Acute kidney injury (AKI) is a devastating complication of acute illness that affects adults and children across multiple settings worldwide and is associated with the development and progression of chronic kidney disease, increased mortality and increased resource utilization. Over the past two decades, standardization of criteria for AKI diagnosis and staging and the publication of multicentre studies have led to improved understanding of the AKI spectrum and provided insights into the heterogeneity of patient characteristics, processes of care and the environmental and sociodemographic factors that influence care delivery and outcomes. Substantial advances have been made in the utilization of electronic health records, biomarkers and care bundles- structured sets of evidence-based treatment practices- to improve the clinical management of AKI. The emerging fields of artificial intelligence and digital health may also provide ways to reduce the burden of this disease. However, these developments have occurred mainly in high-income countries and have yet to improve care delivery or outcomes in low-resource regions. Progress in the development of specific treatments for AKI is limited, and important gaps in knowledge and clinical practice remain, particularly in relation to the 5R framework (risk, recognition, response, renal support and rehabilitation) for managing AKI. An urgent need exists to address the wide variation and inequities in AKI management worldwide.

Recent advancements in the field artificial intelligence (AI), particularly in the architecture of convolutional neural network (CNN) architecture, have revolutionized medical imaging by enabling accurate image recognition. However, the application of AI in identifying degenerative musculoskeletal disorders, specifically on plain radiographs, is still poorly explored. The aim of this study is to classify cuff tear arthropathy (CTA) and glenoid erosion using AI on plain shoulder radiographs, using the Hamada and Favard classification systems. We used a publicly available CNN trained for image recognition and trained it using a diverse dataset of 6733 shoulder and clavicle X-ray images covering various clinical conditions. The performance of the network was evaluated in detail on a validation set of 561 images. Metrics such as sensitivity, specificity, Youden's index, and Area Under Curve (AUC) in the receiver operating characteristics curve analysis, were used for evaluation. AUC was the primary measure of accuracy. The network showed exceptional performance in identifying Hamada grades 3 and 4, achieving AUCs of 0.95, 95% CI [0.91-0.98] for both categories. While performance was slightly lower for Hamada grades 0-2 and glenoid erosion, with AUCs ranging from 0.81 to 0.91, it still demonstrated considerable accuracy. Similar results were obtained for Favard although we could not validate the network for more advanced stages due to lack of data. Our study demonstrates the network's robust capability to identify CTA on plain radiographs, comparable to earlier studies focused on osteoarthritis. Notably, the network excelled in later disease stages characterized by pronounced pathology. The ability to achieve such performance with a heterogeneous dataset bodes well for the real-world implementation of AI technology. However, it is crucial to acknowledge the potential influence of using a validation set instead of a dedicated test set, warranting further investigation.

The study explores and highlights the direct relationship between contemporary knowledge, paradigms, aims, and public expectations in the field of artificial intelligence (AI) and its definitions. For the purpose of this research, the division of the stages of the development of AI was used for analogies to the seasons; spring to winter of AI. The history of AI is covered here only in the range necessary to point out this relationship with the possibilities of derivingthe resulting definitions. Examples of period-typical AI definitions are given for each period. This historical excursion is then used as a background for thinking about the form (not the content) of the definition of AI that is appropriate to the current state of the field and its paradigm, focus, and use. The current discussion on the shape of the definition of AI within the framework of EUlegislation is outlined. The form of a suitable definition of AI for the present is examined from the perspective of interested parties, such as multinational entities, business organisations, and other stakeholders, and is compared with some already valid definitions of these entities. A paradox in the definitions of AI, which are always too “narrow and broad at the same time” from a certainpoint of view, is pointed out. Finally, the possibility of deploying a fractal defi nition with a fixed rational-moral core but changing content with respect to the levels at which it is applied is explored within a conceptual ideation. This operational fractal definition could, in principle, resolve the ever-present “broadness-narrowness” paradox.

