BACKGROUNDAdvanced pancreatic cancer (PC) is associated with a poor prognosis. The integration of Chinese and Western medicine (ICWM) has shown promising clinical efficacy. Nonetheless, the existing body of research assessing the efficacy and safety of this integrative approach is limited, hindering the provision of robust evidence-based support for clinical decision-making.AIMTo assess the short-term and long-term efficacy and safety of ICWM compared with Western medicine (WM) as a standalone treatment for advanced PC.METHODSWe enrolled 136 patients with advanced PC admitted to Henan Provincial Hospital of Traditional Chinese Medicine and Henan Provincial People’s Hospital from 2019 to 2024. Patients were randomly assigned to the ICWM or WM group (n = 66 or n = 70, respectively) according to treatment modality. The long-term efficacy was evaluated using survival analyses. Short-term efficacy was assessed by analyzing the tumor response, serum tumor markers, and immune function before and after treatment. Treatment safety was assessed by monitoring bone marrow suppression and hepatic and renal function impairment.RESULTSThe median overall survival was 12.91 months and 10.64 months in the ICWM and WM groups, while the median progression-free survival was 5.12 months and 3.55 months, respectively. The disease control rate was significantly higher in the ICWM group than that in the the WM group, while the myelosuppression was significantly milder. The serum tumor markers carbohydrate antigen (CA) 19-9 and CA125 showed a significant downward trend before and after treatment in the ICWM group, whereas only CA19-9 showed a significant decrease in the WM group. Post-treatment, both groups showed an upward trend in natural killer cells and CD3+, CD4+, and CD4+/CD8+ lymphocytes compared with pre-treatment, with the ICWM group exhibiting a more pronounced increase. The two groups showed no significant differences in hepatic and renal function impairment.CONCLUSIONICWM extended survival in patients with advanced PC, improved long-term efficacy, controlled local lesions, reduced serum tumor markers, enhanced immune function, and improved short-term outcomes, with a favorable safety profile.

Digital platforms hold promise to scale implementation of population screening. We tailored the Genetics Adviser platform to provide education, decision support, consent, and result return in a genomic newborn screening (gNBS) study, BabyScreen + . Participants were surveyed and interviewed on the usability and value of Genetics Adviser. Genetics Adviser was used by 1048 participants and 1007 (96%) provided feedback. The majority (96%, n = 963) found the platform easy to navigate, with 85% (n = 851) spending <20 min online. Participants demonstrated excellent understanding, over 80% answering at least 6/8 knowledge questions correctly. Only 7% (12/173) of participant-initiated contacts with the study team were for genetic counselling. Interview participants valued the online process. We demonstrate the successful use of a digital platform for a genomic screening program. This model is streamlined, providing consistent, user-friendly education to support decision-making with minimal input from healthcare practitioners. Further evaluation in diverse populations will be essential for future use.

Identifying and analyzing the cooperative effectiveness of regional innovation systems is an important prerequisite for reflecting the operational efficiency and effect of regional innovation systems. The regional innovation system is a complex system network composed of many subsystems. Therefore, it is crucial to propose an effective measurement and evaluation method to reflect the essential characteristics of collaboration. Based on the principle of order parameters in synergitics and the idea of cloud model, this paper proposes the mechanism of order parameters membership cloud in the operation of regional innovation system, then proposes the evaluation process of cooperative order degree of regional innovation system based on order parameters membership cloud, and gives the identification method of order parameters in the cooperative order degree process of regional innovation system, and then builds the order parameters member cloud model. According to the possible results, the evaluation criteria and management and control strategies of cooperative order degree are constructed. Finally, taking the innovation system of 30 administrative regions in China as an example, the measurement and evaluation method proposed in this paper is applied to verify the scientific rationality of the model and method proposed in this paper. The method and control strategy proposed in this paper can provide method support for decision-making of relevant departments and enterprises of regional innovation management.

