Browse Articles
Discover research articles across all indexed journals
Using large language models for enhanced fraud analysis and detection in blockchain based health insurance claims
Abstract Traditional health insurance claim processing systems are plagued by inefficiencies and vulnerabilities, often resulting in significant financial losses due to fraudulent activities. Existing fraud detection methods are largely manual, time-consuming, and inadequate for handling the complexity and scale of modern fraudulent schemes. Moreover, the trust-based relationships between insurers and healthcare providers lack mechanisms to ensure data integrity and prevent manipulation. While several blockchain-based systems have been proposed to improve transparency and tamper resistance, they typically focus on structured data and predefined fraud types, offering limited adaptability and analytical insight. This paper proposes a novel solution leveraging blockchain technology and Large Language Models (LLMs) to transform fraud detection. The system uses Ethereum smart contracts (SCs) to securely store medical records and claim details on a decentralized, tamper-proof ledger that ensures data integrity, traceability, and accountability. This immutable data is accessed by an LLM via a Retrieval-Augmented Generation (RAG) system, which enables intelligent retrieval and analysis of relevant clinical information to detect fraud patterns and inconsistencies. To support complex scenarios involving free-text documents, unstructured clinical data, such as lab reports, are stored using decentralized off-chain storage and retrieved during LLM analysis. In addition, an LLM-powered chatbot also allows insurance providers to interact with the system in natural language for claim inquiries, explanations, and summaries. The architecture, sequence diagrams, and implementation algorithms outline the development process, while testing scenarios demonstrate the system’s ability to detect fraud such as inflated costs, unnecessary treatments, and unrendered services. Evaluation using both synthetic and public clinical datasets showed strong performance, with the LLM achieving up to 99% fraud detection accuracy. Cost, security, and scalability analyses confirm the system’s practicality and resilience, with the complete detection process executing in just 13 seconds. By overcoming the limitations of traditional systems, this framework offers a scalable and adaptable approach for healthcare and other domains. The SCs and source code are publicly available on GitHub.
Ultrahigh-Throughput Multiplexed Screening of Purified Protein from Cell-Free Expression Using Droplet Microfluidics
Correlation analysis and comprehensive evaluation of dam safety monitoring at Silin hydropower station
Abstract Dam failures pose catastrophic risks to human life and property, necessitating robust safety monitoring systems for risk mitigation. However, the specific contributions of distinct monitoring modalities to dam safety remain inadequately characterized, particularly regarding their differential impacts on structural integrity assessment. This study investigates the correlation between diverse monitoring modalities and dam structural safety through a comprehensive analysis of the Silin Hydropower Station dam. We analyzed 324 datasets collected from nine types of monitoring sensors installed across 36 dam cross-sections. Statistical analyses including one-way ANOVA, cluster analysis, and principal component analysis (PCA) were employed to quantify the influence patterns of monitoring parameters. The safety impact levels of all 36 cross-sections were systematically ranked, establishing a prioritized reference framework to inform decision-making in dam safety management. Unlike conventional dam safety assessments that predominantly rely on subjective empirical judgments, this study introduces an objective methodology integrating principal component analysis (PCA) of heterogeneous monitoring data across multiple dam cross-sections. The analytical outcomes were systematically quantified, hierarchically ranked, and visualized through multidimensional mapping techniques. The results demonstrated that variations in fissure (X2), horizontal displacement (X3), tilt (X4), stress (X6), soil-displacement (X8), and denotes water-level (X9) exerted highly significant effects on dam safety (p < 0.001). The first two principal components cumulatively accounted for 74.1876% of the total variance, with eigenvalues reaching 6.6769. In the comprehensive evaluation, cross-section T4 (T4) obtained the maximum score (0.8500), while cross-section T35 (T35) showed the minimum score (0.0175). In conclusion, the analysis revealed that X9, X8, X2, X3, and X4 exerted significant impacts on dam safety, while cross-section T4 achieved the highest comprehensive evaluation score. This approach employs Principal Component Analysis (PCA) with integrated scoring to reduce multivariate dimensionality, enabling rapid identification of key monitoring sections critical to dam safety, and demonstrates broad applicability for dam safety monitoring.
