Search Results (129)

As artificial intelligence (AI) is increasingly used in classrooms, tutoring systems, and learning platforms, it is essential that these tools are not only powerful, but also easy to understand, fair, and supportive of real learning. Many current AI systems can generate fluent responses or accurate predictions, yet they often fail to clearly explain their decisions, reflect students’ cultural contexts, or give learners and educators meaningful control. This gap can reduce trust and limit the educational value of AI-supported learning. This paper introduces the PEARL framework, a human-centered approach for designing and evaluating explainable AI in education. PEARL is built around five core principles: Pedagogical Personalization (adapting support to learners’ levels and curriculum goals), Explainability and Engagement (providing clear, motivating explanations in everyday language), Attribution and Accountability (making AI decisions traceable and justifiable), Representation and Reflection (supporting fairness, diversity, and learner self-reflection), and Localized Learner Agency (giving learners control over how AI explains and supports them). Unlike many existing explainability approaches that focus mainly on technical performance, PEARL emphasizes how students, teachers, and administrators experience and make sense of AI decisions. The framework is demonstrated through simulated examples using an AI-based tutoring system, showing how PEARL can improve feedback clarity, support different stakeholder needs, reduce bias, and promote culturally relevant learning. The paper also introduces the PEARL Composite Score, a practical evaluation tool that helps assess how well educational AI systems align with ethical, pedagogical, and human-centered principles. This study includes a small exploratory mixed-methods user study (N = 17) evaluating example AI tutor interactions; no live classroom deployment was conducted. Together, these contributions offer a practical roadmap for building educational AI systems that are not only effective, but also trustworthy, inclusive, and genuinely supportive of human learning. Full article

Extracellular vesicles (EVs) are known as potential biomarkers for several diseases; nevertheless, the degree of technical and biological variability is not yet adequately characterized. Because pre-analytical factors such as blood collection time and EV subpopulation could confound biomarker studies, we performed a pilot study systematically quantifying methodological and biological variability including EV-Array (surface proteins), and proteome characterization of cargo. Plasma samples from six healthy adults were collected at two time points (morning and afternoon) and plasma was analyzed with EV-Array, and isolated EVs were analyzed using nanoparticle tracking analysis (NTA), and label-free mass spectrometry (LC-MS/MS). Methodological repeatability was high for NTA particle size (3.3% CV) and LC-MS (8.2% CV), and lower for EV-Array surface markers (22.6% CV). Variations between samples were reasonable for NTA-size, EV-Array and LC-MS/MS (5–21%) and substantially lower than between-subject variation. No evidence of systemic morning–afternoon shifts in particle size and concentration or EV cargo was observed, although small effects cannot be excluded. The same was true for most surface markers, but minor but statistically significant reductions in a few specific surface markers occurred in afternoon EV-Array samples. In this pilot we therefore do not observe any major systemic diurnal bias in healthy individuals in samples collected a.m. vs. p.m. Despite the small sample size, this study underscores the importance of accounting for individual variability and methodological standardization when designing EV-based biomarker research. Full article

Efficient irrigation planning in data-scarce regions remains challenging due to limited access to localized meteorological data, reliance on complex computer-based models, and the technical knowledge required to deploy them at the field scale. Hence, the need for accessible, smartphone-based tools that simplify soil water balance calculations using public data to support practical decision-making in resource-limited contexts. This smartphone-based application estimates Net and Gross Irrigation Requirements using a Soil Water Balance (SWB) framework. The app combines region-specific empirical formulations for Effective Rainfall (Pe) calculation. The application utilizes user-supplied crop and irrigation parameters and meteorological data available in the public domain and operates at multiple temporal scales (daily, 10-day, weekly, and monthly), thereby supporting flexible irrigation schedules. The performance of app was evaluated through simulation-based benchmarking against FAO-CROPWAT 8.0 using harmonized inputs across five representatives agro-climatic region: Central India, Southern Vietnam, Northern Thailand, Western Bangladesh, and Central Sri Lanka. Quantitative comparison showed deviations within ±5% for Effective Rainfall, crop evapotranspiration, Net Irrigation, and Gross Irrigation, and low mean bias values (−2.8% to +3.3%) show the absence of systematic over- or under-estimation compared to CROPWAT model. The application also demonstrated responsiveness to climatic variability. Although the validation is limited to few representative locations and assumed minimal runoff conditions, the results suggest that the proposed method is technically consistent and feasible in practice. This study demonstrates smartphone-based application as a decision support for field-level irrigation planning and water resource management, particularly in data-limited agricultural contexts. Full article

