Hasil untuk "artificial intelligence"

LIU Donglin, ZHOU Xia, DAI Jianfeng, XIE Xiangpeng, TANG Yi, LI Juanshi

Integrated energy systems in buildings are an effective means to achieve low-carbon buildings. To further tap into their demand-side flexibility adjustable potential and carbon reduction potential, and reasonably allocate the interests of various entities in the building integrated energy system, a bi-level optimization scheduling strategy for building integrated energy system considering virtual energy storage in buildings under Stackelberg game framework is proposed. First, the thermal inertia of the cooling and heating system inside the building and the flexibility of the cooling and heating load are considered to leverage the virtual energy storage function of the building and improve system flexibility in the game model. Then, the genetic algorithm is used to solve the upper-level pricing model of energy operators, updating the purchase and sale electricity prices set by upper-level leaders, while the CPLEX solver is used to solve the lower-level problem, optimizing equipment output, demand response, and electricity trading plans. Finally, the proposed model is verified by case studies that it can effectively improve the economic performance and low-carbon characteristics of building integrated energy systems.

Sheryl L. Rifas-Shiman, Li Yi, Izzuddin M. Aris et al.

Plain English summary Many women experience suboptimal cardiovascular health during midlife, increasing their risk of heart disease. Research suggests that exposure to greenspace, such as vegetation and trees, may benefit heart health, but most studies have relied on satellite images rather than street-level views. In this study, we used Google Street View images and artificial intelligence to measure visible greenspace around the homes of 767 women from the Project Viva study in Eastern Massachusetts. We assessed greenspace exposure from trees, grass, and other plants within 500 m of their homes between 2012 and 2016 and examined its association with cardiovascular health scores about five years later, accounting for individual and neighborhood socioeconomic status. We evaluated cardiovascular health scores using the Life’s Essential 8 (LE8) construct, which includes behavioral factors (diet, physical activity, sleep, and smoking) and biomedical factors (body mass index, blood pressure, blood lipids, and blood glucose). Our results showed that women living in areas with more total greenspace—particularly more trees—had higher overall cardiovascular health scores. Living in an area with higher amounts of visible trees was associated with a healthier diet, better sleep, higher physical activity levels, smoking avoidance, and healthier body weight, blood pressure, and blood sugar levels. These findings suggest that increasing trees in neighborhoods could support better cardiovascular health among women in midlife. Policies promoting urban tree planting may be a valuable public health strategy to improve heart health and overall well-being.

Joshua P Schiff, Beatriz Guevara, Amir Ahari et al.

Abstract Background Adaptive radiotherapy (ART) is an advanced form of image-guided radiotherapy that involves the re-contouring and re-planning of a patient’s treatment plan, either while the patient is on the table (online) or in between fractions (offline). ART allows for the adjustment of a treatment plan to respect a patient’s changes in internal anatomy, something that is critical in the treatment of gastrointestinal (GI) malignancies in which the mobile and radiosensitive GI tract plays a key role in driving toxicity. Herein we review the indications for both online and offline ART in GI cancers. Main text Online ART plays a critical role in the treatment of pancreatic cancer when using stereotactic body radiotherapy (SBRT). A variety of ART workflows have demonstrated that ART allows for the safe dose-escalated treatment of locally advanced pancreatic cancer. In addition to pancreatic cancer, there are now a bevy of data demonstrating that ART plays a key role in the treatment of liver cancers and abdominal oligometastases when using SBRT and allows for the safe delivery of single-fraction abdominal SBRT. While lower GI cancers are generally not treated with SBRT-like doses, both online and offline ART workflows have been shown to potentially reduce toxicity in patients with anal and rectal cancers. Improved integration of artificial intelligence and direct-to-unit workflows in ART hold promise that the overall process can become more efficient, allowing for more widespread adoption in GI radiation oncology. Conclusions ART is an expanding radiotherapy paradigm in which a patient’s treatment plan is adjusted to match observed changes in patient anatomy and has been successfully incorporated into the treatment of a variety of GI cancers. The successful implementation of workflows in pancreatic cancer, liver cancers, and lower GI cancers, amongst others, as well as incorporation into multi-center clinical trials, suggest that ART will continue to play a critical role of GI radiation oncology for years to come. As improvements in efficiency and access allow for increasing use of ART world-wide, we predict that ART will continue to play a critical part in the management of patients with GI malignancies.

Oscar Girón-Vallejo, Bernardo Garcia-Nuñez, Isidoro Narbona-Arias et al.

