Search Results (257)

Background and Objectives: Diffusion models, as a recent advancement in generative modeling, have become central to high-resolution image synthesis and reconstruction. Their rapid progress has notably shaped computer vision and health informatics, particularly by enhancing medical imaging and diagnostic workflows. However, despite these developments, researchers continue to face challenges due to the absence of a structured and comprehensive discussion on the use of diffusion models within clinical imaging. Methods: This systematic review investigates the application of diffusion models in medical imaging for diagnostic purposes. It provides an integrated overview of their underlying principles, major application areas, and existing research limitations. The review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines and included peer-reviewed studies published between 2013 and 2024. Studies were eligible if they employed diffusion models for diagnostic tasks in medical imaging; non-medical studies and those not involving diffusion-based methods were excluded. Searches were conducted across major scientific databases prior to the review. Risk of bias was assessed based on methodological rigor and reporting quality. Given the heterogeneity of study designs, a narrative synthesis approach was used. Results: A total of 68 studies met the inclusion criteria, spanning multiple imaging modalities and falling into eight major application categories: anomaly detection, classification, denoising, generation, reconstruction, segmentation, super-resolution, and image-to-image translation. Explainable AI components were present in 22.06% of the studies, clinician engagement in 57.35%, and real-time implementation in 10.30%. Overall, the findings highlight the strong diagnostic potential of diffusion models but also emphasize the variability in reporting standards, methodological inconsistencies, and the limited validation in real-world clinical settings. Conclusions: Diffusion models offer significant promise for diagnostic imaging, yet their reliable clinical deployment requires advances in explainability, clinician integration, and real-time performance. This review identifies twelve key research directions that can guide future developments and support the translation of diffusion-based approaches into routine medical practice. Full article

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder characterized by cognitive and motor decline. Early detection remains challenging, as traditional neuroimaging and neuropsychological assessments often fail to capture subtle, preclinical changes. Recent advances in digital health and artificial intelligence (AI) offer new opportunities to identify non-invasive biomarkers of cognitive impairment. In this study, we propose an AI-driven framework for early AD based on handwriting motion data captured using a sensor-integrated Smart Pen. The system employs an inertial measurement unit (MPU-9250) to record fine-grained kinematic and dynamic signals during handwriting and drawing tasks. Multiple machine learning (ML) algorithms—Logistic Regression, Support Vector Machine (SVM), Random Forest (RF), and k-Nearest Neighbors (kNN)—and deep learning (DL) architectures, including one-dimensional Convolutional Neural Networks (1D-CNN), Long Short-Term Memory (LSTM), and a hybrid CNN-BiLSTM network, were systematically evaluated. To address data scarcity, we implemented a Sim-to-Real Domain Adaptation strategy, augmenting the training set with physics-based synthetic samples. Results show that classical ML models achieved moderate diagnostic performance (AUC: 0.62–0.76), while the proposed hybrid DL model demonstrated superior predictive capability (accuracy: 0.91, AUC: 0.96). These findings underscore the potential of motion-based digital biomarkers for the automated, non-invasive detection of AD. The proposed framework represents a cost-effective and clinically scalable informatics solution for digital cognitive assessment. Full article

The continuous advancement of modern weaponry has intensified the pursuit of next-generation ballistic protection systems that integrate lightweight architectures, superior flexibility, and high energy absorption efficiency. This review provides a technological overview of current trends in the design, processing, and performance optimization of metallic, ceramic, polymeric, and composite materials for ballistic applications. Particular emphasis is placed on the role of advanced surface coatings and nanostructured interfaces as enabling technologies for improved impact resistance and multifunctionality. Conventional materials such as high-strength steels, alumina, silicon carbide, boron carbide, Kevlar^®, and ultra-high-molecular-weight polyethylene (UHMWPE) continue to dominate the field due to their outstanding mechanical properties; however, their intrinsic limitations have prompted a transition toward nanotechnology-assisted solutions. Functional coatings incorporating nanosilica, graphene and graphene oxide, carbon nanotubes (CNTs), and zinc oxide nanowires (ZnO NWs) have demonstrated significant enhancement in interfacial adhesion, inter-yarn friction, and energy dissipation. Moreover, multifunctional coatings such as CNT- and laser-induced graphene (LIG)-based layers integrate sensing capability, electromagnetic interference (EMI) shielding, and thermal stability, supporting the development of smart and adaptive protection platforms. By combining experimental evidence with computational modeling and materials informatics, this review highlights the technological impact of coating-assisted strategies in the evolution of lightweight, high-performance, and multifunctional ballistic armor systems for defense and civil protection. Full article