Significant advances in neuroscience have improved the ability of physicians to diagnose and manage neurological and psychiatric disorders in patients. The use of neuroscience evidence in criminal trials in developed countries has increased significantly in the last two decades. This rapid increase has raised questions among the legal and scientific communities about the effects that these technologies can have on judicial decision-makers. The role of neuroscience in criminal liability is a topic that has been discussed in recent years. The purpose of this article is to review the use of neuroscience evidence in the criminal justice system, as well as current research examining the effects of neuroscience evidence on judicial decision-makers in criminal cases. This review is warranted given legal and scientific concerns about the impact of potential bias. The present study was conducted and analyzed using a documentary method and with reference to research published in the last four years. Some argue that neuroscience is irrelevant in the criminal court, while others believe that it can help prove the lack of control of behavior by many criminals. However, the truth is likely somewhere in between, as certain types of neuroscience evidence may be useful and relevant in criminal trials. This article describes recent advances in neuroscience in the fields of functional neuroimaging and artificial intelligence "deep learning" algorithms, and examines the legal and ethical challenges and potential benefits and drawbacks.

The artificial intelligence (AI) field holds significant promise to revolutionize healthcare, including transfusion medicine (TM). This study explored AI use in TM, education and research among International Society of Blood Transfusion (ISBT) members. A mixed methodology was employed. A survey was conducted June to November 2024. Eighteen participants were interviewed. A total of 218 ISBT members from 67 countries responded to the survey, 43.5% of which use AI. Most users (91.1%) have used Generative AI (GenAI); 82.3% indicated they were self-taught. Application to clinical TM was reported by 54.4%, and 87.3% reported a positive impact. A third of respondents (32.7%) indicated the use of AI in their institutions, commonly GenAI tools. More than two-thirds indicated use in TM education, research or both, and 71.1% indicated a positive impact on their institution's operations. Use in education included preparing lectures and generating questions. Use in research included brainstorming ideas, statistical analysis, coding, data interpretation, manuscript drafting and proofing. Survey respondents reported various challenges in adopting AI, including lack of access to AI resources or expertise (78%), cost (74%), difficulty in hiring AI professionals (73%) and data privacy concerns (72%). Concerns raised during interviews included accuracy of information, regulatory constraints and risks on intellectual ability and employment. There is general interest in the use of AI in TM practice, education and research. Barriers to adoption include access to the technology and lack of AI professionals. Educational resources must be expanded. Regulatory constraints and privacy and trust concerns need to be addressed.

Multiple sequence alignments (MSAs) have been traditionally used for making inferences about site-specific diversity in proteins. Recent advancements in the field of artificial intelligence have highlighted the potential of protein language models (pLMs) to capture similar protein properties. Unlike MSAs, pLMs can make inferences from single sequences, without the need for a set of aligned sequences. In this study, we introduce a variation of the Context-Dependent Entropy metric, based on pLM embeddings instead of MSA input, to assess protein site conservation and variability. We test this metric using versions of two popular pLMs (ESM-2 and protT5) fine-tuned on the diversity of different Influenza A virus subtype hemagglutinin proteins. Our study demonstrates how our pLM entropy metric can capture which sites are more likely to change in a specific sequence context and how fine-tuning pLMs on a set of evolutionarily related proteins (evotuning) can improve the models’ understanding of the group’s diversity.

Artificial intelligence (AI) has, in the recent past, experienced a rebirth with the growth of generative AI systems such as ChatGPT and Bard. These systems are trained with billions of parameters and have enabled widespread accessibility and understanding of AI among different user groups. Widespread adoption of AI has led to the need for understanding how machine learning (ML) models operate to build trust in them. An understanding of how these models generate their results remains a huge challenge that explainable AI seeks to solve. Federated learning (FL) grew out of the need to have privacy-preserving AI by having ML models that are decentralized but still share model parameters with a global model. This study sought to examine the extent of development of the explainable AI field within the FL environment in relation to the main contributions made, the types of FL, the sectors it is applied to, the models used, the methods applied by each study, and the databases from which sources are obtained. A systematic search in 8 electronic databases, namely, Web of Science Core Collection, Scopus, PubMed, ACM Digital Library, IEEE Xplore, Mendeley, BASE, and Google Scholar, was undertaken. A review of 26 studies revealed that research on explainable FL is steadily growing despite being concentrated in Europe and Asia. The key determinants of FL use were data privacy and limited training data. Horizontal FL remains the preferred approach for federated ML, whereas post hoc explainability techniques were preferred. There is potential for development of novel approaches and improvement of existing approaches in the explainable FL field, especially for critical areas. OSF Registries 10.17605/OSF.IO/Y85WA; https://osf.io/y85wa.