The purpose of this study was to examine morphological asymmetries in male youth judokas using an integrated assessment combining three-dimensional (3D) body scanning and bioelectrical impedance analysis (BIA), and to determine how these asymmetries relate to judo-specific performance. Twenty-seven competitive male youth judokas were evaluated for bilateral girth, segmental length, and lean mass asymmetries across upper- and lower-limb segments. The Absolute Asymmetry index, expressed as a percentage for individual body segments, and the average body symmetry across all variables were calculated, and associations with performance were assessed using the Special Judo Fitness Test (SJFT). Significant right-dominant asymmetries were found in elbow girth p < 0.001, forearm girth p < 0.001, thigh girth p = 0.028, and leg muscle mass p = 0.008. Upper-limb asymmetries were the primary contributors to total-body asymmetry, reflecting the unilateral gripping and rotational demands typical in judo. Only calf girth asymmetry was significantly associated with SJFT performance, with greater asymmetry linked to poorer outcomes, indicating a specific rather than general asymmetry–performance relationship (r = 0.405; p = 0.037). These findings underscore the importance of early detection of segment-specific asymmetries and suggest that rapid digital anthropometry is a practical tool for monitoring morphological development in youth judokas. Early targeted interventions may support balanced technical execution, enhance performance, and reduce the risk of uneven loading patterns as athletes progress to higher age categories and competition levels. Full article

Large language models (LLMs) and other foundation models are rapidly being woven into enterprise analytics workflows, where they assist with data exploration, forecasting, decision support, and automation. These systems can feel like powerful new teammates: creative, scalable, and tireless. Yet they also introduce distinctive risks related to opacity, brittleness, bias, and misalignment with organizational goals. Existing work on AI ethics, alignment, and governance provides valuable principles and technical safeguards, but enterprises still lack practical frameworks that connect these ideas to the specific metrics, controls, and workflows by which analytics teams design, deploy, and monitor LLM-powered systems. This paper proposes a conceptual governance framework for enterprise AI and analytics that is explicitly centered on LLMs embedded in analytics pipelines. The framework adopts a three-layered perspective—model and data alignment, system and workflow alignment, and ecosystem and governance alignment—that links technical properties of models to enterprise analytics practices, performance indicators, and oversight mechanisms. In practical terms, the framework shows how model and workflow choices translate into concrete metrics and inform real deployment, monitoring, and scaling decisions for LLM-powered analytics. We also illustrate how this framework can guide the design of controls for metrics, monitoring, human-in-the-loop structures, and incident response in LLM-driven analytics. The paper concludes with implications for analytics leaders and governance teams seeking to operationalize responsible, scalable use of LLMs in enterprise settings. Full article

The use of machine learning (ML) and deep learning (DL) in lung cancer detection and classification offers great promise for improving early diagnosis and reducing death rates. Despite major advances in research, there is still a significant gap between successful model development and clinical use. This review identifies the main obstacles preventing ML/DL tools from being adopted in real healthcare settings and suggests practical advice to tackle them. Using PRISMA guidelines, we examined over 100 studies published between 2022 and 2024, focusing on technical accuracy, clinical relevance, and ethical aspects. Most of the reviewed studies rely on computed tomography (CT) imaging, reflecting its dominant role in current lung cancer screening workflows. While many models achieve high performance on public datasets (e.g., >95% sensitivity on LUNA16), they often perform poorly on real clinical data due to issues like domain shift and bias, especially toward underrepresented groups. Promising solutions include federated learning for data privacy, synthetic data to support rare subtypes, and explainable AI to build trust. We also present a checklist to guide the development of clinically applicable tools, emphasizing generalizability, transparency, and workflow integration. The study recommends early collaboration between developers, clinicians, and policymakers to ensure practical adoption. Ultimately, for ML/DL solutions to gain clinical acceptance, they must be designed with healthcare professionals from the beginning. Full article