Three-dimensional (3D) modeling and printing technologies are increasingly used in pediatric surgery, offering improved anatomical visualization, surgical planning, and personalized approaches to complex conditions. Compared to standard imaging, patient-specific 3D models—virtual or printed—provide a more intuitive spatial understanding of congenital anomalies, tumors, and vascular anomalies. This review compiles evidence from pediatric surgical fields including oncology, abdominal, and thoracic surgery, highlighting the clinical relevance of 3D applications. The technological workflow—from image segmentation to computer-aided design (CAD) modeling and multimaterial printing—is described, emphasizing accuracy, reproducibility, and integration into hospital systems. Several clinical cases are presented: neuroblastoma, cloacal malformation, conjoined twins, and two cases of congenital diaphragmatic hernia (one with congenital pulmonary airway malformation, CPAM). In each, 3D modeling enhanced anatomical clarity, increased surgeon confidence, and supported safer intraoperative decision-making. Models also improved communication with families and enabled effective multidisciplinary planning. Despite these advantages, challenges remain, such as production time, cost variability, and lack of standardization. Future directions include artificial intelligence-based automation, expanded use of virtual and mixed reality, and prospective validation studies in pediatric cohorts. Overall, 3D modeling represents a significant advance in pediatric precision surgery, with growing evidence supporting its safety, clinical utility, and educational value.

Haoran Zhu, Yu Ding

Nanobodies, derived from naturally occurring heavy-chain antibodies in camelids (VHHs) and sharks (V_NARs), are unique single-domain antibodies that have garnered significant attention in therapeutic, diagnostic, and biotechnological applications due to their small size, stability, and high specificity. This review first traces the historical discovery of nanobodies, highlighting key milestones in their isolation, characterization, and therapeutic development. We then explore their structure–function relationship, emphasizing features like their single-domain architecture and long CDR3 loop that contribute to their binding versatility. Additionally, we examine the growing interest in multiepitope nanobodies, in which binding to different epitopes on the same antigen not only enhances neutralization and specificity but also allows these nanobodies to be used as controllable modules for precise antigen manipulation. This review also discusses the integration of AI in nanobody design and optimization, showcasing how machine learning and deep learning approaches are revolutionizing rational design, humanization, and affinity maturation processes. With continued advancements in structural biology and computational design, nanobodies are poised to play an increasingly vital role in addressing both existing and emerging biomedical challenges.

Vinit Mehta, Charu Sharma, Karthick Thiyagarajan

With the rapid advancement of artificial intelligence and robotics, the integration of Large Language Models (LLMs) with 3D vision is emerging as a transformative approach to enhancing robotic sensing technologies. This convergence enables machines to perceive, reason, and interact with complex environments through natural language and spatial understanding, bridging the gap between linguistic intelligence and spatial perception. This review provides a comprehensive analysis of state-of-the-art methodologies, applications, and challenges at the intersection of LLMs and 3D vision, with a focus on next-generation robotic sensing technologies. We first introduce the foundational principles of LLMs and 3D data representations, followed by an in-depth examination of 3D sensing technologies critical for robotics. The review then explores key advancements in scene understanding, text-to-3D generation, object grounding, and embodied agents, highlighting cutting-edge techniques such as zero-shot 3D segmentation, dynamic scene synthesis, and language-guided manipulation. Furthermore, we discuss multimodal LLMs that integrate 3D data with touch, auditory, and thermal inputs, enhancing environmental comprehension and robotic decision-making. To support future research, we catalog benchmark datasets and evaluation metrics tailored for 3D-language and vision tasks. Finally, we identify key challenges and future research directions, including adaptive model architectures, enhanced cross-modal alignment, and real-time processing capabilities, which pave the way for more intelligent, context-aware, and autonomous robotic sensing systems.

