Diverse Domains Research Articles

Exploring Applications of Artificial Intelligence in Critical Care Nursing: A Systematic Review

Background: Artificial intelligence (AI) has been increasingly employed in healthcare across diverse domains, including medical imaging, personalized diagnostics, therapeutic interventions, and predictive analytics using electronic health records. Its integration is particularly impactful in critical care, where AI has demonstrated the potential to enhance patient outcomes. This systematic review critically evaluates the current applications of AI within the domain of critical care nursing. Methods: This systematic review is registered with PROSPERO (CRD42024545955) and was conducted in accordance with PRISMA guidelines. Comprehensive searches were performed across MEDLINE/PubMed, SCOPUS, CINAHL, and Web of Science. Results: The initial review identified 1364 articles, of which 24 studies met the inclusion criteria. These studies employed diverse AI techniques, including classical models (e.g., logistic regression), machine learning approaches (e.g., support vector machines, random forests), deep learning architectures (e.g., neural networks), and generative AI tools (e.g., ChatGPT). The analyzed health outcomes encompassed postoperative complications, ICU admissions and discharges, triage assessments, pressure injuries, sepsis, delirium, and predictions of adverse events or critical vital signs. Most studies relied on structured data from electronic medical records, such as vital signs and laboratory results, supplemented by unstructured data, including nursing notes and patient histories; two studies also integrated audio data. Conclusion: AI demonstrates significant potential in nursing, facilitating the use of clinical practice data for research and decision-making. The choice of AI techniques varies based on the specific objectives and requirements of the model. However, the heterogeneity of the studies included in this review limits the ability to draw definitive conclusions about the effectiveness of AI applications in critical care nursing. Future research should focus on more robust, interventional studies to assess the impact of AI on nursing-sensitive outcomes. Additionally, exploring a broader range of health outcomes and AI applications in critical care will be crucial for advancing AI integration in nursing practices.

Automated orthodontic diagnosis via self-supervised learning and multi-attribute classification using lateral cephalograms

BackgroundMalocclusion, characterized by dental misalignment and improper occlusal relationships, significantly impacts oral health and daily functioning, with a global prevalence of 56%. Lateral cephalogram is a crucial diagnostic tool in orthodontic treatment, providing insights into various structural characteristics.MethodsThis study introduces a pre-training approach using multi-center lateral cephalograms for self-supervised learning, aimed at improving model generalization across diverse clinical data domains. Additionally, a multi-attribute classification network is proposed, leveraging attribute correlations to optimize parameters and enhance classification performance.ResultsComprehensive evaluation on both public and clinical datasets showcases the superiority of the proposed framework, achieving an impressive average accuracy of 90.02%. The developed Self-supervised Pre-training and Multi-Attribute (SPMA) network achieves a best match ratio (MR) score of 71.38% and a low Hamming loss (HL) of 0.0425%, demonstrating its efficacy in orthodontic diagnosis from lateral cephalograms.ConclusionsThis work contributes significantly to advancing automated diagnostic tools in orthodontics, addressing the critical need for accurate and efficient malocclusion diagnosis. The outcomes not only improve the efficiency and accuracy of diagnosis, but also have the potential to reduce healthcare costs associated with orthodontic treatments.

Modern Approaches to Protein Constructions: A Comprehensive Review of Computational Tools and Databases for De Novo Protein Design and Engineering

ABSTRACTThe field of protein engineering has witnessed transformative advancements, with computational tools and databases driving novel innovations in de novo protein design. This review consolidates and critiques a comprehensive range of modern computational resources, offering a unique focus on their applications across diverse domains, including protein stability prediction, posttranslational modification analysis, and mutation effect evaluation. Key contributions include a detailed examination of tools integrating machine learning and artificial intelligence to enhance predictive accuracy and streamline protein engineering workflows. By highlighting underexplored tools and novel methodologies, such as advanced protein–ligand interaction predictors and neural network–based stability assessment models, this study establishes itself as a unique reference for researchers aiming to develop tailored proteins for therapeutic, industrial, and biomedical applications.

Processing and consolidation of open data on public procurement in France (2015-2023).