According to the Church–Turing thesis, the limit of what is computable is bounded by Turing machines. Following from this, given that general computable functions formally describe the notion of recursive mechanisms, it is sometimes argued that every organismic process that specifies consistent cognitive responses should be both limited to Turing machine capabilities and amenable to formalization. There is, however, a deep intuitive conviction permeating contemporary cognitive science, according to which mental phenomena, such as consciousness and agency, cannot be explained by resorting to this kind of framework. In spite of some exceptions, the overall tacit assumption is that whatever the mind is, it exceeds the reach of what is described by notions of computability. This issue, namely the nature of the relation between cognition and computation, becomes particularly pertinent and increasingly more relevant as a possible source of better understanding the inner workings of the mind, as well as the limits of artificial implementations thereof. Moreover, although it is often overlooked or omitted so as to simplify our models, it will probably define, or so we argue, the direction of future research on artificial life, cognitive science, artificial intelligence, and related fields.

Mangosteen is one of the leading export commodities from Indonesia. Despite its great economic potential, only about 25% of Indonesian mangosteens meet export standards, mainly due to visual defects such as yellow sap and spots on the skin of the fruit. The process of sorting export worthy mangosteens has been done manually, which tends to be time consuming and inconsistent. Therefore, this study aims to utilize artificial intelligence technology in building a real-time image recognition model to improve the efficiency and accuracy of the export-quality mangosteen sorting process. This study uses you only look once version 8 (YOLOv8) as an image recognition model with YOLOv8 variants, including nano, small, medium, large, and extra large variants. The results of the study using 4,014 primary and 255 secondary data of mangosteen, the highest performance is reached by YOLOv8 medium 82% of accuracy, 0.856 of mean average precision (mAP)50, and 0.616 of mAP50-95. This result is obtained from 70% training, 20% validation, and 10% testing data with epoch stop 85. These results indicate that the model can provide good performance in mangosteen export quality classification. This research contributes to the fields of agricultural technology and artificial intelligence by offering an innovative solution to a practical problem, enhancing efficiency, accuracy, and scalability in export-quality mangosteen sorting.

Artificial intelligence offers students more personalised and adaptive learning, which encourages educators to better understand students' learning processes. This study aims to determine the readiness levels of medical faculty students in medical artificial intelligence and to examine whether these levels vary based on gender and class year. The study was conducted with 322 medical students. Research data were collected using the "Medical Artificial Intelligence Readiness Scale for Medical Students." Results showed that medical students rated themselves as moderate in the "cognition" and "vision" dimensions, slightly higher in the "ability" and "ethics" dimensions, and overall at a "neutral" level in medical artificial intelligence readiness. Compared to females, males showed significant differences at a "small effect" level in cognition, ability factors and overall scores. Regarding class levels, significant differences were found between 2nd graders and both 5th and 6th graders, favouring the 2nd graders at an "intermediate effect" level. In the cognition dimension, there was also a significant difference between the 2nd and 4th grades in favour of the 2nd grade and at the level of "small effect." In order to increase medical artificial intelligence readiness of students, it is important to comprehensively include the subject in the medical school curriculum and to develop it according to needs. In future research, long-term follow-up studies aimed at improving medical students' education in the field of medical artificial intelligence (AI) are considered to be very beneficial. Furthermore, future studies should also consider potential changes in medical AI readiness that may occur over time.

Ethics and Policy Challenges in Applying Artificial Intelligence in Medicine.

