搜索结果：Artificial life

Deep vein thrombosis (DVT) is the formation of thrombi in the deep venous system, most often in the lower extremities. Although usually not life-threatening, DVT requires timely diagnosis to prevent complications such as pulmonary embolism and post-thrombotic syndrome. The growing demand for image interpretation has generated interest in applying artificial intelligence (AI) to automated DVT detection. This scoping review analyzes the performance of artificial intelligence in diagnosing DVT using computed tomography (CT), magnetic resonance imaging (MRI), and ultrasound (US). We conducted a search across seven databases from inception to May 2025 using terms related to deep vein thrombosis, artificial intelligence, and machine learning. Eligible studies were limited to those evaluating DVT diagnosis using CT, MRI, or ultrasound. Two independent reviewers selected eligible studies, and quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2). Eleven studies published between 2021 and 2025 met the inclusion criteria. Some of the AI algorithms included RetinaNet, Deep R-Belief Neural Networks, and Sooty Tern Optimization. US-based models were the most studied algorithms, with sensitivities and specificities ranging from 68 to 100% and 70-100%, respectively. The MRI-based model achieved sensitivities, specificities, and accuracies of 95% to 97%. One CT-based model demonstrated a sensitivity of 83%. Studies evaluated across multiple imaging datasets showed high sensitivities, specificities, and precision of 96% or higher. Future research should prioritize multicenter validation and integration of clinical factors. In addition, explainable frameworks capable of integrating multiple imaging datasets must be developed with attention to workflow efficiency and cost-effectiveness to support clinical translation. The results indicate that AI is best situated as a supplementary tool rather than a replacement for expert interpretation in DVT diagnosis.

Grapes (Vitis vinifera L.) are recognized as the third most popular fruit worldwide. Nevertheless, table grapes are highly perishable. Accurate prediction of shelf-life duration and nutritional components is crucial in supply chain of fresh fruit and vegetables. In present study, we developed the machine learning models to predict the shelf-life and total anthocyanin content (TAC) of table grapes intelligently. The artificial neural network (ANN) accurately estimated shelf-life, with a mean absolute error (MAE) of 1.624 days and a coefficient of determination (R2) of 0.9873, whereas extreme gradient boosting (XGBoost) effectively predicted TAC (MAE 0.0113 mg C3G g-1； R2 0.9355). SHapley Additive exPlanations (SHAP)-based sensitivity analysis revealed the berry abscission rate, weight loss rate and postharvest treatment were critical factors influencing quality performance, improving model interpretability. These findings successfully provide a reliable approach for monitoring grape quality, providing novel insight into reducing the loss of fresh agricultural produce.

Parents of children with achondroplasia face sustained caregiving demands that may affect multiple dimensions of well-being. Despite growing recognition of these challenges, no validated, condition-specific instrument exists to assess the quality of life (QoL) of parents of children with achondroplasia. This study aimed to develop, and pilot test the Quality of Life of Parents of Children with Achondroplasia (QOLA) questionnaire. QOLA was developed using a multi-phase mixed-methods design in accordance with established standards for developing self-reported outcome measures for caregivers and parents. Phase 1 comprised semi-structured qualitative interviews with 17 parents of children with achondroplasia to identify relevant QoL domains and language. Interview data were analysed using qualitative content analysis and informed systematic item generation (Phase 2). Conceptual structure was examined through researcher-led card sorting (Phase 3) and two rounds of international card sorting following translation (Phase 4). The resulting 63-item questionnaire across eight domains was pilot-tested in a cross-sectional, multi-country study with embedded cognitive debriefing in Germany, Italy, and Portugal (total N = 50). The final pilot version of QOLA comprised 63 items across eight domains covering healthcare experiences, challenges and support, physical health, mental health, social life and relationships, coping, family and daily life, and worries and future concerns. Item-level missing data were minimal, and no pronounced floor or ceiling effects were observed. Internal consistency was acceptable to good for domains (α = 0.624-0.821) and good for the total scale (α = 0.798). Inter-domain correlations were generally moderate to strong. Cognitive debriefing was highly acceptable and relevant across countries, with some suggestions for further refinement. QOLA shows strong preliminary evidence of acceptability and internal consistency and addresses a key measurement gap in achondroplasia research. Further large-scale psychometric validation is warranted.