Artificial intelligence (AI) is becoming more and more integrated into manufacturing processes, revolutionizing conventional production, like CNC (Computer Numerical Control) machining. This study analyzes how large language models (LLMs), exemplified by ChatGPT, behave when tasked with G-code optimization for improving surface quality and productivity of automotive metal parts, with emphasis on systematically documenting failure modes and limitations that emerge when general-purpose AI encounters specialized manufacturing domains. Even if software programming remains essential for highly regulated sectors, free AI tools will be increasingly used due to advantages like cost-effectiveness, adaptability, and continuous innovation. The condition is that there is sufficient technical expertise available in-house. The experiment carried out involved milling three identical parts using a Haas VF-3 SS CNC machine. The G-code was generated by SolidCam and was optimized using ChatGPT considering user-specified criteria. The aim was to improve the quality of the part’s surface, as well as increase productivity. The measurements were performed using an ISR C-300 Portable Surface Roughness Tester and Axiom Too 3D measuring equipment. The experiment revealed that while AI-generated code achieved a 37% reduction in cycle time (from 2.39 to 1.45 min) and significantly improved surface roughness (Ra—arithmetic mean deviation of the evaluated profile—decreased from 0.68 µm to 0.11 µm—an 84% improvement), it critically eliminated the pocket-milling operation, resulting in a non-conforming part. The AI optimization also removed essential safety features including tool length compensation (G43/H codes) and return-to-safe-position commands (G28), which required manual intervention to prevent tool breakage and part damage. Critical analysis revealed that ChatGPT failures stemmed from three factors: (1) token-minimization bias in LLM training leading to removal of the longest code block (31% of total code), (2) lack of semantic understanding of machining geometry, and (3) absence of manufacturing safety constraints in the AI model. This study demonstrates that current free AI tools like ChatGPT can identify optimization opportunities but lack the contextual understanding and manufacturing safety protocols necessary for autonomous CNC programming in production environments, highlighting both the potential, but also the limitation, of free AI software for CNC programming. Full article

While face recognition systems are increasingly deployed in critical domains, they remain vulnerable to presentation attacks and exhibit significant demographic bias, particularly affecting African populations. This paper presents a fairness-aware Presentation Attack Detection (PAD) system using Local Binary Patterns (LBPs) with novel ethnicity-aware processing techniques specifically designed for African contexts. Our approach introduces three key technical innovations: (1) adaptive preprocessing with differentiated Contrast-Limited Adaptive Histogram Equalization (CLAHE) parameters and gamma correction optimized for different skin tones, (2) group-specific decision threshold optimization using Equal Error Rate (EER) minimization for each ethnic group, and (3) three novel statistical methods for PAD fairness evaluation such as Coefficient of Variation analysis, McNemar’s significance testing, and bootstrap confidence intervals representing the first application of these techniques in Presentation Attack Detection. Comprehensive evaluation on the Chinese Academy of Sciences Institute of Automation-SURF Cross-ethnicity Face Anti-spoofing dataset (CASIA-SURF CeFA) dataset demonstrates significant bias reduction achievements: a 75.6% reduction in the accuracy gap between African and East Asian subjects (from 3.07% to 0.75%), elimination of statistically significant bias across all ethnic group comparisons, and strong overall performance, with 95.12% accuracy and 98.55% AUC. Our work establishes a comprehensive methodology for measuring and mitigating demographic bias in PAD systems while maintaining security effectiveness, contributing both technical innovations and statistical frameworks for inclusive biometric security research. Full article

This systematic review focused on the validity of markerless motion capture (MMC) systems used for human movement assessment during tasks that involve physical interaction with objects. Five electronic databases were searched until May 2025. Eligible studies (i) assessed the validity of an MMC system, (ii) required human participants to perform tasks that involved physical interaction with objects (e.g., lifts, carrying, gait with loads), (iii) employed a marker-based reference system, and (iv) reported at least one kinematic metric. Risk of bias was assessed using the SURE checklist. A best-evidence synthesis was conducted to classify the level of evidence across included studies. Fifteen studies met eligibility (median = 21 participants per study). In general, MMC systems presented good performance in capturing the waveforms related to movement (i.e., high associations with reference systems), but its level of precision (i.e., the magnitude of differences to the reference systems) still requires improvement regarding tasks involving human–object interactions. Most tasks analyzed were lifts, gait with load, squatting and reaching/manipulation, and technical gestures. There was strong evidence for the validity of MMC for implementation during lifting tasks. In summary, markerless motion capture (MMC) systems exhibit promising evidence of validity for some human–object interaction tasks, that is, especially when lifting as strong evidence was observed across studies on this type of task. In contrast, some evidence for tasks including gait under load, squatting, reaching, or touchscreen interaction is limited, moderate, or conflicting. Notwithstanding these limitations, most studies were observed to have moderate- to high-quality methodology. Additional research is required to optimize protocols to study the measurement error aspects of MMC under human–object interaction in real-world environments. Full article