Oleksandr Tymoshchuk

The expansion of nuclear technologies in various industries, combined with the constant threat of radiation-related incidents, highlights the urgent need for effective radiation education. This study is devoted to an empirical investigation of the effectiveness of artificial intelligence tools (neurological models of artificial intelligence) in detecting and correcting of artificial intelligence tools (neurological models of artificial intelligence) in detecting and correcting misconceptions about radiation (ionising radiation). We empirically evaluate the effectiveness of artificial intelligence (AI) tools in detecting and correcting these misconceptions among university students, focusing on different cognitive, cognitive-activity, and systemic-axiological levels. A pedagogical experiment was conducted with 168 students of Ukrainian universities using control questionnaires to assess the effectiveness of the selected artificial intelligence tools. The experiment involved presenting students with a series of statements designed to identify misconceptions related to factual knowledge (e.g., radiation units, background levels), conceptual understanding (e.g., the difference between radiation and radioactivity, effects of low-dose exposure), and application/evaluation (e.g., risk assessment, protective measures). AI tools, including natural language processing models for text analysis and machine learning algorithms for misconceptions classification, were used to provide personalised feedback and targeted corrective information. The results show that AI achieved high accuracy (80-98%) in eliminating misconceptions about factual knowledge. However, the effectiveness decreased for misconceptions requiring deeper conceptual understanding (73-78%) and is much lower for those involving complex knowledge assessment and application (24-36%). These findings indicate that while AI has significant potential to improve basic radiation literacy and provide automated feedback, its current capabilities are limited in addressing more multidimensional and complex misconceptions. Further research is needed to develop more sophisticated AI-based integrations that can effectively target higher-order cognitive skills and promote a more complete understanding of radiation science and its implications. This study contributes to this field by providing empirical evidence on the strengths and weaknesses of AI in radiation education, and offers practical recommendations for the further development and implementation of customised AI-based learning tools.

Sonal Allana, Rozita Dara, Xiaodong Lin et al.

Explainable Artificial Intelligence (XAI) is a crucial pathway in mitigating the risk of non-transparency in the decision-making process of black-box Artificial Intelligence (AI) systems. However, despite the benefits, XAI methods are found to leak the privacy of individuals whose data is used in training or querying the models. Researchers have demonstrated privacy attacks that exploit explanations to infer sensitive personal information of individuals. Currently there is a lack of defenses against known privacy attacks targeting explanations when vulnerable XAI are used in production and machine learning as a service system. To address this gap, in this article, we explore Privacy Enhancing Technologies (PETs) as a defense mechanism against attribute inference on explanations provided by feature-based XAI methods. We empirically evaluate 3 types of PETs, namely synthetic training data, differentially private training and noise addition, on two categories of feature-based XAI. Our evaluation determines different responses from the mitigation methods and side-effects of PETs on other system properties such as utility and performance. In the best case, PETs integration in explanations reduced the risk of the attack by 49.47%, while maintaining model utility and explanation quality. Through our evaluation, we identify strategies for using PETs in XAI for maximizing benefits and minimizing the success of this privacy attack on sensitive personal information.

Niloofar Shadab, Tyler Cody, Alejandro Salado et al.

Engineering methodologies predominantly revolve around established principles of decomposition and recomposition. These principles involve partitioning inputs and outputs at the component level, ensuring that the properties of individual components are preserved upon composition. However, this view does not transfer well to intelligent systems, particularly when addressing the scaling of intelligence as a system property. Our prior research contends that the engineering of general intelligence necessitates a fresh set of overarching systems principles. As a result, we introduced the "core and periphery" principles, a novel conceptual framework rooted in abstract systems theory and the Law of Requisite Variety. In this paper, we assert that these abstract concepts hold practical significance. Through empirical evidence, we illustrate their applicability to both biological and artificial intelligence systems, bridging abstract theory with real-world implementations. Then, we expand on our previous theoretical framework by mathematically defining core-dominant vs periphery-dominant systems.

Njabulo Mziwandile Zulu, Hamed Hashemi, Kaniki Tumba

Gas hydrate inhibition using chemicals has been under continuous investigation, and several modelling studies have been published since its inception. Since it is not always feasible to conduct experimental research, it is especially crucial to forecast the conditions under which gas hydrates may form and dissociate in the presence of chemical inhibitors. As a result, a reliable forecasting tool is vital. This article provides an exhaustive review of various modelling methodologies in the context of gas hydrate chemical inhibition. The key aspects of empirical models, thermodynamic models, kinetic models, artificial intelligence-based models and quantum chemistry-based models are presented. Critical analysis of each modelling approach has been performed, highlighting strengths, limitations, and areas where further investigations are still crucial. Rapid progress has been made with respect to gas hydrate modelling approaches in the context of chemical inhibition; however, further research is still vital to bridge the gaps that have been identified in this review. Potential improvements to existing models have been proposed, particularly in terms of integrating experimental data and utilizing hybrid approaches, which could serve as valuable future directions for the field.

P. Sasikala, R. Ravichandran

This study explores the transformative potential of Artificial Intelligence (AI) in education by analyzing its impact on student learning outcomes. Through a comprehensive literature review, the research synthesizes current findings on the integration of AI in educational settings, examining both the benefits and challenges it presents. The study explores into AI's role in personalizing learning experiences, enhancing student engagement, and improving academic performance. Ethical considerations such as data privacy and algorithmic bias are also assessed. This research also identifies existing gaps in the literature and suggests avenues for future inquiry, contributing to a deeper understanding of how AI can be effectively and responsibly integrated into education to optimize student success.