There is interest in the use of recent single-cell spatial transcriptomic technologies to gain biological insights into disease mechanisms. Previously, we characterized the use of herpes simplex virus 1 (HSV-1) induced neuroinflammation in 2D dissociated cells from human cerebral organoids (dcOrgs) to model molecular and transcriptomic readouts associated with Alzheimer's disease (AD). In this work, we generated two datasets by using single-cell non-spatial RNA sequencing and single-cell spatial RNA sequencing technologies on HSV-1-infected 2D dcOrgs and HSV-1-infected 3D cerebral organoids (cOrgs). We conducted cell type assignment for the cells in the 2D dcOrgs and 3D cOrgs, by using single-cell non-spatial RNA sequence data from human fetal brains and adult post-mortem brains, to infer the transcriptomic effects of AD-associated in-vitro perturbations through viral infections linked to AD. We evaluated computational and machine learning methods, including the use of multi-layer perceptrons (MLPs), and we used cross-2D/3D platform comparisons as a benchmark to evaluate the artificial neural network models. In the process, we found that the use of MLPs can lead to high validation rates for assigning cell type identities from 2D and 3D human cerebral organoids to cell types found in human adult post-mortem brain samples. Furthermore, the use of these technologies and systems enabled the identification of pseudotime trajectories and cell clusters associated with the viral transcriptional life cycle. We identified several cell types, including endothelial cells and astrocytes, with significantly more clustered cell-cell nearest neighbor distances in infected 3D cOrgs compared to mock 3D cOrgs. Permutation tests revealed that these differences in nearest neighbor distances are unlikely to be driven by overall structural differences between individual infected 3D cOrgs and mock 3D cOrgs, such as differences in the density of cells. Given that there are more large-scale single-cell non-spatial (2D) RNA sequence datasets that had been generated from human post-mortem brain samples, compared to single-cell spatial (3D) RNA sequence datasets from human post-mortem brain samples, the development of data integration approaches by using artificial neural networks such as MLPs, across 2D and 3D single-cell transcriptomics datasets generated from human post-mortem brain samples and human in-vitro systems such as brain organoids is likely to be critical to gain novel insights into neurodegenerative diseases such as AD.

In the context of accelerating integration of artificial intelligence (AI) technologies into education, rapidly assessing students' AI attitude is crucial for advancing digital transformation in education.The aim of this study is to develop and validate a new five-item version of the Artificial Intelligence Attitude Scale (AIAT-5). For examining the quality of the scale, a total sample of 1576 middle and high school students (study 1: 858; study 2: 614) from diverse regions in China completed the survey. Results: (1) Study1: AIAT-5 exhibited strong internal consistency (Cronbach's α = 0.923, McDonald's ω = 0.923, Guttman's λ6 = 0.911) and good discrimination (α = 3.201 ~ 4.513). Both EFA and EGA supported a robust unidimensional structure. (2) Study 2: CFA results indicated excellent model fit for unidimensional structure (CFI = 0.993, TLI = 0.976, and RMSEA = 0.028), and all factor loadings ranged from 0.879 to 0.953. Furthermore, AI attitude was positively correlated with knowledge of generative AI, life satisfaction, internet addiction, and epistemic curiosity. Overall, within the specific population of Chinese middle and high school students, the AIAT‑5 demonstrates initial validation as a reliable and practically efficient tool for large‑scale assessment of student AI attitude.

Artificial hydration (AH) in terminally ill cancer patients remains ethically and clinically controversial. Although evidence suggests limited benefit near the end of life, AH is frequently administered, particularly in East Asian settings where it may carry symbolic meaning. This study conceptualized physician decision-making as a clinical spectrum and examined demographic, clinical, and ethical factors associated with AH decision orientations. A nationwide cross-sectional survey of palliative care-trained physicians in Taiwan assessed clinical practices and ethical domains (autonomy, beneficence, non-maleficence, justice, cultural, and emotional factors). Scenario-based scores were used to derive continuation, withdrawal, and variability indices. Two-step cluster analysis identified decision-orientation profiles. Group differences were analyzed using ANOVA, chi-square tests, and multinomial logistic regression. Among 377 respondents, four decision-orientation clusters emerged: contextual/proportional, selective continuation, conservative/continuation-leaning balancing, and consistent withdrawal. Several demographic and professional characteristics differed across clusters. Ethical domains also differed significantly. Multinomial regression showed that cultural and emotional factors were associated with contextual/proportional orientation, whereas beneficence independently predicted selective continuation orientation. Hydration volume and consideration of life expectancy differed across clusters, supporting behavioral distinctions among decision orientations. AH decision-making reflects multiple context-sensitive physician orientations shaped by both ethical considerations and professional characteristics, rather than a binary continuation-withdrawal model.