Public procurement can be defined as the process by which public contracting authorities purchase goods, services, and works from private suppliers. To ensure transparency and prevent favoritism and corruption, public contracts adhere to strict procedures based on calls for tenders and award notices, which are publicly accessible online. This paper introduces a tabular dataset derived from the processing and consolidation of online public procurement notices. It provides a detailed overview of public contracts awarded in France between 2015 and 2023. The dataset encompasses approximately one million contractual relationships between public authorities and companies, in over 300,000 contracts, spanning all sectors and public institutions. It includes 113 variables covering contract characteristics (procedure, subject matter, award criteria, clauses, etc.), award outcomes (award price, number of bids, etc.), as well as information on contracting authorities (type, location, main activity, etc.) and awarded firms (size, legal status, main activity, age, location, etc.). This unprecedented dataset, both in accuracy and scope, provides reliable and detailed information on every advertised contract in France for nearly a decade, making it valuable for empirical research in diverse domains such as economics, geography, law, and political science.

BCA: Besiege and Conquer Algorithm

This paper introduces a bio-inspired meta-heuristic algorithm, the Besiege and Conquer Algorithm (BCA), developed to tackle complex and high-dimensional optimization problems. Drawing inspiration from the concept of symmetry and guerrilla warfare strategies, the BCA incorporates four core components: besiege, conquer, balance, and feedback. The besiege strategy strengthens exploration, while the conquer strategy enhances exploitation. Balance and feedback mechanisms maintain a dynamic equilibrium between these capabilities, ensuring robust optimization performance. The algorithm’s effectiveness is validated through benchmark test functions, demonstrating superior results in comparison with existing methods, supported by Friedman rankings and Wilcoxon signed-rank tests. Beyond theoretical and experimental validation, the BCA showcases its real-world relevance through applications in engineering design and classification problems, addressing practical challenges. These results underline the algorithm’s strong exploration, exploitation, and convergence capabilities and its potential to contribute meaningfully to diverse real-world domains.

Encapsulation of fluorescent carbon dots into mesoporous SiO2 colloidal spheres by surface functionalization-assisted cooperative assembly for high-contrast latent fingerprint development.

Exploiting solid powder fluorescence holds significant potential in diverse domains including medicine and forensics. Conventional fingerprint detection methods often fall short due to low contrast, sensitivity, and high toxicity. To addressing these challenges, we present a novel method for latent fingerprint detection using fluorescent carbon dots (CDs) encapsulated into mesoporous SiO2 colloidal spheres through a surface functionalization-assisted cooperative assembly. The synthesized monodisperse CDs@SiO2 and CDs@m-SiO2 spheres, with their adjustable porosity and size, improve fingerprint visibility and contrast on various substrates such as glass, stainless steel, and plastic. Our method demonstrates a high contrast, selectivity, and sensitivity in fingerprint detection, offering an environmentally friendly and healthy alternative to conventional techniques. The results indicate potential applications beyond forensics, including in optoelectronic and biomedical fields, showcasing a significant step forward in materials science for forensic analysis.

KnowbasAI: Knowledge-Based AI system for Custom Recommendations

Ontology-based recommendation engines are a type of recommendation system that incorporates ontologies to enhance the recommendation process. An ontology is a formal representation of knowledge that defines the relationships between various concepts within a specific domain. In the context of recommendation engines, an ontology organizes and structures information about users, items, and their attributes, facilitating more intelligent and context-aware recommendations. KnowbasAI is an innovative Knowledge-Based AI System that revolutionizes custom recommendations in diverse domains. Leveraging a comprehensive ontology-driven knowledge base, KnowbasAI captures structured and unstructured data, including user interactions, product descriptions, and domain-specific knowledge. By assimilating this wealth of information, the system gains a profound understanding of user preferences and item attributes. KnowbasAI's recommendation engine uses a hybrid strategy that combines content-based filtering with collaborative filtering techniques. While content-based filtering aligns user preferences with pertinent items, collaborative filtering discovers user commonalities. Even for inexperienced or specialized users, the integration of various approaches guarantees precise and varied recommendations. KnowbasAI's clear and understandable suggestion approach is a key benefit. By using cutting-edge methods, the system offers insightful analyses of the decision-making process, boosting user confidence in the suggestions. With about 90% accuracy, it assists in achieving privacy and personalized recommendations for the users.

