Search Results (92)

Today, data are no longer confined to numerical values arranged in row-by-column matrices or stored neatly within relational databases. One of the defining characteristics of big data is its high variety, encompassing unstructured and multimodal forms such as text, audio, images, and video. These data types dominate contemporary domains including social media, digital humanities, biomedical research, education, and surveillance systems. Yet these data types remain difficult to manage and analyze using traditional data management architectures. To cope with this shift, modern data management systems must move beyond schema-driven designs and incorporate multimodal artificial intelligence capable of understanding, integrating, and reasoning across heterogeneous data modalities. This article examines how multimodal AI, in particular large multimodal foundation models, can be leveraged to support the ingestion, representation, organization, and analysis of unstructured data. It discusses emerging multimodal data management frameworks, outlines a conceptual pipeline for multimodal data analysis, and highlights key challenges related to scalability, interpretability, and governance. By situating multimodal AI at the core of data management, this work argues that effective data analysis in the era of big data requires systems that treat meaning, context, and cross-modal relationships as first-class computational objects rather than afterthoughts. Full article

Background: Colorectal cancers (CRCs) harboring the BRAFV600E (V600E) mutation exhibit aggressive clinical behavior and chemotherapy resistance, yet the underlying mechanisms remain poorly understood. Ferroptosis, which is driven by iron-dependent lipid peroxidation, has emerged as a potential therapeutic vulnerability. This study aimed to explore whether the NF-κB/Lipocalin 2 (LCN2) pathway modulates cisplatin sensitivity through Fenton-reaction-induced ferroptosis in BRAFV600E-overexpressing CRC cells. Methods: BRAF mutation status and the expression of LCN2, PTGS2, and cleaved caspase3 were examined in clinical CRC specimens by immunohistochemistry. The correlations between ferroptosis and apoptosis markers and 5-year survival rate were evaluated in TCGA datasets. CRC cells with LCN2 knockdown/knockout or BRAF/V600E/LCN2 overexpression were established to assess the proliferation, lipid metabolism, iron levels, and NF-κB/LCN2 signaling under cisplatin treatment. In vivo studies were employed with BALB/c xenograft models. Results: V600E-mutant clinical specimens exhibit significantly reduced expression of the ferroptosis marker PTGS2, and the iron metabolism regulators LCN2. Within a KRAS-mutant cellular model, V600E overexpression attenuated cisplatin-induced ferroptosis through suppression of the NF-κB/LCN2 signaling axis, leading to impairment of Fenton-reaction-mediated lipid peroxidation. Restoration of LCN2 expression re-sensitized V600E-overexpressing cells to cisplatin both in vitro and in vivo. Interestingly, inhibition of apoptosis contributes to the resistance of cisplatin induced ferroptosis in V600E overexpression cells, implying a crosstalk between ferroptosis and apoptosis within the therapeutic resistance. Conclusions: Our findings show that the NF-κB/LCN2 axis drives Fenton-reaction-induced ferroptosis to promote the vulnerability of V600E overexpression CRC cells within a KRAS-mutant background to cisplatin. LCN2 restoration partially overcomes V600E overexpression resistance both in vitro and in vivo, suggesting LCN2 as a promising therapeutic target. The crosstalk between ferroptosis and apoptosis may offer potential strategies to overcome chemotherapy resistance of this high-risk CRC subtype. Full article

Record linkage is a fundamental step in ensuring the quality of data by detecting duplicate records within different databases. Nevertheless, dealing with big, imbalanced databases and ensuring data confidentiality is still difficult in terms of performance and precision. This paper introduces a new Privacy-Preserving Record Linkage (PPRL) method named PPRL-Stack, which uses the Bloom filter encoding technique to hide information and a Stack Ensemble structure for classification. The proposed model consists of Support Vector Machine (SVM) as a base learner and Logistic Regression (LR) as a meta-classifier in combination with the application of Sorted Neighborhood Method (SNM) technique to bring down the time complexity to O(N log N). Experiments conducted on the Freely Extensible Biomedical Record Linkage (FEBRL) and North Carolina Voter Registration (NCVR) databases prove that the proposed PPRL-Stack can obtain nearly perfect discrimination with an F1-score of 0.9921. Particularly, our proposed architecture is more than 340 times and 40 times faster than the latest Siamese Bidirectional Long Short-Term Memory (Bi-LSTM) architecture in training and validation stages, respectively. Full article