This study analyses the progress and persistent challenges in achieving Sustainable Development Goal 8—Decent Work and Economic Growth (SDG 8)—in the province of Alto Amazonas, Loreto, Peru, a territory characterized by structural informality exceeding 80%. A mixed-methods design was employed, integrating a survey of 500 economically active residents, semi-structured interviews with local authorities and business representatives, and a documentary review of official data from the National Institute of Statistics and Informatics (INEI) and the Ministry of Economy and Finance (MEF). Quantitative results reveal uneven economic growth driven mainly by low-value primary sectors, with 41.2% of workers lacking social protection and 51.4% reporting discriminatory practices. Although 70% expressed interest in entrepreneurship, only 37.8% achieved business formalization. Qualitative findings highlight a strong dependence on public investment, limited private-sector diversification, and an entrepreneurial ecosystem with high motivation but insufficient institutional support. The study concludes that structural constraints—informality, credit restrictions, territorial inequality, and weak institutional coordination—continue to hinder SDG 8 achievement. It recommends integrated policies that promote labor formalization, financial inclusion, productive diversification, and sustainable micro-enterprise development to align economic dynamism with social protection and territorial cohesion. Full article

This review investigates the application of Explainable Artificial Intelligence (XAI) in biomedical informatics, encompassing domains such as medical imaging, genomics, and electronic health records. Through a systematic analysis of 43 peer-reviewed articles, we examine current trends, as well as the strengths and limitations of methodologies currently used in real-world healthcare settings. Our findings highlight a growing interest in XAI, particularly in medical imaging, yet reveal persistent challenges in clinical adoption, including issues of trust, interpretability, and integration into decision-making workflows. We identify critical gaps in existing approaches and underscore the need for more robust, human-centred, and intrinsically interpretable models, with only 44% of the papers studied proposing human-centred validations. Furthermore, we argue that fairness and accountability, which are key to the acceptance of AI in clinical practice, can be supported by the use of post hoc tools for identifying potential biases but ultimately require the implementation of complementary fairness-aware or causal approaches alongside evaluation frameworks that prioritise clinical relevance and user trust. This review provides a foundation for advancing XAI research on the development of more transparent, equitable, and clinically meaningful AI systems for use in healthcare. Full article

Hydrogel additive manufacturing underpins soft tissue models, biointerfaces, and soft robotics. The coupled choices of formulation, rheology, and process conditions limit the progress. This review maps how artificial intelligence links composition to printability across direct ink writing, inkjet, vat photopolymerization, and laser-induced forward transfer, and how vision-guided control improves fidelity and viability during printing. Interpretable predictors connect routine rheology to strand stability, data-driven classifiers chart droplet regimes, and optical dose models with learning enhance voxel accuracy. Polymer informatics, including BigSMILES based representations, supports generative screening of precursors and crosslinkers. Bayesian optimization and active learning reduce experimental burden while honoring biological constraints, and emerging autonomous platforms integrate in situ sensing with rapid iteration. A strategic framework outlines a technological progression from current open-loop data gathering toward real-time closed-loop correction and ultimately predictive fault prevention through digital twins. The synthesis provides quantitative routes from formulation through process to function, establishing a practical foundation for predictive, reproducible hydrogel manufacturing and application-oriented design. Full article