Ants’ Collective Intelligence: What Could We Learn?

Ants, albeit appearing diminutive and trivial within the broader context of nature, demonstrate an extraordinary capacity for collective problem-solving and exhibit intellect that beyond expectations for individual members of their species. This phenomenon, termed ants’ collective intelligence, has elicited attention and appreciation from both experts and enthusiasts. This article examines the complex mechanisms of ants’ collective intelligence, highlighting its essential features, ramifications across diverse domains, and the significant insights it can provide to human. Embark on an intriguing exploration of the ant kingdom, where collaboration, communication, and decentralized decision-making coalesce to create a complex system that offers significant lessons for human pursuits.

A novel cyber threat intelligence platform for evaluating the risk associated with smart agriculture

The rapid proliferation of Internet of Things (IoT) devices has brought about a profound transformation in our daily lives and work environments. However, this proliferation has also given rise to significant security challenges, as cybercriminals increasingly target IoT devices to exploit vulnerabilities and gain access to sensitive data. This escalating threat landscape poses a severe issue across diverse domains where IoT is deployed, including agriculture, healthcare, and surveillance. In the realm of agriculture, where farmers have historically contended with pests and environmental challenges, a new adversary has emerged in the form of cyber criminals. The agriculture sector has witnessed a surge in cyber-attacks targeting smart agriculture solutions despite being a relatively recent addition to the industry. Farmers may not have control over the actions of cyber adversaries, but they possess the ability to make informed purchasing decisions when adopting smart farming solutions and implementing fundamental security measures, such as robust user credentials and regular system updates. In this regard, this research introduces a groundbreaking approach to addressing the cybersecurity concerns associated with smart agriculture–deception technology. Overall, deception technology involves the creation of deceptive elements, including decoys, traps, and false information, designed to divert cybercriminals away from genuine data and systems where this research presents a novel cyber threat intelligence platform that leverages deception technology to assess and mitigate the risks associated with smart agriculture as the first of its kind research. Based on the insights derived from the experimental work, actionable recommendations would be provided to relevant stakeholders on how to mitigate cyber risks and bolster the security posture of IoT-enabled smart agriculture. Overall, this innovative approach represents a significant step towards safeguarding the increasingly interconnected world of smart agriculture, offering a promising avenue for defending against the escalating cyber threats faced by this vital industry.

Effect of micro and nanoemulsions on wound healing: a review

Emulsions, categorised by their particle size into microemulsions (<100 nm) and nanoemulsions (<200 to <100 nm), hold a position of burgeoning significance in the field of wound healing. Beyond their established role in diverse medical domains, emulsions have emerged as promising agents for drug delivery with wound-healing properties. This review explores a comprehensive analysis of the available literature to shed light on the impact of micro and nanoemulsions on wound healing. Covering research articles published from 2011 to 2022, this study meticulously follows the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) guidelines while employing bibliometric analysis tools, PRISMA, and VOSviewer. Through an extensive exploration of databases such as PubMed, Scopus, and ScienceDirect, this study identifies and includes 26 articles that align with the rigorous criteria for investigating emulsions in wound healing. Our bibliometric analysis underscores 'wound healing' as the most prominent keyword in this corpus. In conclusion, this study contributes substantially to the research by combining insights into the effects of nano and microemulsions on wound healing, offering a foundation for future investigations and applications within emerging nanotechnology.

