Search Results (88)

The justice system has increasingly applied AI techniques for legal judgment to enhance efficiency. However, most AI techniques focus on decision-making outcomes, failing to capture the deliberative nature of the real-world judicial process. To address these challenges, we propose a large language model-based multi-agent framework named AgentsBench. Our approach leverages multiple LLM-driven agents that simulate the discussion process of the Chinese judicial bench, which is often composed of professional and lay judge agents. We conducted experiments on a legal judgment prediction task, and the results show that our framework outperforms existing LLM-based methods in terms of performance and decision quality. By incorporating these elements, our framework reflects real-world judicial processes more closely, enhancing accuracy, fairness, and societal consideration. While the simulation is based on China’s lay judge system, our framework is generalizable and can be adapted to various legal scenarios and other legal systems involving collective decision-making processes. Full article

Legal actors often discuss emotion-based decisions and reasoned evaluation of the facts as distinct and opposite methods through which jurors can reach conclusions. However, research suggests that emotion can have an indirect effect on juror decisions by changing the way that jurors evaluate the facts of the case. In three studies (N = 713, N = 677, N = 651), we tested whether mock jurors’ negative moral emotions towards the defendant predicted their evaluations of unrelated case evidence and in turn their case judgments and whether judicial rehabilitation could reduce this effect. Participants read a civil case and were randomly assigned to either receive judicial rehabilitation or not. Then, they completed measures relating to their negative moral emotions towards the defendant, their agreement with plaintiff and defense evidence, and case judgments. When participants reported increased negative emotions towards the defendant, they agreed more with unrelated plaintiff evidence and less with unrelated defense evidence. In turn, they voted liable more often and awarded more in damages. Judicial rehabilitation did not reduce this effect. This research provides support for the idea that there is a more complicated relationship between emotion and decisions than legal actors suggest. Specifically, negative emotions towards the defendant are associated with a pro-plaintiff evaluation of evidence and pro-plaintiff judgments. Full article

Fluid management is a critical component in the treatment of patients suffering with major burns. Clinicians must carefully balance judicious resuscitation with the risks of over- or under-resuscitation. We aimed to identify factors associated with survival in burns patients and determine the importance of resuscitation practices. Patients requiring admission to Burns Services in the United Kingdom between 1 April 2022 and 31 March 2023 were included in the National Burns Audit project on fluid resuscitation practices, to evaluate factors associated with survival and Critical Care Length of Stay (CCLoS). A total of 198 patients were included in the analyses, with median age of 51 years (interquartile range, (IQR) 35–62 years), median Total Burn Surface Area (TBSA%) of 27.5% (IQR 20–40%), and median Baux score 82.5 (IQR 66–105). The following were found to be significant for survival: younger age, smaller TBSA%, lower Baux score and independence from renal replacement therapy. Neither the mechanism of burns nor the fluid resuscitation volumes appeared to influence survival. Although interventions such as tracheostomy or the number of surgical procedures did not appear to affect survival, fluid replacement of more than 6 mL/kg/%TBSA independently predicted longer CCLoS. Volume of fluid resuscitation, within the limits examined in this cohort, did not impact likelihood of survival. Full article