The growing global demand for energy, driven by population growth and industrial development, has increased the importance of renewable sources such as wind energy. In this context, Türkiye has made remarkable progress in expanding its wind energy capacity, particularly in the Aegean Region. The Bergama district, located in the northern part of İzmir, stands out as a promising area for sustainable wind power plant investments due to its favorable average wind speeds of 8–9 m/s measured at a hub height of 100 m. This study proposes an intelligent fuzzy multi criteria decision framework to determine the most suitable sites for wind power plant installation in the Bergama region. The evaluation process is structured around four main criteria, economic, technical, environmental, and social, each comprising five sub-criteria. Six alternative locations are comparatively assessed using an integrated Fuzzy Analytic Hierarchy Process and Fuzzy Weighted Sum Model approach. The combined model enabled effective handling of uncertainty in decision parameters and provided a consistent ranking of alternatives. Based on the results, Site 6 emerged as the most suitable location due to its superior wind resource characteristics, technical feasibility, and accessibility advantages, and the proposed approach offers a decision support framework for regional planners to guide strategic wind energy development.

Introduction: Patient-reported outcomes (PROs) are increasingly valued in oncology for capturing treatment tolerability and quality of life, and they are emerging as important data sources for precision-medicine and AI-driven clinical workflows. While the EQ-5D-5L questionnaire remains a widely used standardized instrument, dynamic electronic PROs (ePROs) collected via mobile applications generate richer, higher-frequency longitudinal data. Their alignment with established PRO measures, however, is not well-understood, limiting their integration into routine care and downstream analytic applications. In the prospective OGIPRO trial (KEK-ZH 2021-D0051), patients with HER2-positive breast cancer reported well-being and symptoms via the Medidux ePRO platform alongside weekly EQ-5D-5L assessments. In this retrospective analysis, we used linear mixed-effects modeling to examine associations between: (i) dynamic ePRO well-being and the EQ-5D-5L visual analogue scale (VAS); (ii) dynamic ePRO symptom grades and EQ-5D-5L domain sums; (iii) ePRO symptom grades and EQ-5D-5L disutility using the EQ-5D-5L value set for Germany. Materials and Methods: The analytic dataset comprised 13,699 dynamic ePRO data points (3376 well-being ratings and 10,323 symptom grades across 91 symptom types) from 53 patients, forming high-frequency longitudinal patient trajectories. Of these, 252 and 226 time-aligned observations, respectively, were used for direct comparison with EQ-5D-5L VAS and domain scores. Results: Dynamic ePRO well-being showed strong agreement with EQ-5D-5L VAS (β = 1.061, 95% CI: 1.015–1.107), with low between-patient variability. In contrast, the agreement between aggregated ePRO symptom grades and EQ-5D-5L domain sums was weaker (β = 0.404, 95% CI: 0.307–0.501) and more heterogeneous across patients. The same applied to the agreement between ePRO symptom grades and EQ-5D-5L disutility (β = 0.213; 95% CI: 0.151–0.275). Discussion: Dynamic ePRO well-being aligns closely with EQ-5D-5L VAS scores, supporting its use as a pragmatic substitute in clinical and research settings. Aggregated symptom grades, however, showed limited concordance with EQ-5D-5L domains, indicating the need for more granular analyses on larger datasets. Conclusions: Overall, dynamic ePRO systems provide validated, high-resolution longitudinal patient data and represent a scalable foundation for patient monitoring and data-driven decision support in oncology, including future AI-based precision-medicine applications.

Introduction The transformation of chronic disease management is increasingly driven by the integration of AI and multimodal data analytics, enabling precise, individualized, and scalable healthcare interventions. Despite the growing availability of longitudinal and heterogeneous health data, conventional methods are constrained in their ability to model the complex, patient-specific dynamics inherent to chronic conditions. Traditional clinical decision support systems rely on rigid, population-level models that inadequately address inter-patient variability, multi-condition comorbidities, and evolving disease trajectories. Methods To overcome these limitations, we propose a computational framework that utilizes multimodal big data to enable personalized diagnosis, prognosis, and therapeutic optimization. At the core of this framework is the Patient-Adaptive Transition Tensor Network (PATTN), a tensorized dynamical model that captures individual-specific disease evolution through structured latent state representations and high-order temporal dependencies. Complementing this is the Trajectory-Aligned Intervention Recalibration (TAIR), an adaptive decision-making strategy that continuously aligns predicted and observed health trajectories, facilitating real-time treatment policy refinement. This unified pipeline integrates latent trajectory modeling, condition-aware modular representation, and personalized policy optimization. Results and Discussion Experimental evaluations on large-scale multimodal datasets demonstrate superior performance in outcome prediction accuracy, intervention personalization, and trajectory alignment, underscoring the practical applicability of the system in chronic care settings. By combining patient-specific temporal modeling with adaptive therapeutic recalibration, this framework represents a significant advancement toward scalable, intelligent, and individualized chronic disease management leveraging AI and big data infrastructures.