An automatic approach for adapting open government data to linked OD with enhanced visualization and user-friendly query composer

Purpose This study aims to introduce a fully automatic process for publishing and leveraging open government data (OGD) as linked OD (LOD), called the LOD-GePEx for LOD generation, publication and exploitation. It empowers developers to harness open-source data, creating practical and beneficial citizen applications. Design/methodology/approach The LOD-GePEx approach is a three-step process. First, it transforms OGD into LOD, adhering to the four LD principles. Second, it publishes the generated LOD in the OGD portal, following accessibility best practices. Third, it exploits the published LOD through an interface that offers multiple services. In addition, this paper carried out functional and performance tests to prove the proposed approach’s effectiveness and performance. Findings The evaluation phase demonstrated that the LOD-GePEx approach is effective and efficient. It transforms OGD into LOD with exploitation services to enable insightful data analysis for stakeholders without intensive human workloads. Besides, it is generic; it can be used for several data sets and any case study. Research limitations/implications The approach struggles with speed index and interactivity time in resource description framework and LOD publication generation, particularly with large government data sets. However, this work has prioritized the comprehensive exploration of the end-to-end process and handling of diverse data domains. This work implies that adopting this approach will facilitate the publication and exploitation of OGD. Originality/value The proposed approach surpasses the limitations of related works, mainly focusing on particular use cases and manual data linking, which require intensive human workloads and few or no data insights. Besides, the proposed approach is the only approach that addresses the end-to-end process of transforming, publishing and exploiting OGDs as LODs.

Enhancing Food Image Recognition by Multi-Level Fusion and the Attention Mechanism

As a pivotal area of research in the field of computer vision, the technology for food identification has become indispensable across diverse domains including dietary nutrition monitoring, intelligent service provision in restaurants, and ensuring quality control within the food industry. However, recognizing food images falls within the domain of Fine-Grained Visual Classification (FGVC), which presents challenges such as inter-class similarity, intra-class variability, and the complexity of capturing intricate local features. Researchers have primarily focused on deep information in deep convolutional neural networks for fine-grained visual classification, often neglecting shallow and detailed information. Taking these factors into account, we propose a Multi-level Attention Feature Fusion Network (MAF-Net). Specifically, we use feature maps generated by the Convolutional Neural Networks (CNNs) backbone network at different stages as inputs. We apply a self-attention mechanism to identify local features on these feature maps and then stack them together. The feature vectors obtained through the attention mechanism are then integrated with the original input to enhance data augmentation. Simultaneously, to capture as many local features as possible, we encourage multi-scale features to concentrate on distinct local regions at each stage by maximizing the Kullback-Leibler Divergence (KL-divergence) between the different stages. Additionally, we present a novel approach called subclass center loss (SCloss) to implement label smoothing, minimize intra-class feature distribution differences, and enhance the model’s generalization capability. Experiments conducted on three food image datasets—CETH Food-101, Vireo Food-172, and UEC Food-100—demonstrated the superiority of the proposed model. The model achieved Top-1 accuracies of 90.22%, 89.86%, and 90.61% on CETH Food-101, Vireo Food-172, and UEC Food-100, respectively. Notably, our method not only outperformed other methods in terms of the Top-5 accuracy of Vireo Food-172 but also achieved the highest performance in the Top-1 accuracies of UEC Food-100.

Normative data for teleneuropsychological testing: findings from a Canadian adult cohort

Objective Use of teleneuropsychological services has greatly increased since the beginning of the COVID-19 pandemic. The present study aimed to create normative data for a neuropsychological test battery of diverse cognitive domains in a Canadian population. Method A sample (n = 291) of adults aged 19 or older completed a comprehensive neuropsychological assessment (i.e. memory, executive function, etc.) via Zoom. Participants included those with a COVID-19 diagnosis (n = 146) and participants who had not contracted COVID-19 (n = 145). Data were stratified by age group as follows: 19–34, 35–49, 50–64, 65–79. Linear bivariate regression in the entire sample and groups stratified by age was used to test the relationship between age and test scores. Test scores were converted to z-scores using the mean and standard deviation for that group, with z-scores then transformed into normative scores for each test. Results Age was a significant predictor of scores for all tests except for FAS, HVLT-R (Retention, Recognition), and Digit Span (Forwards, Backwards). After raw test scores were regressed onto age for each group, age was no longer a significant predictor for most test scores, with exceptions for each age group. Conclusions This study created normative data for a diverse teleneuropsychological test battery in a Canadian population. Standard­ized scores generally fell within the average range, with the exception of TOPF and JLO scores, which may be explained by high educational attainment and virtual testing environment, respectively. The results suggest that the teleneuropsychological testing environment results in similar performance to in-person assessment.