Background: Chronic kidney disease (CKD) and its advanced stage, end-stage renal disease (ESRD), affect millions worldwide and are associated with a paradoxical hemostatic imbalance—marked by both increased thrombotic and bleeding risks—which complicates anticoagulant use and demands clearer, evidence-based clinical guidance. Design: This study is a critical review synthesizing the current literature on anticoagulant therapy in CKD and ESRD, with emphasis on altered pharmacokinetics, clinical complications, and therapeutic adjustments. Data Sources: PubMed, Scopus, and Google Scholar were searched for articles discussing anticoagulation in CKD/ESRD, focusing on pharmacokinetics, clinical outcomes, and dosing recommendations. Study Selection: Studies examining the safety, efficacy, and pharmacokinetics of anticoagulants—including heparin, low-molecular-weight heparin (LMWH), warfarin, and direct oral anticoagulants (DOACs)—in CKD and ESRD populations were included. Data Extraction and Synthesis: Key findings were summarized to highlight the dose modifications, therapeutic considerations, and clinical challenges in managing anticoagulation in CKD/patients with ESRD. Emphasis was placed on balancing thrombotic and bleeding risks and identifying gaps in existing guidelines. Results: Patients with CKD and ESRD exhibit a paradoxical hypercoagulable state marked by platelet dysfunction, altered coagulation factors, and vascular endothelial damage. This condition increases the risk of thrombotic events, such as deep vein thrombosis (DVT) and pulmonary embolism (PE), while simultaneously elevating bleeding risks. Hemodialysis and CKD-associated variables further complicate the management of coagulation. Among anticoagulants, unfractionated heparin (UFH) is preferred due to its short half-life and adjustability based on activated partial thromboplastin time (aPTT). Low-molecular-weight heparins (LMWHs) offer predictable pharmacokinetics but require dose adjustments in CKD stages 4 and 5 due to reduced clearance. Warfarin necessitates careful dosing based on the estimated glomerular filtration rate (eGFR) to maintain an international normalized ratio (INR) ≤ 4, minimizing bleeding risks. Direct oral anticoagulants (DOACs), particularly Apixaban, are recommended for patients with eGFR < 15 mL/min or those on dialysis, although data on other DOACs in CKD remain limited. The lack of comprehensive guidelines for anticoagulant use in CKD and ESRD highlights the need for individualized, patient-centered approaches that account for comorbidities, genetics, and clinical context. Conclusions: Managing anticoagulation in CKD/ESRD is challenging due to complex coagulation profiles and altered pharmacokinetics. Judicious dosing, close monitoring, and patient-centered care are critical. High-quality randomized controlled trials are needed to establish clear guidelines and optimize therapy for this vulnerable population. Full article

Background/Objectives: Urinary tract infections (UTIs) are frequently diagnosed empirically, often leading to overtreatment and rising antimicrobial resistance. This study aimed to develop and evaluate machine learning (ML) models that predict urine culture outcomes using routine urinalysis and demographic data, supporting more targeted empirical antibiotic use. Methods: A real-world dataset comprising 8065 urinalysis records from a hospital laboratory was used to train five ensemble ML models, including random forest, XGBoost (eXtreme gradient boosting), extra trees, voting classifier, and stacking classifier. Models were developed using 10-fold stratified cross-validation and assessed via clinically relevant metrics including specificity, sensitivity, likelihood ratios, and diagnostic odds ratios (DORs). To enhance screening utility, threshold optimization was applied to the best-performing model (XGBoost) using the Youden index. Results: XGBoost and random forest demonstrated the most balanced diagnostic profiles (AUROC: 0.819 and 0.791, respectively), with DORs exceeding 21. The voting and stacking classifiers achieved the highest specificity (>95%) and positive likelihood ratios (>10) but exhibited lower sensitivity. Feature importance analysis identified positive nitrites, white blood cell count, and specific gravity as key predictors. Threshold tuning of XGBoost improved sensitivity from 70.2% to 87.9% and reduced false negatives by 82%, with an associated NPV of 96.4%. The adjusted model reduced overtreatment by 56% compared to empirical prescribing. Conclusions: ML models based on structured urinalysis and demographic data can support clinical decision-making for UTIs. While high-specificity models may reduce unnecessary antibiotic use, sensitivity trade-offs must be considered. Threshold-optimized XGBoost offers a clinically adaptable tool for empirical treatment decisions by improving sensitivity and reducing overtreatment, thus supporting the more personalized and judicious use of antibiotics. Full article

