-wire system, unequal currents pass through the three phases. Due to it, a heavy current flows in the neutral wire, which not only adds the losses, but also puts constraints on three phases' loads. In this paper, we have presented a practical approach for load balancing. First, we have considered the existing three-phase load system where the supply is a three-phase unbalanced supply. Before balancing the load, it is necessary to compensate the current in neutral wire. A nature-inspired moth–flame optimization (MFO) algorithm is used to propose a scheme for balancing of current in neutral wire. The information of a distributed single-phase load was used to balance the currents in a three-phase system. The feeder phase and load profiles of each single-phase load are used to reconfigure the network using an optimization process. By balancing the current of three phases, the current of the neutral conductor in substation transformers was reduced to almost zero. [ABSTRACT FROM AUTHOR]

Abstract: We use focused beams of Ga+, Au+ and Si++ ions to induce local microstructural changes in single crystal silicon. The ions were delivered as single spot pulses into thin Si membranes that could subsequently be imaged and annealed in situ in a transmission electron microscope. For each ion, the focused ion beam implantation created an array of amorphous regions in the crystalline membrane. Annealing causes solid phase epitaxial regrowth to take place, but we show that the resulting microstructure depends on the ion species. For Ga+ and Au+, precipitates remain after recrystallization, while for Si++, dislocation loops form around the periphery of each implanted spot. We attribute these loops to defects formed during solid phase epitaxial regrowth, with controlled placement of the loops possible. [Copyright &y& Elsevier]

Decoupled data and control planes in Software Defined Networks (SDN) allow them to handle an increasing number of threats by limiting harmful network links at the switching stage. As storage, high-end servers, and network devices, Network Function Virtualization (NFV) is designed to replace purpose-built network elements with VNFs (Virtualized Network Functions). A Software Defined Network Function Virtualization (SDNFV) network is designed in this paper to boost network performance. Stateful firewall services are deployed as VNFs in the SDN network in this article to offer security and boost network scalability. The SDN controller's role is to develop a set of guidelines and rules to avoid hazardous network connectivity. Intruder assaults that employ numerous socket addresses cannot be adequately protected by these strategies. Machine learning algorithms are trained using traditional network threat intelligence data to identify potentially malicious linkages and probable attack targets. Based on conventional network data (DT), Bayesian Network (BayesNet), Naive-Bayes, C4.5, and Decision Table (DT) algorithms are used to predict the target host that will be attacked. The experimental results shows that the Bayesian Network algorithm achieved an average prediction accuracy of 92.87%, Native–Bayes Algorithm achieved an average prediction accuracy of 87.81%, C4.5 Algorithm achieved an average prediction accuracy of 84.92%, and the Decision Tree algorithm achieved an average prediction accuracy of 83.18%. There were 451 k login attempts from 178 different countries, with over 70 k source IP addresses and 40 k source port addresses recorded in a large dataset from nine honeypot servers. [ABSTRACT FROM AUTHOR]