Machine Learning Interatomic Potential for Modeling the Mechanical and Thermal Properties of Naphthyl-Based Nanotubes.

Two-dimensional (2D) nanomaterials are at the forefront of potential technological advancements. Carbon-based materials have been extensively studied since synthesizing graphene, which revealed properties of great interest for novel applications across diverse scientific and technological domains. New carbon allotropes continue to be explored theoretically, with several successful synthesis processes for carbon-based materials recently achieved. In this context, this study investigates the mechanical and thermal properties of DHQ-based monolayers and nanotubes, a carbon allotrope characterized by 4-, 6-, and 10-membered carbon rings, with a potential synthesis route using naphthalene as a molecular precursor. A machine-learned interatomic potential (MLIP) was developed to explore this nanomaterial's mechanical and thermal behavior at larger scales than those accessible through the first-principles calculations. The MLIP was trained on data derived from the DFT/PBE (density functional theory/Perdew-Burke-Ernzerhof) level using ab initio molecular dynamics (AIMD). Classical molecular dynamics (CMD) simulations, employing the trained MLIP, revealed that Young's modulus of DHQ-based nanotubes ranges from 127 to 243 N/m, depending on chirality and diameter, with fracture occurring at strains between 13.6 and 17.4% of the initial length. Regarding thermal response, a critical temperature of 2200 K was identified, marking the onset of a transition to an amorphous phase at higher temperatures.

Soliton solutions of the (2 + 1)-dimensional Jaulent-Miodek evolution equation via effective analytical techniques

In this study, we investigate the -D Jaulent-Miodek (JM) equation, which is significant due to its energy-based Schrödinger potential and applications in fields such as optics, soliton theory, signal processing, geophysics, fluid dynamics, and plasma physics. Given its broad utility, a rigorous mathematical analysis of the JM equation is essential. The primary objective of this work is to derive exact soliton solutions using the Modified Sub-Equation (MSE) and Modified Auxiliary Equation (MAE) techniques. These solutions are computed using Maple 18, and encompass a variety of wave structures, including bright solitons, kink solitons, periodic waves, and singular solitons. The potential applications of these solutions span diverse domains, such as nonlinear dynamics, fiber optics, ocean engineering, software engineering, electrical engineering, and other areas of physical science. Through numerical simulations, we visualize the physical characteristics of the obtained soliton solutions using three distinct graphical formats: 3D surface plots, 2D contour plots, and line plots, based on the selection of specific parameter values. Our results demonstrate that the MSE and MAE techniques are not only efficient but also straightforward in extracting soliton solutions for the JM equation, outperforming other existing methods. Furthermore, the solutions presented in this study are novel, representing contributions that have not been previously reported in the literature.

Bi-Partitioned Feature-Weighted K-Means Clustering for Detecting Insurance Fraud Claim Patterns

The weighted K-means clustering algorithm is widely recognized for its ability to assign varying importance to features in clustering tasks. This paper introduces an enhanced version of the algorithm, incorporating a bi-partitioning strategy to segregate feature sets, thus improving its adaptability to high-dimensional and heterogeneous datasets. The proposed bi-partition weighted K-means (BPW K-means) clustering approach is tailored to address challenges in identifying patterns within datasets with distinct feature subspaces, such as those in insurance claim fraud detection. Experimental evaluations on real-world insurance datasets highlight significant improvements in both clustering accuracy and interpretability compared to the classical K-means, achieving an accuracy of approximately 91%, representing an improvement of about 38% over the classical K-means algorithm. Moreover, the method’s ability to uncover meaningful fraud-related clusters underscores its potential as a robust tool for fraud detection. Beyond insurance, the proposed framework applies to diverse domains where data heterogeneity demands refined clustering solutions. The application of the BPW K-means method to multiple real-world datasets highlights its clear superiority over the classical K-means algorithm.