The phenomenal rise of artificial intelligence (AI) in the last decade, and its evolution as a versatile addition to various fields, necessitates its usage for novel purposes in multidimensional fields like the manufacturing industry. Even though AI has been rigorously studied for process optimization, wastage reduction, and other quintessential aspects of the manufacturing industry, there has been limited focus on worker safety as a theme in the current literature. Safety standards contribute to worker safety, but there is no one-size-fits-all approach in these standards or policies, which warrants evaluation and integration of new ideas and technologies to reach the closest to ideal standards. This includes but is not limited to health, regulation of operations, predictive maintenance, and automation and control. The rise of Industry 4.0 and the migration towards Industry 5.0 facilitate easy integration of advanced technologies like AI into the manufacturing industry with real-time predictive capabilities, and this can help reduce human errors and mitigate hazards in processes where sensitivity is crucial or hazards are frequent. Keeping the future outlook in focus, AI can contribute to training workers in risk-free environments, promote engineering education for easy adaptation to new technology, and reduce resistance to changes in the industry. Furthermore, there is an urgent need for standards and regulations to govern and integrate AI technologies judiciously into the manufacturing industry, which holds AI models and their creators accountable for their decisions. This could further extend to preventing the adversarial use of new technology. This study exhaustively discusses the potential and ongoing contributions of this technology to the safety of workers in the manufacturing industry. Full article

The Integrated Valuation of Ecosystem Services and Trade-Offs (InVEST) model with the distributed hydrological model Soil and Water Assessment Tool (SWAT) were implemented. The SWAT model quantifies and visualizes water production and groundwater reserves in the Mudanjiang River Basin, employing the groundwater runoff modulus method to calculate groundwater recharge in the basin. This study aims to assess the model’s applicability in cold basins and subsequently analyze groundwater distribution characteristics, water reserves, and the exploitable volume. It serves as a reference for the judicious allocation of groundwater resources and the preservation of the local aquatic ecosystem. The study indicates the following: (1) Utilizing the monthly runoff data from the Mudan River hydrologic station, SWAT simulation and calibration were conducted, yielding a determination coefficient (R²) of 0.75 and a Nash–Sutcliffe efficiency coefficient (NS) of 0.77, thereby satisfying fundamental scientific research criteria. The water yield predicted by the InVEST model aligns closely with the water resources bulletin of the research region. (2) The data from the water production module of the InVEST model indicate that the average annual water production during the research period was 6.725 billion m³, with an average annual water production depth of 148 mm. In 2018, characterized by ample water supply, the water output was at its peak, with a depth of 242 mm. In 2014, the water depth recorded was merely 16 mm. (3) Throughout the study period, the average annual flow of the Mudan River was 4.2 billion m³, whereas the groundwater reserve was 24.13 (10⁸ m³·a⁻¹). In 2013, the maximum groundwater reserve was 38.42 (10⁸ m³·a⁻¹), while the minimum reserve in 2014 was 2.36 (10⁸ m³·a⁻¹), suggesting that the region was predominantly experiencing sustainable exploitation. (4) The mean groundwater runoff modulus is 0.28 L/(s·km²), with a peak annual recharge of 15.4 (10⁸ m³·a⁻¹) in 2013 and a lowest recharge of just 3.2 (10⁸ m³·a⁻¹) in 2011. Full article

This study presents a novel estimator that combines the Kibria–Lukman and ridge estimators to address the challenges of multicollinearity in Conway–Maxwell–Poisson (COMP) regression models. The Conventional COMP Maximum Likelihood Estimator (CMLE) is notably susceptible to the adverse effects of multicollinearity, underscoring the necessity for alternative estimation strategies. We comprehensively compare the proposed COMP Modified Kibria–Lukman estimator (CMKLE) against existing methodologies to mitigate multicollinearity effects. Through rigorous Monte Carlo simulations and real-world applications, our results demonstrate that the CMKLE exhibits superior resilience to multicollinearity while consistently achieving lower mean squared error (MSE) values. Additionally, our findings underscore the critical role of larger sample sizes in enhancing estimator performance, particularly in the presence of high multicollinearity and over-dispersion. Importantly, the CMKLE outperforms traditional estimators, including the CMLE, in predictive accuracy, reinforcing the imperative for judicious selection of estimation techniques in statistical modeling. Full article