Background/Objectives: A substantial proportion of eyes with exudative macular diseases show a suboptimal response to standard-dose anti-vascular endothelial growth factor (anti-VEGF) therapy, and evidence regarding clinical outcomes after same-agent double-dose escalation remains limited. We evaluated the functional and anatomical outcomes of same-agent double-dose anti-VEGF therapy in such eyes. Methods: In this observational real-world cohort study, we retrospectively reviewed eyes with exudative macular diseases that were switched to same-agent double-dose anti-VEGF therapy because of a suboptimal response to prior standard-dose treatment. Eligibility required clinician-assessed persistent or recurrent exudative activity during the standard-dose treatment course despite at least three prior anti-VEGF injections, based on the longitudinal treatment and imaging history. Insufficient visual response and inadequate durability were considered additional, potentially overlapping features supporting dose escalation. Primary outcomes were functional and anatomical changes after switching. Secondary outcomes included injection interval, time-to-event outcomes, and associated baseline factors. Results: A total of 104 patients (111 eyes) were included. Mixed-effects modeling showed a small longitudinal improvement in BCVA of 0.007 logMAR per month (p = 0.028) and an estimated CST decrease of approximately 2.1% per month (p < 0.001). In visit-specific analyses, BCVA did not differ significantly from baseline at the first three post-switch visits and differed significantly only at the final follow-up (p = 0.004); however, the median BCVA was 0.70 logMAR at both baseline and final follow-up, and the median within-eye change was 0.00 logMAR (IQR, −0.25 to 0.10). CST was significantly lower than baseline at all post-switch visits (all p < 0.01). Among the 73 eyes with calculable paired interval data, injection intervals did not differ significantly between the standard-dose and double-dose phases (p = 0.289). Baseline pigment epithelial detachment was associated with a less favorable BCVA change (p = 0.002), and worse baseline visual acuity was associated with a higher risk of subretinal fibrosis (hazard ratio, 2.854; p = 0.002). Conclusions: In this uncontrolled pre–post cohort, anatomical changes after same-agent double-dose escalation were more consistent than functional changes. The BCVA change was statistically detectable but small in magnitude and of uncertain clinical relevance. No significant interval extension was observed among the 73 eyes with complete paired interval data during the relatively short follow-up. Full article

Background/Objectives: Acute type A aortic dissection (ATAAD) has high preoperative mortality, and an interpretable multivariable model based on clinically accessible data is crucial for clinical risk stratification. Methods: The data for this study were obtained from West China Hospital, Sichuan University. Centroid regression was used to construct the predictive model, with logistic regression, classification and regression tree, explainable boosting machine and extreme gradient boosting as the reference. Variables were screened by iterative selection, the literature review and clinical experience. Model performance was evaluated by accuracy, sensitivity, precision, Youden’s index, AUROC and AUPRC. Results: The vital signs and tests of 361 ATAAD patients during the first 24 h of their first admission were included in the final analysis. Centroid regression outperformed logistic regression, achieving accuracy (90.7% vs. 81.5%), sensitivity (0.813 vs. 0.741), specificity (0.983 vs. 0.900), Youden’s index (0.796 vs. 0.641), AUROC (area under the receiver operating characteristic curve, 0.953 vs. 0.843) and AUPRC (area under the precision–recall curve, 0.978 vs. 0.863) in the test set. It revealed that the use of α-blocker (the weights w = −1.20) and hydrochlorothiazide (w = −1.20), clinical features like dyspnea (w = −0.94), chest pain (w = 0.91) and lactate dehydrogenase (w = −0.95) were variables that had the greatest impact on model prediction. Conclusions: The centroid regression model not only has relatively high predictive performance and interpretability but also can be easily implemented in hospital systems to provide a practical and cost-effective tool for ATAAD preoperative risk stratification. Full article