Valerio Nardone, Federica Marmorino, Marco Maria Germani et al.

The integration of multidisciplinary tumor boards (MTBs) is fundamental in delivering state-of-the-art cancer treatment, facilitating collaborative diagnosis and management by a diverse team of specialists. Despite the clear benefits in personalized patient care and improved outcomes, the increasing burden on MTBs due to rising cancer incidence and financial constraints necessitates innovative solutions. The advent of artificial intelligence (AI) in the medical field offers a promising avenue to support clinical decision-making. This review explores the perspectives of clinicians dedicated to the care of cancer patients—surgeons, medical oncologists, and radiation oncologists—on the application of AI within MTBs. Additionally, it examines the role of AI across various clinical specialties involved in cancer diagnosis and treatment. By analyzing both the potential and the challenges, this study underscores how AI can enhance multidisciplinary discussions and optimize treatment plans. The findings highlight the transformative role that AI may play in refining oncology care and sustaining the efficacy of MTBs amidst growing clinical demands.

Yue Xing, Wensheng Gan, Qidi Chen

The development history of landscape architecture (LA) reflects the human pursuit of environmental beautification and ecological balance. With the advancement of artificial intelligence (AI) technologies that simulate and extend human intelligence, immense opportunities have been provided for LA, offering scientific and technological support throughout the entire workflow. In this article, we comprehensively review the applications of AI technology in the field of LA. First, we introduce the many potential benefits that AI brings to the design, planning, and management aspects of LA. Secondly, we discuss how AI can assist the LA field in solving its current development problems, including urbanization, environmental degradation and ecological decline, irrational planning, insufficient management and maintenance, and lack of public participation. Furthermore, we summarize the key technologies and practical cases of applying AI in the LA domain, from design assistance to intelligent management, all of which provide innovative solutions for the planning, design, and maintenance of LA. Finally, we look ahead to the problems and opportunities in LA, emphasizing the need to combine human expertise and judgment for rational decision-making. This article provides both theoretical and practical guidance for LA designers, researchers, and technology developers. The successful integration of AI technology into LA holds great promise for enhancing the field's capabilities and achieving more sustainable, efficient, and user-friendly outcomes.

T. J. Hart, Chloe Engler Hart, Spencer Hopson et al.

Despite considerable progress in developing artificial intelligence (AI) algorithms for prostate cancer detection from whole slide images, the clinical applicability of these models remains limited due to variability in pathological annotations and existing dataset limitations. This article proposes a novel approach to overcome these challenges by leveraging a Bayesian framework to seamlessly integrate new data, and present results as a panel of annotations. The framework is demonstrated by integrating a Bayesian prior with one trained AI model to generate a distribution of Gleason patterns for each pixel of an image. It is shown that using this distribution of Gleason patterns rather than a ground-truth label can improve model applicability, mitigate errors, and highlight areas of interest for pathologists. Additionally, we present a high-quality, hand-curated dataset of prostate histopathological images annotated at the gland level by trained pre-medical students and verified by an expert pathologist. We highlight the potential of this adaptive and uncertainty-aware framework for developing clinically deployable AI tools that can support pathologists in accurate prostate cancer grading, improve diagnostic accuracy, and create positive patient outcomes.

Elsa Lawrence, Adham El-Shazly, Srijit Seal et al.

Modern life sciences research is increasingly relying on artificial intelligence approaches to model biological systems, primarily centered around the use of machine learning (ML) models. Although ML is undeniably useful for identifying patterns in large, complex data sets, its widespread application in biological sciences represents a significant deviation from traditional methods of scientific inquiry. As such, the interplay between these models and scientific understanding in biology is a topic with important implications for the future of scientific research, yet it is a subject that has received little attention. Here, we draw from an epistemological toolkit to contextualize recent applications of ML in biological sciences under modern philosophical theories of understanding, identifying general principles that can guide the design and application of ML systems to model biological phenomena and advance scientific knowledge. We propose that conceptions of scientific understanding as information compression, qualitative intelligibility, and dependency relation modelling provide a useful framework for interpreting ML-mediated understanding of biological systems. Through a detailed analysis of two key application areas of ML in modern biological research - protein structure prediction and single cell RNA-sequencing - we explore how these features have thus far enabled ML systems to advance scientific understanding of their target phenomena, how they may guide the development of future ML models, and the key obstacles that remain in preventing ML from achieving its potential as a tool for biological discovery. Consideration of the epistemological features of ML applications in biology will improve the prospects of these methods to solve important problems and advance scientific understanding of living systems.