Foundation models (FMs) have become a paradigm shift in the field of artificial intelligence, allowing one large-scale pretrained model to be customized for a broad set of downstream tasks using very little task-specific data. These models, which include GPT, CLIP, BERT, and vision transformers, have altered the scope of transfer learning and multimodal understanding and are built on top of enormous datasets and self-supervised learning. The paper provides a broad view of the modern state of foundation models, with an emphasis on their technological foundation, training, and cross-domain use in fields like natural language processing, computer vision, healthcare, robotics and scientific discovery. We also explore the main opportunities that FMs offer, as well as state-of-the-art methods and techniques for the development of foundation models. we discuss their applications in natural language processing, computer vision, healthcare, etc. Furthermore, their limitations and challenges are also investigated. Lastly, future prospects are discussed so that professionals and scientists obtain a better understanding of the importance of foundation models for addressing their research goals.

Rapid developments in the Artificial Intelligence (AI) industry have triggered an increased need for workers with specialized competencies, which has implications for significant variations in salary levels. This research aims to analyze the factors that influence salaries in the AI sector using the multiple linear regression method. The dataset used includes 15,000 AI job vacancies with variables including job and company characteristics. The data was engineered via the one-hot encoding method and divided into two parts: training data (80%) and test data (20%). The analysis results show that the regression model is able to explain 85% of the variation in salary, with an R² value of 0.85 and a Root Mean Square Error (RMSE) of USD 23,221. The three main factors identified as having a significant influence on salaries in the AI field are work experience, company location, and the industry in which the company operates. The experience factor reflects the skills and knowledge developed over many years, which can increase productivity (Rony et al., 2023). Company location also plays an important role, as the cost of living and demand for skilled labor varies by region (Badran, 2019). Additionally, the specific industry in which an employee works influences salary, given that more developed industries can often offer higher compensation (Huang, 2025). This research makes a significant empirical contribution to the understanding of compensation structures in the AI labor market.

This special issue presents papers with a shared focus on the exploration or use of concepts relating to embodiment within the field of computing education. We first summarize many of the related theoretical frameworks that have been based on embodied perception and action within the fields of psychology and artificial intelligence. We then introduce the 14 papers of the special issue. These cover a wide range of topics, many involving young learners, and they address both typical topics in introductory computing and the emerging field of AI education.

The connection between the field of computer and artificial intelligence has emerged as a potential new tool for enhancing students’ writing performance at the University of Tabuk. The goal of this research was to investigate the use of artificial intelligence to augment scientific and technical writing among computer science students at the University of Tabuk. Thirty-five students aged nineteen to twenty years old participated in this study. This research employed quantitative research design. A survey questionnaire with 11 questions was designed to assess the university students’ perceptions of artificial intelligence use in learning. The study illustrated that students with a positive attitude toward artificial intelligence performed better, improved their language skills, and enhanced their learning. The study also established that artificial intelligence use among students increases academic writing. This research contributes significantly to understanding of how artificial intelligence impacts the scientific and technical writing proficiency of computer science students. Although artificial intelligence shows great potential, it should be seen as an additional tool that enhances rather than replaces the crucial function of human educators. In results indicate that, chatbot was demonstrated to significantly improve the writing skills of computer science students at the University of Tabuk through the use of artificial intelligence, including chatbots.

Abstract A rapid rise in machine learning-based applications has made it one of the most popular areas in the field of artificial intelligence (AI). The most commonly used libraries to implement the algorithms used in these applications are Scikit learn and Weka. It is challenging to test these machines learning based applications due to the Oracle Problem. The problem is when the expected outcome is not known and hence the testing of such applications cannot be performed via traditional testing techniques. One of the solution to the Oracle problem is the use of Metamorphic testing to test the machine learning applications. The code of machine learning algorithms is often ignored, when testing of ML-based applications is done. However, the usage of the machine learning algorithms within the libraries requires formal testing to improve reliability. This work evaluates the Metamorphic relations for machine learning algorithms by finding their kill rate while testing 5 machine learning (ANN, ID3, KNN, Naive Bayes, SVM) classifiers from the Scikit Learn library. This work also calculates the statement coverage, while testing the metamorphic relations. The relationship between the effectiveness of fault detection and code coverage is identified as well.

