Advancing Open Science

Natural antioxidants align with consumer demand for clean-label, sustainable, and health-promoting food solutions. Artificial intelligence (AI) is enabling deeper understanding, more rapid screening, and new application modalities in food systems. Novel deep learning frameworks have been developed to predict interactions between polyphenols and proteins—crucial for understanding how antioxidants affect nutrient bioavailability, therapeutic functions, and food processing behavior. The convergence of AI and natural antioxidants is forging a transformative frontier in food science. This review aims to focus on AI-enabled methods and advances in natural antioxidants, focusing on practical impact and future directions. PubMed, Web of Science, Scopus, and the Cochrane Library databases were searched for relevant articles published up to September 2025. AI accelerates the analysis, design, and personalization of food systems, while natural antioxidants deliver health-promoting, sustainable, and clean-label functionality. Together, they offer promising avenues for safer, fresher, and more nutritious food systems. Continued innovation, multidisciplinary synergy, and thoughtful regulation are essential to unlocking their full potential. Encapsulating essential oils, polyphenols, and curcumin within nanocarriers significantly improves their stability, antimicrobial efficacy, controlled release, and bioavailability, extending their shelf life and application in diverse food formats. Advancing the use of natural antioxidants in food systems must navigate additive classifications, health claim validations, labeling transparency, and regulatory compliance across regions.

Veterinary drugs are widely present in animal manure and manure-based fertilizers, making their safety for use as soil amendments still ambiguous. This study investigated the concentrations of 17 typical veterinary drugs in animal manure and manure-based fertilizers from Shandong Province using solid-phase extraction coupled with high-performance liquid chromatography–tandem mass spectrometry and assessed their environmental risks to soil organisms based on risk quotient values. The established method demonstrated robust performance, with drug recovery rates ranging from 72.9% to 109%. Tetracyclines were identified as the most prevalent contaminants, with mean concentrations of 1522 μg/kg in animal manure and 144 μg/kg in manure-based fertilizers. Drug concentrations in manure-based fertilizers were generally lower than those in animal manure. Livestock manure contained higher drug concentrations compared to poultry manure. Influenced by farming practices, drug concentrations were higher in beef cattle manure than in dairy cattle manure, and higher in broiler manure than in layer manure. Manure-based fertilizers primarily derived their drug content from chicken, cattle, and sheep manure. Tetracyclines in swine and sheep manure posed high risks to soil organisms, while those in beef cattle manure and dairy cattle manure posed medium risks. In contrast, most drugs in manure-based fertilizers exhibited low risks. Comprehensive analysis of both concentration levels and ecological risks indicates that manure-based fertilizers represent a more feasible option for soil amendment. This study provides a theoretical foundation for better understanding the feasibility of applying animal manure and manure-based fertilizers to agricultural land.

We investigate the fundamental trade-off between entropy and the Gini index within income distributions, employing a stochastic framework to expose deficiencies in conventional inequality metrics. Anchored in the principle of maximum entropy (ME), we position entropy as a key marker of societal robustness, while the Gini index, identical to the (second-order) K-spread coefficient, captures spread but neglects dynamics in distribution tails. We recommend supplanting Lorenz profiles with simpler graphs such as the odds and probability density functions, and a core set of numerical indicators (K-spread K₂/μ, standardized entropy Φμ, and upper and lower tail indices, ξ, ζ) for deeper diagnostics. This approach fuses ME into disparity evaluation, highlighting a path to harmonize fairness with structural endurance. Drawing from percentile records in the World Income Inequality Database from 1947 to 2023, we fit flexible models (Pareto–Burr–Feller, Dagum) and extract K-moments and tail indices. The results unveil a concave frontier: moderate Gini reductions have little effect on entropy, but aggressive equalization incurs steep stability costs. Country-level analyses (Argentina, Brazil, South Africa, Bulgaria) link entropy declines to political ruptures, positioning low entropy as a precursor to instability. On the other hand, analyses based on the core set of indicators for present-day geopolitical powers show that they are positioned in a high stability area.

Background: Post-thyroidectomy hypocalcemia is a common complication, yet the optimal perioperative markers for identifying high-risk patients and guiding supplementation remain debated. This study aimed to evaluate factors associated with hypocalcemia at 24 h after total thyroidectomy, identify independent predictors, and assess the reliability of early PTH measurement in determining supplementation needs. Methods: We conducted a single-center prospective cohort study including 200 patients undergoing total thyroidectomy at Genesis Hospital, Thessaloniki, between November 2022 and March 2025. PTH was measured preoperatively, 10 min post-resection, and at 24 and 72 h; calcium and phosphorus were measured preoperatively and postoperatively. Results: Independent predictors of hypocalcemia at 24 h were female sex, preoperative calcium, and PTH at 10 min. Age, pathology, incidental parathyroid excision, and extent of surgery were not significantly associated with hypocalcemia. ROC analysis showed that a preoperative calcium cutoff of 9.47 mg/dL yielded an AUC of 0.73, with 70.1% sensitivity and an NPV of 82%. PTH at 10 min with a cutoff of 24.6 pg/mL yielded an AUC of 0.66, with 70.1% sensitivity and an NPV of 79%. For supplementation needs, PTH at 10 min demonstrated excellent discrimination, with a cutoff of 16.3 pg/mL at 24 h and 14.1 pg/mL at 72 h. Conclusions: Preoperative calcium and PTH measured 10 min after thyroid removal are useful markers for predicting hypocalcemia after total thyroidectomy, with early PTH also accurately identifying supplementation needs.