This study posits that strategically optimizing irrigation and fertilization regimes can enhance the productivity and water use efficiency (WUE) of alfalfa (Medicago sativa L.), thereby mitigating the constraints imposed by soil impoverishment and water scarcity in forage production systems of arid and semi-arid regions. Conducted over two years, the outdoor pot experiment investigated the effects of water regulation during the branching and bud stages (each at 60–100% θ_0.85, where θ_0.85 = 0.85θfc) and different levels of nitrogen and phosphorus fertilization (0–280 kg/ha each) on alfalfa yield and WUE. Using Response Surface Methodology (RSM) with a Central Composite Design (CCD), we modeled the relationships between input variables and key response parameters: total yield, evapotranspiration (ET), and WUE. The response surface models exhibited high reliability, with coefficients of determination R², adjusted R², predicted R², and adequate precision exceeding 0.94, 0.90, 0.86, and 13.6, respectively. Sensitivity analysis indicated that water regulation during critical growth stages, particularly the branching stage, had the most significant impact on yield and ET, while nitrogen and phosphorus fertilization positively influenced WUE. Within the appropriate range of water management, judicious fertilization significantly enhanced alfalfa production performance, although excessive inputs resulted in diminishing returns. This study identified the optimal conditions for sustainable production: branching stage water regulation (82.26–83.12% θ_0.85) and bud stage water regulation (78.11–88.47% θ_0.85), along with nitrogen application (110.59–128.88 kg/ha) and phosphorus application (203.86–210 kg/ha). These findings provide practical guidelines for improving the sustainability and efficiency of alfalfa production in resource-limited environments. Full article

Anti-amyloid therapies (AATs) are increasingly being recognized as promising treatment options for Alzheimer’s disease (AD). Amyloid-related imaging abnormalities (ARIAs), small areas of edema and microbleeds in the brain presenting as abnormal signals in MRIs of the brain for patients with AD, are the most common side effects of AATs. While most ARIAs are asymptomatic, they can be associated with symptoms like nausea, headache, confusion, and gait instability and, less commonly, with more serious complications such as seizures and death. Cerebral amyloid angiopathy (CAA) has been found to be a major risk for ARIA development. The identification of sensitive and reliable non-invasive biomarkers for CAA has been an area of AD research over the years, but with the approval of AATs, this area has taken on a new urgency. This comprehensive review highlights several potential biomarkers, such as Aβ40, Aβ40/42, phosphorylated-tau217, neurofilament light chain, glial fibrillary acidic protein, secreted phosphoprotein 1, placental growth factor, triggering receptor expressed on myeloid cells 2, cluster of differentiation 163, proteomics, and microRNA. Identifying and staging CAA even before its consequences can be detected via neuroimaging are critical to allow clinicians to judiciously select appropriate candidates for AATs, stratify monitoring, properly manage therapeutic regimens for those experiencing symptomatic ARIAs, and optimize the treatment to achieve the best outcomes. Future studies can test potential plasma biomarkers in human beings and evaluate predictive values of individual markers for CAA severity. Full article

The application of mineral fertilizers can effectively enhance crop yields. However, this potential benefit may be diminished if the use of mineral fertilizers leads to a substantial decline in soil organic carbon (SOC) and an increase in soil greenhouse gas (GHG) emissions. This study aimed to determine the optimal fertilizer combinations and rates for improving SOC and maize yield while reducing GHG emissions in the semi-arid uplands of Kenya. Data were collected from five different fertilizer treatments (N50, N100, N150, N100+manure, and N100+straw) compared to a control (N0) in a long-term experimental field, which was used to run and validate the DNDC model before using it for long-term predictions. The results showed that the combination of mineral fertilizer and straw resulted in the highest SOC balance, followed by that of fertilizer and manure. All fertilized treatments had higher maize grain yields compared to low-fertilizer treatment (N50) and control (N0). Daily CO₂ fluxes were highest in the treatment combining mineral fertilizer and manure, whereas there were no significant differences in N₂O fluxes among the three tested treatments. The findings of this study indicate that the judicious application of mineral fertilizer, animal manure, and straw has great potential in enhancing SOC and maize yields while reducing GHG emissions, thereby providing practical farming management strategies in semi-arid Kenya. Full article