Jeremy Straub, Zach Johnson

Computer system creativity is a key step on the pathway to artificial general intelligence (AGI). It is elusive, however, due to the fact that human creativity is not fully understood and, thus, it is difficult to develop this capability in software. Large language models (LLMs) provide a facsimile of creativity and the appearance of sentience, while not actually being either creative or sentient. While LLMs have created bona fide new content, in some cases - such as with harmful hallucinations - inadvertently, their deliberate creativity is seen by some to not match that of humans. In response to this challenge, this paper proposes a technique for enhancing LLM output creativity via an iterative process of concept injection and refinement. Initial work on the development of the Creative Artificial Intelligence through Recurring Developments and Determinations (CAIRDD) system is presented and the efficacy of key system components is evaluated.

Youzhi Qu, Chen Wei, Penghui Du et al.

During the evolution of large models, performance evaluation is necessarily performed to assess their capabilities and ensure safety before practical application. However, current model evaluations mainly rely on specific tasks and datasets, lacking a united framework for assessing the multidimensional intelligence of large models. In this perspective, we advocate for a comprehensive framework of cognitive science-inspired artificial general intelligence (AGI) tests, aimed at fulfilling the testing needs of large models with enhanced capabilities. The cognitive science-inspired AGI tests encompass the full spectrum of intelligence facets, including crystallized intelligence, fluid intelligence, social intelligence, and embodied intelligence. To assess the multidimensional intelligence of large models, the AGI tests consist of a battery of well-designed cognitive tests adopted from human intelligence tests, and then naturally encapsulates into an immersive virtual community. We propose increasing the complexity of AGI testing tasks commensurate with advancements in large models and emphasizing the necessity for the interpretation of test results to avoid false negatives and false positives. We believe that cognitive science-inspired AGI tests will effectively guide the targeted improvement of large models in specific dimensions of intelligence and accelerate the integration of large models into human society.

Oksana Buhaichuk, Vitalina Nikitenko, Valentyna Voronkova

The relevance of the study is that the digital challenge is important and stimulating, requiring the formation of digital education in the digital economy. The purpose of the article is to develop a model of digital education as a factor of improving the efficiency of digital competencies that contribute to the development of the digital economy. The object of research is the formation of a digital education model as a factor in the implementation of digital literacy. The subject of the study is the impact of the digital education model on the development of the digital economy. The methodology for researching digital education, which cultivates a smart economy, smart governance and smart people, is represented by the Agile methodology (flexible adaptive), based on the use of the values of artificial intelligence and deep learning, which can create effective tools for education, increasing their effectiveness through rapid change. The results of the study: 1) analyzes the formation of digital competencies in the context of the European educational paradigm that contribute to the development of the digital economy; 2) identifies the directions of implementation of digital competencies in the context of the European educational paradigm; 3) reveals digital tools and educational platforms that contribute to the formation of digital education; 4) formulates the concept of quality, inclusive, accessible digital education as a factor in improving digital competencies and adapting education to the digital age; 5) traces the impact of digital education and digital competencies on the development of the digital economy. The concept of digital education contains both its potential and its risks, which can have serious consequences for the future of the educational process if digital literacy is not developed. The combination of four factors – cultural change, technological innovation, national policy leadership and internal development of the digital education system – stimulates the digital transformation of society.

Fábio Eid Morooka, Adalberto Manoel Junior, Tiago F. A. C. Sigahi et al.

Applications of deep learning (DL) in autonomous vehicle (AV) projects have gained increasing interest from both researchers and companies. This has caused a rapid expansion of scientific production on DL-AV in recent years, encouraging researchers to conduct systematic literature reviews (SLRs) to organize knowledge on the topic. However, a critical analysis of the existing SLRs on DL-AV reveals some methodological gaps, particularly regarding the use of bibliometric software, which are powerful tools for analyzing large amounts of data and for providing a holistic understanding on the structure of knowledge of a particular field. This study aims to identify the strategic themes and trends in DL-AV research using the Science Mapping Analysis Tool (SciMAT) and content analysis. Strategic diagrams and cluster networks were developed using SciMAT, allowing the identification of motor themes and research opportunities. The content analysis allowed categorization of the contribution of the academic literature on DL applications in AV project design; neural networks and AI models used in AVs; and transdisciplinary themes in DL-AV research, including energy, legislation, ethics, and cybersecurity. Potential research avenues are discussed for each of these categories. The findings presented in this study can benefit both experienced scholars who can gain access to condensed information about the literature on DL-AV and new researchers who may be attracted to topics related to technological development and other issues with social and environmental impacts.

Yee Hui Yeo, Jamil S. Samaan, Wee Han Ng