Background: Chronic total occlusion (CTO) percutaneous coronary intervention (PCI) remains among the most complex procedures in interventional cardiology, with variable technical success and heterogeneous long-term outcomes. Conventional angiographic scores such as J-CTO and PROGRESS-CTO provide only modest predictive accuracy and neglect critical patient and operator-related factors. Artificial intelligence (AI) and machine learning (ML) have emerged as transformative tools, capable of integrating multimodal data and offering enhanced diagnostic, procedural, and prognostic insights. Methods: We performed a structured narrative review of the literature between January 2010 and September 2025 using PubMed, Scopus, and Web of Science. Eligible studies were peer-reviewed original research, reviews, or meta-analyses addressing AI/ML applications in CTO PCI across imaging, procedural planning, and prognostic modeling. A total of 330 records were screened, and 33 studies met the inclusion criteria for qualitative synthesis. Results: AI applications in diagnostic imaging achieved high accuracy, with deep learning on coronary CT angiography yielding AUCs up to 0.87 for CTO detection, and IVUS/OCT segmentation demonstrating reproducibility > 95% compared with expert analysis. In procedural prediction, ML algorithms (XGBoost, LightGBM, CatBoost) outperformed traditional scores, achieving AUCs of 0.73–0.82 versus 0.62–0.70 for J-CTO/PROGRESS-CTO. Prognostic models, particularly CatBoost and neural networks, achieved AUCs of 0.83–0.84 for 5-year mortality in large registries (n ≈ 3200), surpassing regression-based methods. Importantly, comorbidities and functional status emerged as stronger predictors than procedural strategy. Future Directions: AI integration holds promise for real-time guidance in the catheterization laboratory, robotics-assisted PCI, federated learning to overcome data privacy barriers, and multimodality fusion incorporating imaging, clinical, and patient-reported outcomes. However, clinical adoption requires prospective multicenter validation, harmonization of endpoints, bias mitigation, and regulatory oversight. Conclusions: AI represents a paradigm shift in CTO PCI, providing superior accuracy over conventional risk models and enabling patient-centered risk prediction. With continued advances in federated learning, multimodality integration, and explainable AI, translation from research to routine practice appears within reach. Full article

The cryptocurrency landscape underwent significant changes in 2024 with the regulatory approval of spot Bitcoin ETFs, opening the market to institutional investors and millions of new clients. As Bitcoin reached new price peaks, the market attracted many retail traders using speculative approaches, evidenced by the surge in meme coins at the end of 2024. In such an environment, properly examining trading strategies can offer substantial advantages over the majority of market participants. Many traders, however, fail to test their strategies adequately, limiting evaluations to selected time periods and risking overfitting. This paper introduces the Rolling Strategy–Hold Ratio (RSHR), which uses a rolling-window approach to evaluate how strategies would perform from thousands of different starting points. This method helps mitigate recency bias and provides a more comprehensive understanding of strategy performance across diverse market conditions and cycles. By comparing strategies results against buy-and-hold results, traders can make informed decisions about whether to refine their strategies further or opt for index-based investing or alternative analytical methods. This study demonstrates the RSHR’s applications across technical, on-chain, sentiment analysis, and dollar cost averaging strategies, with initial research suggesting potential applications in traditional markets. Full article

This study aimed to investigate adhesive techniques applied to MIH-affected teeth by analyzing the one-year failure rate of restorations performed using total-etch and self-etch adhesive methods. This systematic review was conducted in accordance with the PRISMA 2020 statement. Eligibility criteria were defined using the PICO acronym. In vivo studies published from 2017 onward were evaluated. Two independent reviewers conducted the search on PubMed, Scopus, Cochrane, and Web of Science. The risk of bias was assessed with RoB2 and the Newcastle-Ottawa Scale. Statistical analysis was performed using the Open Meta [Analyst] software based on the absolute risk of failure. Results were presented as a pooled estimate with a 95% confidence interval (CI) and visualized in forest plots. Four RCTs and one retrospective cohort study were selected for the analysis. Data collected included information such as authors, study design, age, restorations, degree of hypomineralization, protocol, and follow-up. The meta-analysis showed no statistically significant differences between the techniques (p = 0.338) on MIH-affected teeth. This meta-analysis supports the use of both adhesive techniques for managing MIH teeth, emphasizing the need for further studies to examine the specific clinical and technical conditions under which each technique might be more advantageous. Full article