The study aims to explore the intersection of Operations Research (OR), Artificial Intelligence (AI), and simulation by examining how classical OR methodologies strengthen AI models, particularly in machine learning, robotics, natural language processing, and autonomous systems. It further investigates the critical role of simulation in training, testing, and validating AI algorithms, emphasizing its relevance for optimization, intelligent decision-making, and real-world system modelling. The study adopts a comprehensive analytical and literature-based methodology, reviewing foundational OR techniques, simulation principles, and modern AI applications. It synthesizes interdisciplinary research across mathematics, computer science, and engineering, supported by case analyses in robotics, healthcare, transportation, and autonomous systems. Additionally, a demonstration algorithm is developed to simulate automatic gear-control behavior in vehicles, illustrating how simulation models practically support AI-oriented operational decision-making. Findings reveal that OR optimization techniques significantly enhance AI efficiency, particularly in parameter tuning, resource allocation, and adaptive decision-making. Simulation is shown to be indispensable for AI training, offering controlled, safe, scalable, and cost-effective environments. The study identifies persistent challenges—including the reality gap, computational demands, and model bias—yet confirms that simulation and OR jointly accelerate AI development and broaden its practical reliability. The integrated OR–AI–simulation framework is applicable to numerous fields, including autonomous vehicle navigation, robotic motion planning, intelligent healthcare systems, logistics optimization, and smart city management. Industries benefit from improved forecasting, reduced operational costs, enhanced safety, and high-fidelity algorithm testing. Simulated environments also support reinforcement learning, surgical training, autonomous decision-making, and large-scale scenario evaluation, contributing to more efficient and intelligent real-world systems. The study’s novelty lies in its unified perspective that connects classical OR optimization principles with AI advancements through simulation-based experimentation. It uniquely synthesizes concepts from mathematics, computer science, and AI to highlight simulation as a bridge enabling intelligent automation. The inclusion of a practical simulation algorithm for automatic vehicle gear control further demonstrates how OR-driven simulation can concretely operationalize AI-based decision systems. Operations Research (OR) and Artificial Intelligence (AI) have both independently evolved as transformative fields which have shown impact on decision-making and problem-solving across diverse domains. While Operations Research makes available a foundation of mathematical modeling and optimization techniques, Artificial Intelligence sets up intelligence through learning, reasoning, and data-driven methods. This seminar paper presented by us explores how Operations Research gets involved in the development and enhancement of Artificial Intelligence (AI) systems. We, in this paper, have tried to discuss Key applications and case studies to highlight the synergies between these fields, remarkably in optimization, logistics, resource allocation, and automated decision-making. Simulation, an approach of Operations Research, has become a cornerstone in the field of Artificial Intelligence (AI), suggesting an experimental platform for testing hypotheses, training algorithms, and evaluating systems in controlled, cost-effective, and scalable environments. Our paper explores the central role simulations play in advancing Artificial Intelligence (AI) research and applications, with a focus on their integration in machine learning, robotics, and decision-making systems.

ObjectivesThis study aimed to systematically characterize the landscape of artificial intelligence (AI) applications in gynecologic cancers, offering a comprehensive overview of current research trends, influential publications, key contributors, and future research directions. The focus of this study was to provide a quantitative overview of the field's development and trends.Materials and MethodsA structured search was performed in the Web of Science Core Collection to identify original articles on AI use in gynecologic oncology. Two independent reviewers screened and selected studies based on predefined inclusion criteria. Extracted data—including publication trends, author and institutional collaborations, keyword co-occurrence, and citation networks—were analyzed using CiteSpace 6.2.R6 and VOSviewer software.ResultsA total of 2544 articles were included for analysis. Research activity showed a notable acceleration after 2019, reaching its highest output in 2024. China and the United States emerged as dominant contributors, with the Chinese Academy of Sciences and Fudan University leading among institutions. Influential authors such as Sala Evis, Tian Jie, and Scambia Giovanni were identified. Major research themes focused on “Radiomics,” “Deep Learning,” “Radiotherapy,” and cancers including cervical, ovarian, and endometrial. Recent emerging topics included “Digital Pathology,” “Personalized Medicine,” and “Tumor Heterogeneity,” signaling a shift toward precision oncology.ConclusionsThis bibliometric study delineated the evolving field of AI in gynecologic oncology, highlighting dynamic research fronts and gaps.