Oysters bioaccumulate the natural radionuclide 210Po, which is the main contributor to ingestion dose from seafood and with oyster aquaculture expanding globally, there is a need to evaluate radionuclide exposure from this and other radionuclides. Activity concentrations of natural and artificial radionuclides in the oyster Crassostrea gigas from seven Irish farms were determined. Activity concentrations of 40K and 210Po were consistently the highest, with 210Po showing the greatest variability among farms. Using the International Commission on Radiological Protection (ICRP) adult ingestion coefficients, a mean ingestion dose of 62 ± 36 μSv kg-1 of fresh oyster tissue consumed was estimated, with 210Po contributing ≈99% and negligible (≈0.03%) contributions from artificial radionuclides. For an assumed typical Irish consumer (0.69 kg yr-1), the annual mean dose was estimated as 42 ± 25 μSv, which is a similar scale to the total dietary 210Po dose of 65 ± 21 μSv. For heavy consumers (12.1 kg yr-1) this dose could reach 740 ± 441 μSv yr-1, which is several times the typical dietary ingestion dose from all sources of radiation, but similar to other high seafood consumption rate dose estimates. These results underline that 210Po is the primary driver of dose from oysters, with negligible contributions from other natural and artificial contaminants. Consequently, future seafood dose assessments should prioritise 210Po measurements and consider high consumption rate groups. Furthermore, due to a lack of clear information on the consumption rates of oysters, there is a need to establish more accurate data on the consumption rates of seafood.

The Viking missions showcased multiple spaceflight technologies that represented state-of-the-art capabilities: From digital line-scan imaging to the operation of complex onboard laboratories and software-controlled process autonomy. Since Viking, there have been extraordinary, and still accelerating, advancements in computing technology that impact science, society, and exploration. These developments have occurred in both hardware and software and have resulted in increasingly capable devices, advanced programming tools, and algorithmic innovations. The subset of artificial intelligence known as machine learning has emerged as one of the most transformative of these developments; it has major implications for space exploration and for improvements to the search for evidence of life beyond Earth. Those improvements include the integration of data across different scales and increased sensitivity to complex features in data, as well as the generation of adaptive strategies for sampling environments. In this article, the present and future nature of space exploration and astrobiological research is examined through the contextual lens of Viking and through the history and possible future of artificial intelligence.

Traditional methods to practice patient interviewing skills for student pharmacists include the use of actors and role-playing. While effective, they can be difficult to manage with large class sizes, cost, and time to train. Artificial intelligence (AI) and large language models (LLMs) are readily accessible to offer students a scalable solution to practice collecting information from a controlled, dynamic simulated patient (SP). Five programs collaborated to create multiple educational activities that utilized AI as an SP for student pharmacists to practice their patient interviewing skills. Each activity included patient cases focused on self-care related topics. Student participants ranged from professional year one to three (P1-P3). A voluntary, anonymous survey was developed to assess baseline demographics and student perceptions of the activities. Out of 602 students across the five programs, 304 (50.5%) voluntarily completed the survey. Most students were born in Generation Z (1997-2012), identified as white/Caucasian, and had previously used AI. The majority of the students used the written feature to complete the activity. Overall, students reported that the AI activity was easy to use and considered it useful to practice patient interviewing skills. Students from two programs favored the use of the AI activity, while the other three programs favored traditional role-playing with partners. Students stated they missed the real-life human component of practicing with a partner. AI can be utilized as an SP to assist students in practicing their patient interviewing skills with a variety of topics. Students found the LLM easy to use, helped with their interviewing skills, and wanted to see it utilized more in the pharmacy curriculum. Future activities should find ways to incorporate more real-life components.