This paper proposes and validates a method for assessing the resilience of cyber–physical microgrids integrating Photovoltaic (PV) generation and Battery Energy Storage Systems (BESS). The approach combines two operational performance indicators—Voltage Deviation Index (VDI) and Energy Not Supplied (ENS)—with a composite resilience index that captures recovery dynamics following physical and cyber disturbances. The method is implemented in MATLAB Simulink R2022b on the IEEE 33-bus feeder, with PV at bus 6 and a BESS at bus 18. Two stress scenarios are analyzed: (i) loss of the main supply at bus 2 and (ii) a cyber-induced communication failure that triggers local (fallback) operation. Compared with the base case, the proposed strategy reduces VDI by approximately 27% and ENS by 12%, demonstrating significantly improved resilience without noticeable performance penalties.

Background/Objectives: Multidisciplinary care is the gold-standard approach for delivering comprehensive pediatric healthcare. For children undergoing cochlear implant (CI) evaluation, multiple appointments are required to assess candidacy, set realistic expectations, and counsel families on rehabilitation and the psychosocial impact of hearing loss. Established pediatric CI users also need coordinated follow-up to address ongoing auditory, educational, and psychosocial needs. This study evaluated the satisfaction and family perspectives of the implementation of a virtual, team-based multidisciplinary model for both CI candidates and established CI users. Methods: Thirty-nine children and their families participated in discipline-specific telehealth consultations, including audiology, listening and spoken language (LSL) therapy, psychology, and educational services, followed by a 60 min multidisciplinary team meeting. Team meetings occurred during pre-implantation and at six months post-activation for CI candidates. Team meetings for established CI users were scheduled following completion of individual consultations. Providers summarized findings from their individual visits before transitioning to a caregiver-led discussion. Post-visit surveys assessed satisfaction and perceived benefit from the multidisciplinary model. Results: Thirty-nine dyads were enrolled (11 Pre-CI; 28 Established CI). Caregivers were predominantly mothers (89.7%), most identified as Hispanic (55.3%) and White (71.1%). Over half of children identified as Hispanic (59%) and White (71.8%); most were diagnosed with hearing loss at birth (55.9%). Satisfaction with the virtual model was uniformly high: 100% of caregivers were satisfied or very satisfied, and most rated care quality as “very good” or “excellent.” LSL therapy was most frequently rated as the most beneficial visit (70% Pre-CI; 45% Established CI). Caregivers strongly preferred ongoing team-based care, with 55–80% reporting that they would like it to occur every six months and 95–100% preferring remote meetings. Conclusions: A virtual multidisciplinary model offers a high-quality, family-centered approach for both CI evaluations and ongoing management of established CI users. By integrating simultaneous team-based sessions, this model not only supports the ‘whole child’ but also strengthens the family system by improving communication, streamlining care, and reducing the burden of multiple in-person appointments. Families consistently report high levels of satisfaction with the convenience, clarity, and collaboration provided through virtual team visits. Incorporating routine check-ins with families is essential to ensure their needs are addressed, reinforce progress, and guide timely, targeted interventions that maximize each child’s developmental outcomes.

In gas turbine fire-resistant oil systems, valve actuations induce transient pressure fluctuations and the water hammer effect, causing pressure oscillations and structural vibrations. This study uses a coupled CFD and transient structural simulation to analyze the effects of different valve strategies on pressure wave propagation and structural response. Results show that a higher valve opening rate leads to a more significant water hammer effect, increasing structural deformation and stress. The maximum equivalent stress was verified at 201.9 MPa, maintaining a 30% safety margin and meeting American Society of Mechanical Engineers (ASME) B31.3 requirements. Finally, a “slow-fast-slow” (S-shaped) valve strategy is proposed to significantly improve the system’s pressure response characteristics, providing theoretical and engineering guidance for safe operation.

To support novice learners, the Java programming learning assistant system (JPLAS) has been developed with various features. Among them, code writing problem (CWP) assigns writing an answer code that passes a given test code. The correctness of an answer code is validated by running it on JUnit. In previous works, we implemented a code plagiarism checking function that calculates the similarity score for each pair of answer codes based on the Levenshtein distance. When the score is higher than a given threshold, this pair is regarded as plagiarism. However, a method for finding the proper threshold has not been studied. In addition, AI-generated codes have become threats in plagiarism, as AI has grown in popularity, which should be investigated. In this paper, we propose a threshold selection method based on Tukey’s IQR fences. It uses a custom upper threshold derived from the statistical distribution of similarity scores for each assignment. To better accommodate skewed similarity distributions, the method introduces a simple percentile-based adjustment for determining the upper threshold. We also design prompts to generate answer codes using generative AI and apply them to four AI models. For evaluation, we used a total of 745 source codes of two datasets. The first dataset consists of 420 answer codes across 12 CWP instances from 35 first-year undergraduate students in the State Polytechnic of Malang, Indonesia (POLINEMA). The second dataset includes 325 answer codes across five CWP assignments from 65 third-year undergraduate students at Okayama University, Japan. The applications of our proposals found the following: (1) any pair of student codes whose score is higher than the selected threshold has some evidence of plagiarism, (2) some student codes have a higher similarity than the threshold with AI-generated codes, indicating the use of generative AI, and (3) multiple AI models can generate code that resembles student-written code, despite adopting different implementations. The validity of our proposal is confirmed.