Under the dual background of the “double carbon” commitment and service consumption upgrading, urban tourism hotels urgently need accurate and continuous efficiency evaluation. Addressing the limitations of traditional data envelopment analysis (DEA) models, such as sensitivity to noise and neglect of environmental heterogeneity, this paper proposes an improved DEA evaluation system by integrating slack-based measure non-radial measurement, bootstrap correction, and hierarchical regression. The system analyzes input-output-externality coupling to identify management redundancy, technical inefficiencies, and scale mismatches, offering differentiated improvement paths. Using panel data from star-rated hotels (2018–2023), the model's robustness was validated. A visual radar framework further aids in identifying bottlenecks and can be extended to carbon emissions and customer satisfaction, forming a closed-loop evaluation system. The study expands DEA's theoretical scope in hospitality and offers practical decision support for targeted government subsidies and refined hotel management.

The cultivation of potatoes is essential for rural food security, and the use of Unmanned Aerial Vehicle Red-Green-Blue (UAV-RGB) imagery allows for precise and cost-effective estimation of yield and identification of varieties, overcoming the limitations of manual assessment. We evaluated four INIA varieties (Bicentenario, Canchán, Shulay and Tahuaqueña) by integrating agronomic measurements (height, number and weight of tubers, leaf health) with color and textural indices derived from RGB orthomosaics. Yield prediction was modeled using Random Forest (RF) and Gradient Boosting (GB); varietal identification was approached with (i) a Convolutional Neural Network (CNN) that classifies RGB images and (ii) classical models such as Random Forest, Support Vector Machines (SVMs), K-Nearest Neighbors (KNNs), Decision Trees and Logistic Regression trained on EfficientNetB0 embeddings. The results showed significant genotypic differences in yield (p &lt; 0.001): Tahuaqueña 13.86 ± 0.27 t ha−1 and Bicentenario 6.65 ± 0.27 t ha−1. The number of tubers (r = 0.52) and plant height (r = 0.23) correlated with yield; RGB indices showed low correlations (r &lt; 0.3) and high redundancy (r &gt; 0.9). RF achieved a better fit (Coefficient of determination, R2 = 0.54; Root Mean Square Error, RMSE = 2.72 t ha−1), excelling in stolon development (R2 = 0.66) and losing precision in maturation due to foliar senescence. In classification, the CNN and RF on embeddings achieved F1-macro ≈ 0.69 and 0.66 (Receiver Operating Characteristic—Area Under the Curve, ROC AUC RF = 0.89), with better identification of Bicentenario and Shulay. We conclude that UAV-RGB is a cost-effective alternative for phenotypic monitoring and varietal selection in high Andean contexts. These findings support the integration of UAV-RGB imagery into breeding and monitoring pipelines in resource-limited Andean systems.

The automation of radiology services has significantly improved access to radiological imaging for accurate diagnosis of diseases and injuries. However, the expansion of radiological equipment, the adoption of telemedicine, and the integration of AI-powered clinical decision support systems necessitate upgrades to existing medical image storage and processing solutions. This article reviews modern compression methods for radiological images, which offer higher compression ratios, im­proved image quality, and faster encoding/decoding times compared to the standards defined by the DICOM specification. It is established that radiological images possess unique characteristics—such as high noise levels, locally symmetric regions (similar patches), and the presence of multiple sequential frames in a single study—which, when accounted for in compression algorithms, can enhance compression efficiency. Implementing advanced data compression approaches can increase the fault tolerance of high-load medical systems and reduce costs associated with the storage, transmission, and processing of diagnostic studies.