Late-life depression (LLD) is common and disabling in older adults, and current pharmacological or invasive treatments are often limited by comorbidity and tolerability. Nutrition is a modifiable target with potential clinical value. This review examines the epidemiological association between nutritional status and LLD. It synthesizes evidence on essential nutrients and dietary patterns, discusses potential underlying mechanisms, and evaluates the clinical utility of nutritional assessment tools-such as the Geriatric Nutritional Risk Index (GNRI), the Mini Nutritional Assessment (MNA) combined with the Geriatric Depression Scale (GDS), and the Comprehensive Geriatric Assessment (CGA)-for identification and intervention. Overall, malnutrition and nutritional risk are consistently associated with greater depressive symptom burden in later life, and emerging data suggest that chrono-nutrition (particularly higher energy intake at breakfast) may be a relevant, under-recognized modifier of depression risk. Mechanistically, nutrition may influence LLD through neuroinflammation, neuroplasticity, gut-brain axis signaling, and oxidative/mitochondrial pathways. Pattern-based strategies appear most actionable: higher adherence to Mediterranean-type diets and Mediterranean-Dietary Approaches to Stop Hypertension Intervention for Neurodegenerative Delay (MIND) diets is generally linked to fewer depressive symptoms, whereas Western/processed-food patterns are generally associated with adverse outcomes; for plant-based approaches, dietary quality is critical. For nutrient-focused interventions, effects remain heterogeneous; benefits may depend on baseline deficiency, inflammatory status, and for omega-3, the eicosapentaenoic acid (EPA) to docosahexaenoic acid (DHA) ratio and dose. For clinical implementation, we highlight an assessment-to-intervention workflow integrating GNRI-based rapid stratification, MNA plus GDS screening, and CGA-guided multidisciplinary management for frail or complex patients. Future research should prioritize adequately powered randomized controlled trials (RCTs) with standardized protocols, biomarker-informed and genotype-aware stratification, and interdisciplinary translation to optimize nutrition-based prevention and comprehensive management of LLD.

Generalization is a fundamental criterion for evaluating learning effectiveness, a domain where biological intelligence excels yet artificial intelligence faces challenges. In biological learning and memory, the well-documented spacing effect shows that appropriately spaced intervals between learning trials significantly improve behavioral performance. While multiple theories have been proposed to explain its underlying mechanisms, one compelling hypothesis is that spaced training promotes integration of input and innate variations, thereby enhancing generalization to novel but related scenarios. Here, we examine this hypothesis by introducing a bio-inspired spacing effect into artificial neural networks, integrating input and innate variations across spaced intervals at neuronal, synaptic, and network levels. These spaced ensemble strategies yield significant performance gains across benchmark datasets and network architectures. Biological experiments on Drosophila further validate the complementary effect of appropriate variations and spaced intervals in improving generalization, which together reveal a convergent computational principle of biological learning and machine learning.

Health technology assessment bodies increasingly emphasise the importance of preference-weighted health-related quality of life (HRQoL) evidence. However, such measures are often absent in clinical trial publications. It is not yet clear how frequently clinical trials have incorporated these measures over the past five decades, how the use of preference-weighted HRQoL instruments has evolved over time, and how trends differ across disease areas, countries and global regions. This study aims to (1) assess changes over time in the proportions of clinical trials using each preference-weighted HRQoL instrument in adults, and (2) model secular trends in the adoption of these instruments across disease areas, countries and regions. The study will provide a comprehensive, systematic assessment of the use of preference-weighted HRQoL instruments in clinical trials since 1976 and develop a scalable approach for large-scale evidence synthesis. We will identify clinical trials involving humans published in English since 1976 through systematic searches of MEDLINE, Embase, Cochrane Library and Web of Science. We will focus on generic preference-weighted HRQoL instruments for adults, including EQ-5D-3L, EQ-5D-5L, Short Form 6 Dimensions, 12-Item Short Form Health Survey (SF-12), Health Utility Index 2, Health Utility Index 3, Assessment of Quality of Life (AQoL) series (AQoL-4D, AQoL-6D, AQoL-7D, AQoL-8D), Quality of Well-Being Scale (QWB), QWB Self-Administered (QWB-SA), 15D and Patient-Reported Outcomes Measurement Information System (PROMIS) with the Preference Scoring System (PROPr). Screening and data extraction will be automated using natural language processing (NLP) pipeline or large language models (LLMs). To determine the most accurate approach, we will benchmark NLP and LLM performance against a manually curated reference dataset of 5000 randomly sampled articles reviewed independently by three reviewers. Model performance will be evaluated using classification metrics including accuracy, recall and F1-score. Annual counts and proportions of trials using each instrument will be calculated, stratified by disease area, country and region. Trends will be modelled using basis-splines (B-splines) with 2 or 3 degrees of freedom and Bayesian spline regression to estimate secular changes in both absolute numbers and proportions of instrument use over time. This study uses only published literature and does not involve human participants or individual-level data. All results will be reported in aggregate form, with no identifiable information. Formal ethics approval is therefore not required. Findings will be disseminated via peer-reviewed publications and conference presentations, and aggregated data and analysis code will be made publicly available to support transparency and reproducibility.