Raw lacquer, a natural polymer derived from the bast of lacquer trees (Toxicodendron vernicifluum), is renowned as the “King of Coatings” due to its exceptional film-forming properties, abrasion resistance, corrosion resistance, and biocompatibility. However, its inherent limitations—including stringent drying conditions, slow curing rates, deep coloration, and difficult application—have severely restricted its modernization and widespread adoption. This review systematically summarizes recent research advances in the modification and application of raw lacquer, focusing on four major modification strategies: (1) Nanocomposite modification—incorporating functional nanofillers such as Al₂O₃, cellulose nanofibrils (CNF), polydopamine (PDA) melanin-like nanoparticles, and SiO₂ to significantly enhance film hardness, compactness, UV-aging resistance, and drying kinetics. (2) Chemical structure modification—employing molecular design strategies including aminoanthraquinone grafting, tung oil blending, water-based emulsification, and terpene/allyl group functionalization to improve hydrophobicity, flexibility, fast-drying properties, and achieve dual photo/oxygen curing. (3) Biomass synergistic composites—utilizing natural polymers such as chitosan and lignin, along with bio-inspired adhesion mechanisms (e.g., PDA), to confer advanced functionalities including antibacterial and antifouling properties. (4) Curing behavior regulation—precisely controlling drying kinetics through inorganic salt ion microenvironment engineering, nonionic surfactants, and salicylaldehyde Schiff base-based driers. Building upon these foundations, this review further expands on the emerging high-value applications of modified lacquer in preventive conservation of cultural heritage, advanced functional coatings (anti-corrosion, super-hydrophobicity, flame retardancy), biomedical materials (hemostasis, antibacterial activity, drug-controlled release, water treatment adsorption), and intelligent responsive flexible electronics. Finally, addressing challenges including weak fundamental research, bottlenecks in green industrialization, and lack of standardization, future development directions are proposed encompassing interdisciplinary innovation, sustainable modification strategies, integration of multifunctional intelligent systems, and big data-driven research paradigms, aiming to provide theoretical guidance and technical references for the high-value utilization and modernization of lacquer resources. Full article

Background: Steroid resistance indicates poor prognosis in pediatric nephrotic syndrome, but predictive models and risk factors for steroid-resistant nephrotic syndrome (SRNS) remain poorly understood. Methods: We searched PubMed, Embase, Scopus, CNKI, SinoMed, Wanfang, and VIP (inception to 1 March 2025) for studies developing SRNS prediction models or identifying risk factors. Odds ratios and AUC were pooled using random-effects meta-analysis. Risk of bias was assessed with PROBAST and Newcastle-Ottawa Scale. Results: Out of 2264 studies, 23 were included. Prediction models were mainly developed using logistic regression (16/17, 94.1%). The most frequently reported predictors included erythrocyte sedimentation rate and vitamin D binding protein. The reported AUC ranged from 0.75 to 0.88. Only one model had undergone external validation with an accuracy of 0.94. A total of 22 independent risk factors were identified, five of which—low birth weight, decreased urine output, hypertension, serum albumin, and serum IgM—were not in existing models. In total, 76% of model studies and 26% of risk factor analyses were at high or moderate risk of bias. Conclusions: Existing SRNS prediction models reported apparent discrimination but had a high risk of bias and very limited external validation, which substantially restricts their current clinical applicability. Several relevant risk factors remain unincorporated. Future research should prioritize rigorous model development and multi-center external validation. Full article

Water-soluble fullerene derivatives exhibit a wide range of fascinating biological properties, including antioxidant, antiviral, antitumor, antibacterial, and myogenic effects. During the initial stage of research, most of the reported data on the biological activity of fullerenes were obtained using complex, inseparable mixtures of regiomers with a big focus on fullerenols as the most accessible form of water-soluble fullerene-based compounds. However, during the past decade, significant progress has been made in the synthesis of various isomerically pure water-soluble fullerene derivatives, which opens up possibilities for more directed investigations of their biological activity. In this review, we will highlight current methods for the straightforward synthesis of different types of water-soluble fullerene derivatives with well-defined molecular structures. Special attention will be paid to the possibilities of the precise control of the number, types, and positions of functional groups on the fullerene cage. We will also discuss the opportunities for and challenges within the biomedical applications of water-soluble fullerene derivatives. Full article