The chronic evolving neurodegenerative disorder Alzheimer's Disease (AD) presents with memory deficits, cognitive impairment, and loss of abilities. AD prevalence is increasing as the world ages, necessitating more precise and easily obtainable diagnostic and treatment approaches. Artificial Intelligence (AI) technologies, and more specifically, machine learning and deep learning, have become game-changers in Alzheimer's disease patient care, including optimizing care, enabling early diagnosis, supporting differential diagnosis, and predicting disease progression over the last few years. To detect amyloid plaques and hippocampal atrophy, this study investigates how AI-driven imaging analysis, specifically convolutional neural networks applied to MRI (Magnetic Resonance Imaging) and PET (Positron Emission Tomography) scans, provides higher sensitivity and specificity. Artificial intelligence models are also being used to analyze clinical and genomic data to identify biomarkers and support risk stratification. AI-assisted cognitive tests provide scalable, non-invasive, and real-time screening. Telemedicine platforms and AI-based Clinical Decision Support Systems (CDSS) are also improving patient management, particularly in remote or underserved areas. Heterogeneity of data, model explainability, ethics, and regulatory guideline requirements remain issues, despite these latest developments. Beyond recent developments such as federated learning and digital twins, the study comprehensively reviews AI's contributions to AD diagnosis and therapy. It also establishes a guide for future research directions for the ethical and equitable integration of AI in clinical practice.

Generic pediatric patient-reported outcome measures (P-PROMs) have the potential to enhance care and patient-clinician interactions in specialty hospital settings. However, evidence about their feasibility and acceptability is lacking. To identify the feasibility and acceptability of a P-PROM at the point of care among children receiving outpatient care from selected specialty clinics. This nonblinded, pilot feasibility and acceptability randomized clinical trial was conducted from February to June 2024 across 4 pediatric specialty clinics (asthma, sleep, encopresis, and chronic constipation) at The Royal Children's Hospital in Melbourne, Victoria, Australia. Children aged 4 to 17 years (and their caregivers) were eligible for inclusion if they had an appointment at one of the participating clinics during the trial period. Patients and their caregivers were randomly assigned to the intervention or control group. Clinicians (physicians, nurses, and allied health staff) providing specialty care services to eligible patients were also invited to participate. Children and/or their caregivers assigned to the intervention arm were asked to complete a generic P-PROM-the EuroQoL 5-Dimensional Questionnaire for Youth, 5 Levels (EQ-5D-Y-5L)-7 days before their appointment and to indicate which EQ-5D-Y-5L items they would like to discuss with their clinician during their appointment. Responses to the completed P-PROM were displayed to clinicians in the electronic medical record, and children and their caregivers received information to act on the P-PROM items. Clinicians received training, clinical decision support, and resources to support patient actions. Patients assigned to the control arm received standard outpatient care, which excluded completing a P-PROM. Primary outcomes were feasibility and acceptability of the P-PROM. Of 170 eligible patients, 87 children (51.2%) and their caregivers were randomly assigned to the intervention arm (n = 43) or control arm (n = 44). Patients included 44 females (50.6%) with a mean (SD) age of 8.8 (3.2) years, and caregivers included 82 females (94.3%) with a mean (SD) age of 41.8 (6.9) years. Of the 17 eligible clinicians, 14 were included in the study; they reported working in a specialty clinic for a mean (SD) of 9.7 (8.4) years. Thirty-three of 37 caregivers (89.2%) in the intervention arm and 9 of 14 clinicians (64.3%) reported that the P-PROM was acceptable. A P-PROM completion rate of 93.0% (40 of 43 patients) was achieved, indicating feasibility. In this pilot randomized clinical trial, the collection and use of the EQ-5D-Y-5L was feasible and acceptable in routine outpatient specialty pediatric care. Further study should examine the quantitative impacts of the intervention on quality of care and outcomes as well as the impacts over time. ISRCTN Identifier: ISRCTN16030620.

