Yaru Song^*, Huahu Xu, Dikai Fang

Intelligent Automation & Soft Computing, Vol.39, No.3, pp. 397-416, 2024, DOI:10.32604/iasc.2023.040521

Abstract Visual Question Answering (VQA) has sparked widespread interest as a crucial task in integrating vision and language. VQA primarily uses attention mechanisms to effectively answer questions to associate relevant visual regions with input questions. The detection-based features extracted by the object detection network aim to acquire the visual attention distribution on a predetermined detection frame and provide object-level insights to answer questions about foreground objects more effectively. However, it cannot answer the question about the background forms without detection boxes due to the lack of fine-grained details, which is the advantage of grid-based features. In… More >

Huiyan Zhao¹, Xuezhong Chen¹, Zhijian Hu^2,*, Man Chen¹, Bo Xiong³, Jianying Yang¹

FDMP-Fluid Dynamics & Materials Processing, Vol.20, No.6, pp. 1313-1330, 2024, DOI:10.32604/fdmp.2024.048840

Abstract Using the typical characteristics of multi-layered marine and continental transitional gas reservoirs as a basis, a model is developed to predict the related well production rate. This model relies on the fractal theory of tortuous capillary bundles and can take into account multiple gas flow mechanisms at the micrometer and nanometer scales, as well as the flow characteristics in different types of thin layers (tight sandstone gas, shale gas, and coalbed gas). Moreover, a source-sink function concept and a pressure drop superposition principle are utilized to introduce a coupled flow model in the reservoir. A… More > Graphic Abstract

Emran Al-Buraihy, Dan Wang^*

CMC-Computers, Materials & Continua, Vol.79, No.3, pp. 3913-3938, 2024, DOI:10.32604/cmc.2024.048104

Abstract Cross-lingual image description, the task of generating image captions in a target language from images and descriptions in a source language, is addressed in this study through a novel approach that combines neural network models and semantic matching techniques. Experiments conducted on the Flickr8k and AraImg2k benchmark datasets, featuring images and descriptions in English and Arabic, showcase remarkable performance improvements over state-of-the-art methods. Our model, equipped with the Image & Cross-Language Semantic Matching module and the Target Language Domain Evaluation module, significantly enhances the semantic relevance of generated image descriptions. For English-to-Arabic and Arabic-to-English cross-language… More >

Abderrezzaq Benalia^1,2,*, Kerroum Derbal², Zahra Amrouci^2,3, Ouiem Baatache², Amel Khalfaoui⁴, Antonio Pizzi^5,*

Journal of Renewable Materials, Vol.12, No.4, pp. 667-698, 2024, DOI:10.32604/jrm.2024.048306

Abstract This review describes the mechanisms of natural coagulants. It provides a good understanding of the two key processes of coagulation-flocculation: adsorption and charge neutralization, as well as adsorption and bridging. Various factors have influence the coagulation/flocculation process, including the effect of pH, coagulant dosage, coagulant type, temperature, initial turbidity, coagulation speed, flocculation speed, coagulation and flocculation time, settling time, colloidal particles, zeta potential, the effects of humic acids, and extraction density are explained. The bio-coagulants derived from plants are outlined. The impact of organic coagulants on water quality, focusing on their effects on the physicochemical… More > Graphic Abstract

CARLOS ALBERTO-SILVA^*, BRENDA RUFINO DA SILVA

BIOCELL, Vol.48, No.6, pp. 897-904, 2024, DOI:10.32604/biocell.2024.050443

Abstract Venom snake-derived peptides have multiple biochemical, pharmacological, and toxicological profiles, allowing for the discovery of new medicinal products and therapeutic applications. This review specifically examines the fundamental elements of neuroprotection offered by different oligopeptides derived from snake venom. It also includes a brief evaluation of short peptides that are being considered as potential therapeutic agents. Proline-rich peptides and tryptophyllin family peptides isolated from the crude venom of Viperidae family snakes, specifically Bothrops atrox, Bothrops jararaca, and Bothrops moojeni, have been shown to have pro-survival properties, the ability to reduce oxidative stress, and the ability to promote cell viability More >