Forest foods play a vital role in enhancing dietary diversity, human health, and the sustainable use of forest ecosystems. However, structured and machine-readable resources that systematically describe their taxonomic and nutritional attributes remain scarce. To fill this gap, we introduce ForestFoodKG, a comprehensive resource that integrates taxonomic hierarchy and nutritional composition of 1191 forest food items. The resource consists of two components—(i) the ForestFoodKG dataset, containing standardized taxonomic and nutritional records across seven biological levels, and (ii) the ForestFoodKG Knowledge Graph (ForestFoodKG-KG), which semantically links forest food entities using named entity recognition and relation extraction. The constructed graph comprises 4492 entities and 14,130 semantic relations, providing a structured foundation for intelligent querying, nutrition analytics, and ecological informatics. All data were manually verified and made publicly available in CSV format on GitHub. ForestFoodKG serves as the first structured knowledge base for forest foods, promoting data-driven research in nutrition science, sustainable forestry, and knowledge-based decision-making. Full article

Artificial intelligence (AI) has transformed the educational landscape and reshaped learning experiences. Its adoption in higher education is increasing due to the recent plethora of AI tools (AITs) and their associated benefits. Romanian universities face the challenge of integrating AITs in the learning process. Thus, the students’ attitudes and behavioral intentions concerning the use of AITs are meaningful. Technology acceptance models have been widely used to investigate factors that affect the intention to use a technology. ChatGPT (Chat Generative Pre-Trained Transformer) is a popular AIT among students. Therefore, this study presents a conceptual model for successfully adopting ChatGPT in a Romanian Higher Education Institution (HEI). A case study was conducted at the Faculty of Cybernetics, Statistics, and Economic Informatics at the Bucharest University of Economic Studies to test this model. Structural equation modeling (SEM) was used to validate and inspect the model’s network of determinants. The findings indicate that perceived ease of use (PEOU) and perceived usefulness (PU) are key predictors of student satisfaction (S) and trust (T), which in turn promote loyalty (L) to the AIT. The paper provides a novel perspective by distinguishing between forms of social presence, and their impact on students’ satisfaction and trust, thereby enhancing the understanding of student behavior toward AIT adoption. Full article

Advancements in deep learning have revolutionized materials discovery by enabling predictive modeling of complex material properties. However, single-modal approaches often fail to capture the intricate interplay of compositional, structural, and morphological characteristics. This study introduces a novel multimodal deep learning framework for enhanced material property prediction, integrating textual (chemical compositions), tabular (structural descriptors), and image-based (2D crystal structure visualizations) modalities. Utilizing the Alexandriadatabase, we construct a comprehensive multimodal dataset of 10,000 materials with symmetry-resolved crystallographic data. Specialized neural architectures, such as FT-Transformer for tabular data, Hugging Face Electra-based model for text, and TIMM-based MetaFormer for images, generate modality-specific embeddings, fused through a hybrid strategy into a unified latent space. The framework predicts seven critical material properties, including electronic (band gap, density of states), thermodynamic (formation energy, energy above hull, total energy), magnetic (magnetic moment per volume), and volumetric (volume per atom) features, many governed by crystallographic symmetry. Experimental results demonstrated that multimodal fusion significantly outperforms unimodal baselines. Notably, the bimodal integration of image and text data showed significant gains, reducing the Mean Absolute Error for band gap by approximately 22.7% and for volume per atom by 22.4% compared to the average unimodal models. This combination also achieved a 28.4% reduction in Root Mean Squared Error for formation energy. The full trimodal model (tabular + images + text) yielded competitive, and in several cases the lowest, error metrics, particularly for band gap, magnetic moment per volume and density of states per atom, confirming the value of integrating all three modalities. This scalable, modular framework advances materials informatics, offering a powerful tool for data-driven materials discovery and design. Full article