The Human Exposome Project (HEP) aims to chart lifelong environmental exposures and their biological consequences, furnishing the environmental counterpart to the genomic revolution. Yet the fine‑grained, multimodal data streams that fuel exposomics-biospecimens, geolocation traces, wearable‑sensor feeds, and socio‑environmental metadata-raise privacy, justice, and governance questions that may exceed the reach of conventional bioethics. Building on lessons from genomics, biobanking, digital health, and environmental‑justice research, we identify five foundational ethical domains for exposome science: (1) privacy and data sovereignty, (2) informed consent and sustained participant engagement, (3) environmental justice, (4) governance and oversight, and (5) actionability and the responsible return of results,as well as (6)the adherence to research program goals. Similar to the "values in design" construct widely used in the socio-technical field and the "ethics by design" in the artificial intelligence (AI) field, we translate these domains into operational pillars for ethics‑by‑design research practice: dynamic or tiered consent architectures; participatory governance mechanisms such as community advisory boards; embedded ethics research programs; algorithmic‑fairness protocols for artificial‑intelligence analytics; and dedicated review bodies equipped to evaluate longitudinal, sensor‑based, multi‑omics studies. Concrete recommendations include federated data stewardship to minimize re‑identification risk, Evidence‑to‑Decision frameworks that couple exposomic evidence with societal values, and transparent pathways for communicating context‑dependent findings to individuals, communities, and policymakers. Ethical preparedness and action are a prerequisite for the scientific impact and social license of exposome research. Institutionalizing the proposed roadmap-via an international Exposome Ethics Consortium, expanded training for Institutional Review Boards, harmonized regulatory guidance, and sustained community co‑governance-will help protect privacy, promote equity, and foster public trust. Embedding systematic ethical reflection as core infrastructure will enable the Human Exposome Project to realize its promise of precision public health without replicating patterns of opaque surveillance, marginalization, or data commodification. The Human Exposome Project (HEP) represents an ambitious endeavor to characterize lifelong environmental exposures in relation to health. Yet, this vision brings profound ethical challenges: from managing massive, sensitive datasets to ensuring justice for disproportionately exposed communities. This article synthesizes foundational work on exposome ethics, outlines core ethical challenges, and proposes a proactive ethical governance model that ensures scientific integrity and social legitimacy.

Several recent advancements have been made in the field of artificial intelligence (AI) language learning. Given the widespread adoption and enabling power of immersive technologies, as well as the potential applications of Artificial Intelligence Supported Language Learning (AISLL), it is critical to continuously investigate the literature to identify trends and practices in language education research. Of the 89 publications located between 2021 and 2023, 10 were selected based on the criteria for inclusion and exclusion from WoS and Scopus. Using five codes obtained from earlier systematic reviews, the researcher conducted an analysis and synthesis of these studies. The codes were as follows: 1) aim, 2) methodology, 3) sample, 4) country, and 5) outcomes. The systematic review revealed several key trends in AISLL. It was found that universities were the predominant setting for AISLL research, with most studies employing quantitative research methods. The methodologies varied widely, with emphasis on experimental and quasi-experimental designs. The countries represented in the studies were diverse, yet there was a concentration in technologically advanced regions. Significant outcomes reported include improved student performance and positive attitudes toward AI tools in language learning. To better understand AI utilization in language teaching and learning, academics are urged to broaden the scope of future studies and involve students at all educational levels in future AISLL practices.