Integrating control-theoretic predictive deep learning for resource-efficient computing systems

Abstract Deep learning (DL) powers numerous applications on ubiquitous edge devices, but its high resource demands pose a challenge. Approximate computing is often proposed to alleviate this, yet such calculation usually suffers from a fixed level of accuracy loss. We propose a novel control-theoretic approach for predictive DL inference on resource constrained devices. Our system dynamically adjusts approximation levels based on a trade-off between resource utilisation and accuracy, considering future demands. Extensive experiments across diverse domains - human activity recognition, acoustic scene profiling, and computer vision - with various neural network architectures and approximation techniques, demonstrate that our approach achieves up to 50% energy savings while maintaining the desired inference accuracy and incurring minimal runtime overhead. Furthermore, we showcase our method in a real-world deployment on low-power edge devices and confirm its superiority over current state-of-the-art solutions.

A Comparative Analysis of Predictive Monitoring Systems Utilizing Machine Learning and Deep Learning Algorithms

This study offers an in-depth comparative examination of predictive monitoring systems utilizing sophisticated machine learning (ML) and deep learning (DL) algorithms. The investigation delves into the efficacy, advantages, and constraints of ML algorithms, exemplified by Random Forest and Extreme Gradient Boosting will be contrasted with Deep Learning (DL) algorithms including artificial neural networks (ANNs) and LSTM, recurrent neural networks (RNNs) in this study. By evaluating these algorithms across diverse domains, the research aims to discern optimal strategies for predictive monitoring, considering factors like efficiency, real-time processing, and adaptability. The findings contribute valuable insights for practitioners and researchers, informing the selection and deployment of algorithms in predictive monitoring systems

A Survey on WiFi-based Human Identification: Scenarios, Challenges, and Current Solutions

With the evolution of wireless sensing technology, WiFi-based human identification has demonstrated tremendous potential in human-computer interaction and home security. However, most existing research operates in controlled environments, overlooking the complexities of real-world scenarios, such as signal fading, signal interference and uncertainty, and the diversity of application requirements. This article presents a comprehensive analysis of a series of representative research articles on WiFi-based human identification and summarizes five major challenges in achieving high-precision identification in complex application scenarios. Non-line-of-sight (NLOS) user sensing, coexistence user sensing, dynamic group user sensing, cross-domain user sensing, and multi-task sensing. Additionally, this article proposes a series of current solutions, including improving the signal-to-noise ratio (SNR) of NLOS sensing from the perspective of communication signals and sensing models, addressing coexistence sensing issues, designing adaptable tasks for dynamic user groups, leveraging techniques like adversarial learning and transfer learning for cross-domain problems, and employing modular deep learning architectures. By providing a comprehensive overview of WiFi-based human identification, this survey not only offers insights into current research but also charts a roadmap for future investigations. It is anticipated that this survey will stimulate innovative research endeavors and foster the expansion of wireless sensing technology across diverse application domains.

Transitioning from MLOps to LLMOps: Navigating the Unique Challenges of Large Language Models

Large Language Models (LLMs), such as the GPT series, LLaMA, and BERT, possess incredible capabilities in human-like text generation and understanding across diverse domains, which have revolutionized artificial intelligence applications. However, their operational complexity necessitates a specialized framework known as LLMOps (Large Language Model Operations), which refers to the practices and tools used to manage lifecycle processes, including model fine-tuning, deployment, and LLMs monitoring. LLMOps is a subcategory of the broader concept of MLOps (Machine Learning Operations), which is the practice of automating and managing the lifecycle of ML models. LLM landscapes are currently composed of platforms (e.g., Vertex AI) to manage end-to-end deployment solutions and frameworks (e.g., LangChain) to customize LLMs integration and application development. This paper attempts to understand the key differences between LLMOps and MLOps, highlighting their unique challenges, infrastructure requirements, and methodologies. The paper explores the distinction between traditional ML workflows and those required for LLMs to emphasize security concerns, scalability, and ethical considerations. Fundamental platforms, tools, and emerging trends in LLMOps are evaluated to offer actionable information for practitioners. Finally, the paper presents future potential trends for LLMOps by focusing on its critical role in optimizing LLMs for production use in fields such as healthcare, finance, and cybersecurity.