Daily briefing: Why some places on our bodies heal without scars
Collinear Jahn–Teller Ordering Induces Monoclinic Distortion in “Defect-Free” LiNiO<sub>2</sub>
A numerical approach to fractional Volterra–Fredholm integro-differential problems using shifted Chebyshev spectral collocation
In Situ Quantification of Directional Rotation by a Catalysis-Driven Azaindole-<i>N</i>-Oxide–Phenoic Acid Molecular Motor
Vaccinia virus modulates the redox environment by inhibiting reactive oxygen and nitrogen species with increased activity of endogenous antioxidant enzymes
In Situ Quantitative Imaging of Nonuniformly Distributed Molecules in Zeolites
Assessing chemical properties and heavy metals in groundwater resources in a developing country: a baseline study
Unshielded Ino Decahedral Core Copper Nanoclusters Enable Efficient Photocatalytic Sulfonylation of Aryl Halides
Construction and validation of a predictive in-hospital mortality nomogram in patients with staphylococcus aureus bloodstream infection
Abstract We aimed to construct and validate a predictive nomogram to evaluate in-hospital mortality of patients with S.aureus BSI. A 10-year retrospective cohort design was conducted to analyze data from 484 patients diagnosed with S. aureus BSI between 2014 and 2023. Clinical data from 339 patients (2014 to 2021) were harnessed in training cohort to develop a predictive nomogram, which underwent rigorous internal validation. An independent cohort of 145 patients (2022 to 2023) were collected for external validation. The prognostic performance of the model was comprehensively assessed using AUC, calibration curve, and DCA. We ultimately identified several key factors that were incorporated into the final prognostic nomogram: the ECFC score, the CCI score, procalcitonin levels, admission to the intensive care unit, and multimicrobial BSI. Internal validation was assessed via 5-fold cross-validation, repeated 400 times on the training cohort, yielding an average AUC value of 0.930 vs. 0.940 of the total. External validation further confirmed the nomogram’s accuracy, with an AUC value of 0.929. Additionally, the calibration curves and DCAs revealed excellent consistency and substantial net clinical benefits in both cohorts. The development of this predictive nomogram marks a substantial breakthrough in the management of patients with S. aureus BSI.
Time-Domain Observation of Ultrafast Self-Trapped Exciton Formation in Lead-Free Double Halide Perovskites
Pythagorean fuzzy N-bipolar soft sets-based multi-criteria decision-making framework for sustainability evaluation and risk assessment in manufacturing industries
How to thrive as a Latin American researcher abroad
Spin Sensor-Integrated Covalent Organic Frameworks Reveal the Effect of Host–Guest Interactions on the Catalytic Activity of Immobilized Enzymes
A phased online support program improves self-efficacy and reduces distress in type 2 diabetes patients using mixed methods design
Structural Control of Metal-Centered Excited States in Cobalt(III) Complexes via Bite Angle and π–π Interactions
A qualitative interview study exploring the experiences of pain specialists on prescribing opioids for chronic non-cancer pain
Abstract Opioid prescribing for patients with chronic non-cancer pain is common despite issues associated with long-term efficacy, functional improvement, and safety. Pain specialists assess many patients with chronic non-cancer pain, but their experiences of this situation are not well represented in prior qualitative research. The aim of this study was to explore pain specialists’ experiences of prescribing opioids to patients with chronic non-cancer pain. We adhered to the Consolidated Criteria for Reporting Qualitative Research guidelines. Pain specialists in Sweden were recruited by purposive and snowball sampling. Participants were digitally interviewed, audio and video were recorded, and interviews translated verbatim. Data was analyzed through manifest inductive content analysis. Twenty pain specialists were interviewed. Qualitative content analysis revealed that the pain specialists’ experiences were represented by two main categories: (1) Navigating the doctor-patient relationship, and (2) Challenges and opportunities when prescribing opioids. The first main category describes the relational demands associated with opioid prescribing and includes communication, conflicts, managing expectations, and the emotional and ethical aspects of prescribing opioids. The second main category describes handling complexity and heterogeneity, organizational aspects, and the doctor’s due diligence when prescribing opioids. Our results offer new insights into pain specialists’ experiences of prescribing opioids for chronic non-cancer pain, offering health care professionals guidance for responsible opioid prescribing. Pain specialists highlight the need for structured pain assessments and identify system-level improvements, such as allocating sufficient time, enabling team-based care, and continued education initiatives, to support safe opioid use in health care.