Background and Objectives: Medical diagnosis documents often exhibit diverse layouts and formats, posing significant challenges for automated information extraction. Ensuring the privacy of sensitive medical data further complicates the development of effective analysis systems. This study aims to develop a robust and privacy-compliant system for analyzing medical diagnosis documents. Methods: We designed an integrated Optical Character Recognition (OCR) system that processes medical documents regardless of their layout or format. The system first converts bitmap images into machine-readable text using OCR. A document-understanding model is then applied to identify and extract key information. To improve adaptability and accuracy, we employed a mutual learning approach. To address privacy concerns, we generated training data using generative techniques, ensuring compliance with privacy regulations while maintaining dataset quality. Results: The proposed system demonstrated strong performance across a wide variety of document layouts, effectively extracting critical information while adhering to privacy requirements. Conclusions: Our approach offers a practical and efficient solution for processing complex medical diagnosis documents, advancing the field of medical informatics while safeguarding patient privacy. Full article

Virtual exhibitions of cultural heritage (CH) have become a key means for preservation, education, and global dissemination in the digital era. This study provides a comprehensive systematic review and bibliometric analysis of CH virtual exhibition research from 1999 to 2025. A total of 651 valid records were retrieved from the Web of Science Core Collection following the PRISMA 2020 guidelines. Three tools (CiteSpace, VOSviewer, and Bibliometrix) support stronger analysis. Results reveal that the field’s knowledge structure can be organized into the following three interrelated layers: (1) a technology-driven layer (laser scanning, photogrammetry, VR/AR, and multimodal interaction), (2) a systemic application layer (curatorial workflows, digital museums, and immersive storytelling), and (3) a user experience layer (educational impact, gamification, and trust building). These dimensions form a cyclical pyramid framework linking innovation, interpretation and perception. The study identifies persistent regional disparities, with China and Italy leading in publication volume, while countries such as Denmark and Australia achieve higher citation impacts due to advanced policy support and digital strategies. Emerging trends highlight the growing integration of gamified learning, AI-assisted curation, and immersive narrative design. These reflect a paradigm shift from technological demonstration to cultural interpretation. This study establishes a holistic analytical framework for understanding the evolution and future directions of CH virtual exhibitions, providing an essential reference for researchers, curators, and policymakers in the heritage informatics domain. Full article

Major depressive disorder (MDD) affects approximately $4.4\%$ of the global population. Its prevalence is increasing among adolescents and has led to the psychosocial condition known as hikikomori. MDD is typically assessed by self-report questionnaires, which, although informative, are subject to evaluator bias and subjectivity. To address these limitations, recent studies have explored machine learning (ML) for automated MDD detection. Among the input data used, speech signals stand out due to their low cost and minimal intrusiveness. However, many speech-based approaches lack integration with cognitive behavioral therapy (CBT) and adherence to evidence-based, patient-centered care—often aiming to replace rather than support clinical monitoring. In this context, we propose ML models to assess MDD in hikikomori patients using speech data from a real-world clinical trial. The trial is conducted in Italy, supervised by physicians, and comprises an eight-session CBT plan that is clinical evidence-based and follows patient-centered practices. Patients’ speech is recorded during therapy, and the Mel-Frequency Cepstral Coefficients (MFCCs) and wav2vec 2.0 embedding are extracted to train the models. The results show that the Multi-Layer Perceptron (MLP) predicted depression outcomes with a Root Mean Squared Error (RMSE) of $0.064$ using only MFCCs from the first session, suggesting that early-session speech may be valuable for outcome prediction. When considering the entire CBT treatment (i.e., all sessions), the MLP achieved an RMSE of $0.063$ using MFCCs and a lower RMSE of $0.057$ with wav2vec 2.0, indicating approximately a $9.5\%$ performance improvement. To aid the interpretability of the treatment outcomes, a binary task was conducted, where Logistic Regression (LR) achieved $70\%$ recall in predicting depression improvement among young adults using wav2vec 2.0. These findings position speech as a valuable predictive tool in clinical informatics, potentially supporting clinicians in anticipating treatment response. Full article