To promote rational drug use in developing countries, it is important to assess drug use patterns. This study assessed the drug prescription patterns of the family medicine clinic at the Outpatient Pharmacy of the University of Ghana Medical Centre using the World Health Organization's drug use indicators. An analytic, cross-sectional survey with data extracted from patient's electronic medical records was carried out. Questionnaires were given to all prescribers in the family medicine clinic to evaluate factors related to rational medicine use. Frequencies and percentages were employed for description with further analysis, including Zero-inflated Poisson regression and logistic regression, used to determine associations between variables with a 95% confidence interval. Of the 600 participants whose prescriptions were analyzed, 367 (61.17%) were male and 233 (38.83%) were female. The prescribers interviewed were 3 males and 7 females. The mean number of medications per prescription was 1.4 (SD = 1.61), with antibiotics and injections making up 12.74% (n = 107) and 4.17% (n = 35) respectively. Generic prescriptions were 34.88% (n = 293) and those from the Essential Medicines List (EML) were 72.38% (n = 608). Prescriptions with a record of diagnosis were 50.83% (n = 305). Patients with comorbidities were shown to have a 52.2% lower prevalence rate of the total number of medications prescribed compared to those without comorbidities (p-value <0.001). Female patients have 46.4% reduced odds of being prescribed an antibiotic compared to male patients (p-value 0.012). The odds of a patient with a chronic condition being prescribed an antibiotic is 93.2% more than that of a patient without a chronic condition (p-value = 0.025). Additionally, the prevalence of drugs prescribed from the EML for a patient with a chronic condition is 74.4% lower than that prescribed for patients without a chronic condition (p-value = 0.048). There was moderate adherence to rational prescribing. Three prescribing indicators met reference standards, these were: average number of medicines per encounter, percentage of prescriptions with an injectable and percentage of encounters with antibiotics. Rational drug prescribing may be enhanced through training, guidelines, EML distribution, drug and therapeutics committee support and integrated Clinical Decision Support Systems (CDSS).

Path planning for large-scale agricultural fields faces challenges such as irregular field shapes, uncertain boundaries, and the need to balance path efficiency, energy consumption, and coverage quality. To address these problems, this research introduces a strategy-aware hierarchical hybrid optimization framework (HANS) for autonomous agricultural operations. This framework introduces a global principal axis extraction method based on Principal Component Analysis (PCA), utilizing the statistical distribution of field boundaries to guide path direction, thereby improving robustness against boundary noise and irregular geometries. The framework integrates Adaptive Large Neighborhood Search (ALNS) for global exploration and Tabu Search (TS) for local optimization, forming a tightly coordinated hybrid structure. The framework further employs a Pareto-set-based decision support selection strategy to solve a multi-objective optimization model encompassing machine kinematics, turning patterns, and energy-aware cost evaluation. This strategy provides three methods: weighted preference-based compromise solution selection, crowding distance-based diversified solution selection, and single-objective extreme value-based dedicated optimization solution selection. To balance the impact of path length, energy consumption, and coverage rate, we assigned equal or nearly equal weights to them (i.e., (0.33, 0.33, 0.34)). Furthermore, the framework incorporates operators and feedback learning mechanisms specific to agricultural coverage path problems to enable adaptive operator selection and reduce reliance on manual parameter tuning. Simulation results under three representative field scenarios show that compared to fixed-direction planning, HANS improves the average coverage rate by 0.51 percentage points and reduces fuel consumption by 4.34%. Compared to Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Tabu Search (TS), and Simulated Annealing (SA), the proposed method shortens the working path length by 0.37–0.83%, improves coverage rate by 0.34–1.11%, and reduces energy consumption by 0.61–1.03%, while maintaining competitive computational costs. These results demonstrate the effectiveness and practicality of HANS in large-scale autonomous farming operations.

Accurate preoperative assessment of lesion size is crucial for selecting the appropriate endoscopic resection technique. However, the current assessment of lesion size still mainly relies on visual estimation, lacking objective measurement methods. To develop and validate an Endoscopic Virtual Ruler (EVR) based on image detection technology for objective measurement of lesion size before endoscopic treatment. Using computer image recognition technology and laser spot imaging principle, EVR was formed to detect the size of lesions. In vitro animal and human experiments were carried out to verify the accuracy and safety of EVR by comparing its measurement results with the actual size and the visual inspection results of endoscopists. In 30 in vitro tests, the measurement error of EVR was 0.08 ± 0.17cm (95% CI 0.01-0.14), and the relative accuracy of the measurement was 92.80% ± 5.50%( 95% CI 90.75-94.85%). In 58 clinical lesions, the mean error for visual estimation was 0.16 ± 0.66cm (95% CI-0.01 to 0.33), while EVR showed 0.12 ± 0.32cm (95% CI 0.04-0.21). EVR was significantly more accurate (85.68% ± 15.25%) than visual estimation (67.08% ± 22.59%, p < 0.001). EVR was more effective [48 (82.8%) vs 31 (46.6%), p = 0.001]. In the multivariable model, EVR-assisted measurement was independently associated with achieving clinically acceptable accuracy (OR 4.38, 95% CI 1.84-10.43, p = 0.001). EVR also demonstrated higher consistency in lesion size classification (Kappa = 0.764 vs. 0.522, p < 0.001). For lesions < 1cm, EVR misclassified only 12.5% as 1-2cm, significantly less than the 50% misclassification rate with visual estimation (p = 0.034). There was no laser damage side effect. EVR offers an accurate, safe, and objective measurement tool, which is helpful for the formulation of appropriate treatment decisions. ChiCTR2400085998.