Family history, body mass index (BMI), and ethnicity are three key, well-established determinants of susceptibility to type 2 diabetes mellitus (T2DM), reflecting genetic predisposition, modifiable metabolic risk, and biological as well as social influences, respectively. These factors interact in complex, non-linear patterns that are not fully captured by conventional risk prediction models. This review examines how artificial intelligence (AI) and machine learning approaches can integrate these variables to improve risk stratification and early identification of individuals at high risk of T2DM. By leveraging large-scale, longitudinal datasets, data-driven models facilitate the capture of population-level heterogeneity and identify risk patterns that extend beyond static thresholds. Incorporating AI-enhanced prediction tools into clinical and public health settings could enable more timely, targeted, and equitable interventions. Ultimately, integrating advances in AI with a deeper understanding of the interplay between BMI, ethnicity, and genetic predisposition may support more personalised prevention strategies and risk-stratified care pathways for T2DM.

High-quality artificial intelligence (AI) models in endodontics require access to diverse, well-annotated datasets. This review introduces federated learning (FL) as a privacy-preserving framework for collaborative AI in endodontics. In this narrative review, a comprehensive literature search was conducted across databases, encompassing studies published up to April 2026. The search strategy was intentionally broad to facilitate a thorough exploration of the evolution of the relevant concepts. However, to ensure consistency in comprehension and analysis, the review was limited to publications in English. Studies describing the fundamentals and applications of FL were reviewed and comparatively analysed. The narrative review served as the framework for outlining implementation pathways, challenges, and research priorities for its application to diagnostic and decision-support tasks. Traditional centralised training methods face legal and ethical challenges due to data protection regulations. FL allows institutions to retain local patient data while contributing model updates to a central server or decentralised network, thus providing a viable alternative. The article explores FL principles, privacy and security mechanisms, architectures, technical challenges, and adversarial risks. Regulatory and ethical considerations hinge on a mix of advanced technical measures and robust organisational practices. A proposed roadmap for implementing FL includes pilot studies, standardised data processes, clinical validation, and regulatory engagement. FL promises to enhance AI development in endodontics while safeguarding patient privacy, with potential benefits in diagnostics and personalised care. FL could advance AI integration in endodontics, prioritising the protection of patient privacy. This initiative holds the potential to improve diagnostic processes and facilitate personalised treatment approaches. FL enables the development of multicentre AI models without sharing patient data. By leveraging diverse clinical datasets, this approach may improve the accuracy and generalisability of AI systems for endodontic diagnosis, treatment planning, and outcome prediction while preserving patient privacy.

Artificial light at night (ALAN) disrupts the physiology and behavior of coastal marine animals globally, but the cellular mechanisms underlying these effects remain unclear. We defined sleep in the damselfish Chromis viridis, tracked school dynamics within their coral habitats, and determined the acute and chronic effects of ALAN on behavioral interactions, sleep, and neuronal health under both controlled laboratory and natural reef conditions. We found that ALAN increased territorial occupancy, aggression, and nocturnal feeding while reducing sleep duration and consolidation. These sleep disruptions correlated with increased DNA damage in neurons of the dorsal pallium, a brain region involved in sleep-dependent brain functions. Our findings introduce a model that links ALAN-dependent alterations in sleep with neuronal insults in reef-dwelling tropical fish.