MINGYUE HUANG^1,#, JUNFEI DAI^1,#, XIANLIANG ZHONG², JIN WANG², JIANZHONG XU², BO DU^2,*

BIOCELL, Vol.48, No.6, pp. 889-896, 2024, DOI:10.32604/biocell.2024.050097

Abstract Chronic subdural hematoma (CSDH) is a disease characterized by capsuled blood products that progressively occupy the intracranial space, causing intracranial hypertension and compression in the brain. CSDH frequently occurs in all demographics, especially in the elderly, but the pathogenesis of CSDH remains unclear. In this review, we discuss the origin, development, and current treatment strategies of CSDH. For the first time, we analyzed the cellular and molecular compositions of hematoma membranes with a focus on neomembrane formation, a complex early-stage interactive event in hematoma pathogenesis. We hypothesize that in patients with CSDH, dural border cells… More >

MOHD ATHAR^1,*, ANU MANHAS², NISARG RANA², AHMAD IRFAN³

BIOCELL, Vol.48, No.6, pp. 935-944, 2024, DOI:10.32604/biocell.2024.049891

Abstract Computational methods have significantly transformed biomedical research, offering a comprehensive exploration of disease mechanisms and molecular protein functions. This article reviews a spectrum of computational tools and network analysis databases that play a crucial role in identifying potential interactions and signaling networks contributing to the onset of disease states. The utilization of protein/gene interaction and genetic variation databases, coupled with pathway analysis can facilitate the identification of potential drug targets. By bridging the gap between molecular-level information and disease understanding, this review contributes insights into the impactful utilization of computational methods, paving the way for More >

Zhenxiang He^*, Zhenyu Zhao, Ke Chen, Yanlin Liu

CMC-Computers, Materials & Continua, Vol.79, No.2, pp. 3023-3045, 2024, DOI:10.32604/cmc.2024.050281

Abstract The fast-paced development of blockchain technology is evident. Yet, the security concerns of smart contracts represent a significant challenge to the stability and dependability of the entire blockchain ecosystem. Conventional smart contract vulnerability detection primarily relies on static analysis tools, which are less efficient and accurate. Although deep learning methods have improved detection efficiency, they are unable to fully utilize the static relationships within contracts. Therefore, we have adopted the advantages of the above two methods, combining feature extraction mode of tools with deep learning techniques. Firstly, we have constructed corresponding feature extraction mode for… More >

Amal Hkiri^1,*, Hamzah Faraj², Omar Ben Bahri², Mouna Karmani¹, Sami Alqurashi², Mohsen Machhout¹

CMC-Computers, Materials & Continua, Vol.79, No.2, pp. 1835-1854, 2024, DOI:10.32604/cmc.2024.049207

Abstract The Internet of Things (IoT) has witnessed a significant surge in adoption, particularly through the utilization of Wireless Sensor Networks (WSNs), which comprise small internet-connected devices. These deployments span various environments and offer a multitude of benefits. However, the widespread use of battery-powered devices introduces challenges due to their limited hardware resources and communication capabilities. In response to this, the Internet Engineering Task Force (IETF) has developed the IPv6 Routing Protocol for Low-power and Lossy Networks (RPL) to address the unique requirements of such networks. Recognizing the critical role of RPL in maintaining high performance,… More >

Chunhua Wang^1,2, Wenqian Shang^1,2,*, Tong Yi^3,*, Haibin Zhu⁴

CMC-Computers, Materials & Continua, Vol.79, No.2, pp. 1939-1956, 2024, DOI:10.32604/cmc.2024.048135

Abstract The advent of self-attention mechanisms within Transformer models has significantly propelled the advancement of deep learning algorithms, yielding outstanding achievements across diverse domains. Nonetheless, self-attention mechanisms falter when applied to datasets with intricate semantic content and extensive dependency structures. In response, this paper introduces a Diffusion Sampling and Label-Driven Co-attention Neural Network (DSLD), which adopts a diffusion sampling method to capture more comprehensive semantic information of the data. Additionally, the model leverages the joint correlation information of labels and data to introduce the computation of text representation, correcting semantic representation biases in the data, and More >