Sustainable land use requires precise monitoring of soil pollution, yet accurately predicting the spatial distribution of heavy metals often relies on post hoc accuracy comparisons with limited a priori diagnosis. To address the challenge of cost effective environmental monitoring, we conducted a PRISMA guided systematic review (2000–2024) and synthesized 135 studies to develop a mechanism-informed, context aware method selection framework. Evidence revealed three regularities: (i) element–driver coupling is structured (Pb/Cd/Zn predominantly anthropogenic; Cr/Ni geogenic; As/Hg mixed), with dominant influence scales from local to regional; (ii) model performance hinges on alignment between algorithmic assumptions, and context hybrid machine learning models integrating multi-source covariates tend to excel under strong, non-stationary anthropogenic heterogeneity, whereas kriging variants are more robust when geogenic continuity holds; and (iii) applicability is jointly constrained by environmental context, data foundations, and management objectives. Building on these insights, we propose a three-step decision workflow—goal definition, contextual diagnosis, and method matching. This framework serves as a decision support tool that shifts selection from trial and error to a priori alignment, optimizing resource allocation and enhancing the reliability of pollution assessments for sustainable soil remediation and policymaking.

Arable land quality is of the essence for the sustenance of grain production and food security. The continuous monitoring of the physical and chemical properties of arable land is instrumental in facilitating a comprehensive understanding of the evolution patterns of soil quality. This, in turn, provides fundamental evidence that is crucial for the optimization of cultivation practices, the establishment of appropriate plough layers, and the enhancement of soil quality. The near-surface sensing methodologies facilitate the acquisition of soil data at reduced scales, thus signifying a pivotal research trajectory for the procurement of soil-related information. The present study undertakes an examination of the current state of research on acquiring key parameters of farmland soil and provides an overview of the fundamental ground-level techniques employed for the assessment of farmland soil parameters. These techniques encompass single-parameter fixed-point detection, encompassing Soil Moisture Content (SMC), Soil Electrical Conductivity (EC), and nutrient analysis, multi-parameter fusion detection, and dynamic parameter monitoring. The study systematically reviews field sensing methods for major soil physicochemical parameters (such as SMC, Soil Penetration Resistance (SPR), EC, and nutrients) while analyzing the current application of Artificial Intelligence (AI) in soil parameter detection. The present paper proposes a developmental trajectory that shifts from “single-parameter static” to “multi-parameter dynamic” monitoring. This trajectory is proposed as a building upon the analysis of existing research. This evolution emphasizes intelligent algorithm-driven data enhancement to improve detection accuracy, forming a closed-loop progression of “dynamic detection—precise modeling—decision support”. This framework provides a reference for the advancement of soil sensing monitoring technologies and the scaling of precision agriculture applications.

Botox is one of the most frequently performed procedures in cosmetic dermatology, aimed at reducing wrinkles and enhancing facial aesthetics. However, the procedure is technically demanding, time-consuming, and physically fatiguing, and it is prone to both intra- and inter-expert variability. Consequently, automated systems play a crucial role in ensuring accurate and consistent identification of injection points. In this study, we propose an enhanced YOLOv8-based object detection framework by integrating two architectural modules: the Convolutional Block Attention Module (CBAM) and the Bidirectional Feature Pyramid Network (BiFPN), to enable precise detection of Botox injection points on facial images. The proposed approach is evaluated on four clinically relevant subsets of a novel high-resolution dataset, demonstrating consistent improvements over the baseline YOLOv8n architecture. The integration of CBAM and BiFPN results in relative mAP@0.5 gains ranging from 1.3 to 4.2% on the validation sets, with the most significant improvements observed in small wrinkle localization tasks. Overall, the proposed system presents a promising step toward AI-assisted clinical decision support in cosmetic dermatology.