Enhancing fuel efficiency in public transportation requires the integration of complex multimodal data into interpretable, decision-relevant insights. However, traditional analytics and visualization methods often yield fragmented outputs that demand extensive human interpretation, limiting scalability and consistency. This study presents a multi-agent framework that leverages multimodal large language models (LLMs) to automate data narration and energy insight generation. The framework coordinates three specialized agents, including a data narration agent, an LLM-as-a-judge agent, and an optional human-in-the-loop evaluator, to iteratively transform analytical artifacts into coherent, stakeholder-oriented reports. The system is validated through a real-world case study on public bus transportation in Northern Jutland, Denmark, where fuel efficiency data from 4006 trips are analyzed using Gaussian Mixture Model clustering. Comparative experiments across five state-of-the-art LLMs and three prompting paradigms identify GPT-4.1 mini with Chain-of-Thought prompting as the optimal configuration, achieving 97.3% narrative accuracy while balancing interpretability and computational cost. The findings demonstrate that multi-agent orchestration significantly enhances factual precision, coherence, and scalability in LLM-based reporting. The proposed framework establishes a replicable and domain-adaptive methodology for AI-driven narrative generation and decision support in energy informatics. Full article

Background/Objectives: Guibi-tang (GBT) and its variant Kami-guibi-tang (KGBT) are traditional East Asian multi-herb formulas prescribed for memory loss, insomnia, and fatigue. Preclinical data suggest multimodal neuroprotective actions, including cholinergic signaling modulation and activation of the cAMP response element-binding protein (CREB)/extracellular signal-regulated kinase (ERK) pathway; however, clinical evidence for cognitive disorders remains scattered. This scoping review aimed to map the breadth, design characteristics, efficacy signals, and safety profile of GBT and KGBT across the full spectrum of cognitive impairment. Methods: Following the Arksey–O’Malley framework and PRISMA-ScR guidelines, seven databases were searched (MEDLINE, Embase, Cochrane Library, China National Knowledge Infrastructure, ScienceON, Scopus, Citation Information by the National Institute of Informatics) from inception to 31 January 2025, for human studies evaluating GBT or KGBT in subjective cognitive decline, mild cognitive impairment (MCI), dementia, or post-stroke cognitive impairment (PSCI). Two reviewers independently screened, extracted, and charted data on study design, participants, interventions, outcomes, and adverse events. Results: Fifteen studies met the inclusion criteria—nine randomized controlled trials, one crossover trial, and five observational reports—enrolling 555 participants (age range, 59–87 years). All were conducted in the Republic of Korea, Japan, or China. GBT or KGBT, given as monotherapy or adjunctive therapy for 4 weeks to 9 months, produced modest but consistent improvements in global cognition (Mini-Mental State Examination/Montreal Cognitive Assessment), memory domains, activities of daily living, and neuropsychiatric symptoms across MCI, Alzheimer’s disease, and PSCI cohorts. Reported adverse event rates were comparable to or lower than those of placebo, usual care, or conventional drugs, and no serious treatment-related toxicity was identified. Conclusions: Current evidence—although limited by small sample sizes, heterogeneous formulations, short follow-up durations, and regional concentration—indicates that GBT and KGBT are well tolerated and confer clinically meaningful cognitive and functional benefits. Standardized, multicenter, placebo-controlled trials with biomarker end points are warranted to confirm long-term efficacy, clarify mechanisms, and guide integrative clinical use. Full article

In the era of digital intelligence, information literacy (IL) competency has become a critical indicator for measuring the comprehensive quality and sustainable development potential of university’s education. Using Yancheng Institute of Technology as a case study, this study systematically elucidates the connotation and current development status of college students’ IL within the framework of sustainable development. An evaluation index system is constructed, comprising four dimensions: information awareness and attitude, information ethics, law and security, information knowledge and skills, and information integration and innovation. The study employs the Analytic Hierarchy Process (AHP) to determine the weights of indicators at various levels and integrates the Fuzzy Comprehensive Evaluation Method (FCEM) to establish a quantitative assessment model for IL competency. Empirical research demonstrates that the proposed model effectively enables a multidimensional and quantitative evaluation of students’ IL, with results that exhibit sound scientific validity and applicability. Based on the analysis, specific strategies are proposed to enhance students’ IL from the perspectives of curriculum design, teaching models, and library services, thereby providing theoretical references and practical pathways for advancing informatization and sustainable development in higher education. Full article