For the extraction of judicial events for Tibetan, a low-resource language, traditional simple neural network approaches struggle to adequately capture the deep semantics and features of the texts because Tibetan texts are usually lengthy and contain numerous judicial-related entities. To overcome this limitation, this research presents an event extraction model combining deep word representation with hybrid neural networks for the Tibetan judicial domain. The model introduces the Chinese minority pre-trained language model (CINO), which generates dynamic word vector representations, addressing the challenge of modeling the deep semantics inherent in Tibetan texts. During feature extraction, a bidirectional long short-term memory network (BiLSTM) is applied to extract the temporal and contextual dependencies, while a convolutional neural network (CNN) is utilized to capture the local semantic features to construct a comprehensive global semantic representation. Finally, the sequences are decoded through conditional random field (CRF) to generate optimal prediction results, thus achieving the efficient extraction of Tibetan judicial events. The experimental findings indicate that the model outperforms the baselines by achieving F1 scores of 70.47% for trigger detection and 62.99% for argument recognition, with improvements of 16.6% and 16.42%, respectively. These results confirm the effectiveness and superiority of the proposed model. Full article

Legal judgment prediction is a common basic task in the field of Legal AI, aimed at using deep domain models to predict the outcomes of judicial cases, such as charges, legal provisions, and other related tasks. This task has practical applications in environmental law, including legal decision assistance and legal advice, offering a promising and broad prospect. However, most previous studies focus on using high-quality labeled data for strong supervised training in criminal justice, often neglecting the rich external knowledge contained in various charges and laws. This approach fails to accurately simulate the decision-making steps of judges in real scenarios, overlooking the semantic information in case descriptions that significantly impacts judgment results, leading to biased outcomes. In judicial environmental protection, the high overlap and similarity between different charges can cause confusion, and there is a lack of relevant judicial decision labeling datasets. To address this, we propose the External Knowledge-Infused Cross Attention Network (EKICAN), which leverages the robust semantic understanding capabilities of large models. By extracting information such as fact descriptions and court opinions from documents of criminal, civil, and administrative cases related to judicial environmental protection, we construct the Judicial Environmental Law Judgment Dataset (JELJD). We address data imbalance in this dataset using the text generation capabilities of judicial large models. Finally, EKICAN fuses semantic information from different parts with external knowledge to output prediction results. Experimental results show that EKICAN achieves state-of-the-art performance on the JELJD compared to advanced models. Full article

Predictive justice, which involves forecasting trial outcomes, presents significant challenges due to the complex structure of legal judgments. To address this, it is essential to first identify all claims across different categories before attempting to predict any result. This paper focuses on a classification task based on the detection of Article 700 in judgments, which is a rule indicating whether the plaintiff or defendant is entitled to reimbursement of their legal costs. Our experiments show that conventional machine learning models trained on word and document frequencies can be competitive. However, using transformer models specialized in legal language, such as Judicial CamemBERT, also achieves high accuracies. Full article

This paper analyzes the impact of AI systems in the judicial domain, adopting an actor–network theory (ANT) framework and focusing on accountability issues emerging when such technologies are introduced. Considering three different types of AI applications used by judges, this paper explores how introducing non-accountable artifacts into justice systems influences the actor–network configuration and the distribution of accountability between humans and technology. The analysis discusses the actor–network reconfiguration emerging when speech-to-text, legal analytics, and predictive justice technologies are introduced in pre-existing settings and maps out the changes in agency and accountability between judges and AI applications. The EU legal framework and the EU AI Act provide the juridical framework against which the findings are assessed to check the fit of new technological systems with justice system requirements. The findings show the paradox that non-accountable AI can be used without endangering fundamental judicial values when judges can control the system’s outputs, evaluating its correspondence with the inputs. When this requirement is not met, the remedies provided by the EU AI Act fall short in costs or in organizational and technical complexity. The judge becomes the unique subject accountable for the use and outcome of a non-accountable system. This paper suggests that this occurs regardless of whether the technology is AI-based or not. The concrete risks emerging from these findings are that these technological innovations can lead to undue influence on judicial decision making and endanger the fair trial principle. Full article