Accurate assessment of source-load complementarity and system regulation capacity is critical for secure dispatch and planning in high-penetration renewable power systems. Addressing limitations of existing methods—which rely heavily on static metrics, struggle to capture temporal and tail dependence characteristics, and provide insufficient support for dispatch decisions—this paper proposes a multi-level integrated evaluation framework. First, from a source—load matching perspective, we develop a novel complementarity metric, integrating real-time rate of change, temporal consistency, and tail dependency. An improved adaptive noise-complete set empirical mode decomposition combined with a hybrid Copula model is employed to isolate noise and to precisely quantify dynamic dependency structures. Second, we introduce the Minkowski measure and construct a net load fluctuation domain accounting for extreme fluctuations and coupling relationships. Subsequently, combining the Analytic Hierarchy Process (AHP) with probabilistic convolution enables multi-level comparative quantification of resource capacity and fluctuation domain requirements under varying confidence levels. Simulation results demonstrate that the proposed framework not only provides a more robust assessment of source-load complementarity but also quantitatively outputs the adequacy and risk level of system regulation capacity. This delivers hierarchical, actionable decision support for dispatch planning, significantly enhancing the engineering applicability of evaluation outcomes.

Abstract—Practical decision support in farming often requires two complementary capabilities: identifying crops that match local soil–weather conditions and estimating how strongly irrigation may be needed under the same conditions. This paper presents a compact machine-learning workflow that addresses both tasks using a public crop-recommendation dataset. First, a Random Forest classifier maps soil nutrients (N, P, K) and environmental measurements (temperature, humidity, pH, rainfall) to one of 22 crop classes. Second, because public data typically lacks ground-truth irrigation volumes, we define a transparent Irrigation Demand Score (IDS) that increases with thermal stress and decreases with rainfall and humidity, while mildly accounting for pH deviation and nutrient imbalance. Multiple Linear Regression is then trained to predict IDS for interpretability and low-cost deployment. On the held-out test split, the classifier achieves 98.5% accuracy, and the regression attains R2 =0.87 against the engineered score. The overall system is reproducible, lightweight, and suitable for low-instrumentation contexts, while remaining extensible to real sensor-based water measurements in future work. Index Terms—Precision agriculture, crop recommendation, Random Forest, linear regression, irrigation demand score, soil nutrients, machine learning.

Diagnostic errors are a major carehealth concern but remain difficult to study because their identification often requires resource-intensive chart reviews. We aimed to validate a previously proposed automated method for detecting discrepancies between an initial and a later, more definitive diagnosis as a screening tool for potential diagnostic errors in a large, prospective cohort of emergency department (ED) patients. This secondary analysis included 1,204 patients enrolled in the DDxBRO randomized trial, which evaluated the effect of a diagnostic decision support tool on diagnostic quality in four Swiss emergency departments. For each patient, the ED diagnosis was extracted from the ED discharge letter, and the follow-up diagnosis at 14days was obtained from hospital discharge letters, or general practitioner notes. All diagnoses were coded using ICD-10 and manually classified for discrepancies by two blinded ED physicians according to a predefined scheme. The automated method calculated the "similarity" between ICD-10 codes for ED and follow-up diagnoses. Discriminative performance of this method to distinguish between cases with and without diagnostic error was evaluated using receiver operating characteristic (ROC) curves, and sensitivity, specificity, and predictive values were assessed across multiple cutoffs. The automated method showed high and consistent discriminative performance across all algorithms tested, with areas under the ROC curve (AUCs) ranging from 0.94 to 0.95. Using the most sensitive cutoff in the simplest algorithm, all true discrepancies were detected, but 162 cases (15%) were incorrectly flagged as discrepant. The automated method demonstrated high accuracy and shows promise as a practical screening tool to prioritize cases for resource-intensive chart review. NCT05346523.