The longevity industry is rapidly expanding, driven by advances in biotechnology and consumer demand. In this News and Perspectives article, JMIR Correspondent Jenna Congdon reports on recent developments in longevity medicine alongside potential risks and ethical concerns.

Heart failure (HF) affects 11.8% of adults aged 65 and older, reducing quality of life and longevity. Preventing HF can reduce morbidity and mortality. We hypothesized that artificial intelligence (AI) applied to 24-hour single-lead electrocardiogram (ECG) data could predict the risk of HF within five years. To research this, the Technion-Leumit Holter ECG (TLHE) dataset, including 69,663 recordings from 47,729 patients, collected over 20 years, was used. Our deep learning model, DeepHHF, trained on 24-hour ECG recordings, achieved an area under the receiver operating characteristic curve of 0.80 that outperformed a model using 30-second segments and a clinical score. High-risk individuals identified by DeepHHF had a two-fold chance of hospitalization or death incidents. Explainability analysis showed DeepHHF focused on arrhythmias and heart abnormalities. This study highlights the feasibility of deep learning to model 24-hour continuous ECG data, capturing paroxysmal events essential for reliable risk prediction. Artificial intelligence applied to single-lead Holter ECG is non-invasive, inexpensive, and widely accessible, making it a promising tool for HF risk prediction.

Sleep supports cardiovascular, metabolic, neurologic, and psychological health. Beyond duration, circadian alignment is crucial because regular sleep, light exposure, feeding, and activity synchronize central and peripheral clocks. Modern chronodisruption from artificial light, irregular schedules, shift work, late eating, and insufficient sleep misaligns these rhythms and increases chronic disease risk. This review summarizes evidence linking sleep and circadian disruption to cardiovascular disease, metabolic dysfunction, and neurodegeneration. The American Heart Association's Life's Essential 8 recognizes sleep as a core cardiovascular health metric, reflecting evidence that sleep duration, timing, regularity, and quality add predictive value beyond traditional risk factors. We also review data from cohorts, proteomics, wearables, and artificial intelligence showing that objective sleep traits can predict cardiometabolic and neurologic outcomes. Both short and long sleep duration are associated with adverse metabolic outcomes, likely through partly distinct mechanisms. Finally, we discuss clinical implications, including light hygiene, meal timing, sleep disorder treatment, sleep metrics for early risk detection, and links between circadian disruption, gut microbiota metabolic reprogramming, clock-gene regulation, and diet-based strategies targeting gut-liver and metabolic pathways.

The secondary use of health data holds substantial potential for advancing biomedical research, strengthening population health analytics, and enabling artificial intelligence-driven decision-making support. Yet, ensuring that such reuse respects patient autonomy, privacy, and regulatory obligations remains a major challenge. Conventional consent mechanisms are typically static, difficult to revoke, and offer limited transparency or accountability after data disclosure. This review aimed to systematically examine blockchain-based frameworks that enable dynamic, auditable, and revocable consent for the secondary use of health data. A structured literature search was conducted in PubMed, Scopus, and Web of Science covering the period 2020 to 2025. Following PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines, 55 peer-reviewed studies meeting predefined inclusion criteria were analyzed. Data extraction focused on four dimensions: (1) consent life cycle management, (2) auditability and traceability, (3) usability and patient empowerment, and (4) legal and ethical alignment. Findings indicate that blockchain technologies provide a robust foundation for automating consent life cycles, ensuring immutable auditability, and enabling decentralized patient control. Most frameworks used smart contracts, decentralized identifiers, and verifiable credentials to implement programmable and verifiable consent processes. Nevertheless, key challenges persist, including limited usability testing, complexities in real-time revocation propagation, interoperability gaps with clinical systems, and tensions with regulatory requirements such as the General Data Protection Regulation right to erasure. Only a small subset of studies reported real-world deployments or user-centered evaluations. Blockchain offers substantial promise for improving the trustworthiness, transparency, and accountability of consent management for secondary health data use. However, wider adoption requires human-centered design approaches, stronger interoperability through standards such as Fast Healthcare Interoperability Resources, verifiable credentials, and consent receipts, and clearer legal guidance for compliance. Future research should prioritize integrating blockchain-enabled consent infrastructures into national and cross-border digital health ecosystems such as the European Health Data Space to support secure, patient-controlled, and ethically governed secondary data use.