Objectives: In patients with autosomal dominant polycystic kidney disease (ADPKD), total kidney volume (TKV) is the gold standard biomarker for assessing the risk of progression and the need for drug therapy. However, it is a time-consuming process. In this systematic review and meta-analysis, we evaluate the current state of deep learning (DL) algorithms for automatic kidney volume segmentation. Methods: All original research, including the search terms ADPKD, diagnostic imaging, DL, and TKV, was identified in PubMed, Embase, and Ovid MEDLINE databases from January 2000 to 13 October 2024. Articles with insufficient information to assess methodological quality were excluded. Quality was assessed using the “Quality Assessment of Diagnostic Accuracy Studies, Version 2” (QUADAS-2) and Checklist for Artificial Intelligence in Medical Imaging (CLAIM) tools. We focused on the Dice Similarity Coefficient (DSC), bias differences, and time efficiency as outcomes. Results: Nineteen studies were included, with an overall low risk of bias; however, the mean adherence to the CLAIM checklist was 64%. The pooled DSC under the random-effects model was $0.953$ (95% CI: $0.938–0.969$) with relatively low bias for TKV in 5622 ADPKD patients (mean age, 46.1 years; 45% male) and 9180 scans (79% MRI). The average segmentation time was decreased by 75% compared to the ground truth. Performance differences were evident among imaging modalities, MRI sequences, and 3D vs. 2D models, but not among imaging planes. The between-study heterogeneity was low ($I^2=0\%$), and no statistically significant evidence of small-study effects or publication bias was detected. Conclusions: DL models for TKV in ADPKD patients demonstrated high precision compared to manual segmentation in a large, pooled sample with heterogeneous study designs and methods. While clinical implementation is not yet feasible, the current work demonstrates the technical and diagnostic efficacy of image-based DL segmentation models. Full article

Background: Multiple primary lung cancer (MPLC) presents clinical challenges due to its biological complexity. While lobectomy remains standard, limited resection and localized ablation offer comparable efficacy. This systematic review evaluates the safety and efficacy of combining surgical and ablative therapies for MPLC. Methods: A comprehensive search of PubMed, Embase, and Web of Science (January 2000–2025) identified studies involving MPLC patients treated with both surgery and ablation, either concurrently or sequentially. Data on ablation efficacy, adverse events, and prognosis were extracted. A meta-analysis was performed when data pooling was appropriate. The methodological quality and risk of bias of the included studies were assessed using the MINORS and ROBINS-I tools. Publication bias was evaluated through funnel plots and Egger’s linear regression test. Furthermore, one case report on combination therapy was also included. Results: A total of nine studies met the inclusion criteria and were included in the final analysis. All reported a 100% technical success rate for ablation, efficacy rates exceeding 70%, and adverse event rates ranging from 5.0% to 26.7%. Due to significant heterogeneity among studies, a random-effects model was applied. The meta-analysis yielded a pooled ablation efficacy rate of 97.11% (95% CI: 85.81–100.00%) and a pooled adverse event rate of 14.23% (95% CI: 8.07–20.38%), indicating favorable safety and efficacy of the combined therapy. Conclusions: The integration of surgical and ablative therapies offers a safe and effective strategy for managing MPLC and supports a potential paradigm shift from single-modality treatment toward a more personalized, organ-preserving, and patient-centered approach. Full article

Background: Chronic total occlusion percutaneous coronary intervention (CTO PCI) is a complex revascularization procedure requiring advanced techniques to ensure procedural success and safety. Hybrid strategies combining antegrade dissection/re-entry (ADR) and retrograde approaches have become increasingly adopted in contemporary practice. Objectives: To systematically review and synthesize evidence comparing outcomes of ADR and retrograde CTO PCI techniques, with pooled estimates of success rates and adverse events. Methods: This review followed PRISMA 2020 guidelines. We searched PubMed, Cochrane CENTRAL, and Google Scholar for studies published between January 2015 and June 2025. Eligible studies included randomized controlled trials and observational studies reporting outcomes of ADR and/or retrograde CTO PCI. Data extraction was performed by two independent reviewers. Risk of bias was assessed using the Newcastle–Ottawa Scale and the Cochrane RoB 2.0 tool. A random-effects meta-analysis was conducted for consistently reported outcomes. Results: Twenty studies encompassing over 87,000 CTO PCI procedures were included. Pooled analysis of 16 studies demonstrated a technical success rate of 83.4% and a procedural success rate of 84.6%. The in-hospital major adverse cardiac event (MACE) rate was 3.3%. Hybrid strategies integrating ADR and retrograde approaches yielded the highest success rates (86–91%) with acceptable safety profiles. Use of adjunctive tools such as IVUS, dual arterial access, and re-entry devices was associated with improved outcomes. Discussion: Hybrid CTO PCI techniques are safe, effective, and reproducible across diverse clinical settings. When performed by experienced operators using modern adjuncts, these strategies provide durable benefits and should be considered standard for complex occlusions. Limitations include variation in study quality, heterogeneous procedural definitions, and lack of long-term data in several cohorts. Full article