(1) Background: Climate change and inequality are topics of major interest in Mediterranean Archaeology. However, comparatively less attention has been dedicated to how these themes are interlinked in the literature. No scoping review has ever addressed this issue. This study aims to identify major research trends on inequality and climate change in the Mediterranean c. 4000 BC–AD 500. It also pinpoints current research gaps on the topic and nascent areas of enquiry. (2) Method: We performed a scoping review on JSTOR, Scopus, Google Scholar and PubMed in December 2025–January 2026. A modified version of the PRISMA-ScR protocol was followed. We sampled journal articles, book chapters, edited volumes and monographs published between 2015 and 2025 which matched the search and inclusion criteria. Additional searches were done on Google Scholar in February 2026 to expand upon emerging research trends relevant to our topic but largely absent from the scoping review. We manually extracted, charted, analysed and synthesised the data. (3) Results: A total of 154 studies were eligible for the scoping review. We identified six research trends prominent in the sampled literature: 1. the rise and fall of world systems, macroscale causal links, and collapse research; 2. inequality, subalternity, and marginality; 3. agriculture, crops, and diet; 4. natural resource management, and water supply; 5. epistemology and methodology; and 6. natural archives and climate proxy datasets. We also recognised the following research gaps or topics that were comparatively less addressed: collapse research applied to the microscale level and marginalised communities; isotope analysis applied to both climate change and inequality in the same study; biomedical approaches applied to both climate change and inequality in the same study; social marginality as a complex construct in human–climate interactions; and the environmental and climate dimensions of the early Roman expansion, especially regarding marginality and the microscale. Finally, we identified artificial intelligence (AI), Big Data, environmental and climate activism, and the perception of climate hazards by subaltern communities as nascent topics of interest that might rise to prominence in the future. (4) Conclusions: We identified major research trends and gaps on climate change and inequality in the ancient Mediterranean in literature published 2015–2025. We also recognised nascent or unexplored topics. The review is intended as a benchmark for developing novel research on the cutting-edge of Mediterranean Archaeology. Full article

Background: Coronary heart disease (CHD) remains a leading cause of morbidity and mortality worldwide. Mitochondria-associated endoplasmic reticulum membranes (MAMs) have recently emerged as critical mediators in cardiovascular pathophysiology; however, their specific contributions to CHD pathogenesis remain largely unexplored. Objective: This study aimed to identify and validate MAM-related biomarkers in CHD through integrated analysis of transcriptomic sequencing data and Mendelian randomization, and to elucidate their underlying mechanisms. Methods: We analyzed two gene expression microarray datasets (GSE113079 and GSE42148) and one genome-wide association study (GWAS) dataset (ukb-d-I9_CHD) to identify differentially expressed genes (DEGs) associated with CHD. MAM-related DEGs were filtered using weighted gene co-expression network analysis (WGCNA). Functional enrichment analysis, Mendelian randomization, and machine learning algorithms were employed to identify biomarkers with direct causal relationships to CHD. A diagnostic model was constructed to evaluate the clinical utility of the identified biomarkers. Additionally, we validated the two hub genes in peripheral blood samples from CHD patients and normal controls, as well as in aortic tissue samples from a low-density lipoprotein receptor-deficient (LDLR−/−) atherosclerosis mouse model. Results: We identified 4174 DEGs, from which 3326 MAM-related DEGs (DE-MRGs) were further filtered. Mendelian randomization analysis coupled with machine learning identified two biomarkers, DHX36 and GPR68, demonstrating direct causal relationships with CHD. These biomarkers exhibited excellent diagnostic performance with areas under the receiver operating characteristic (ROC) curve exceeding 0.9. A molecular interaction network was constructed to reveal the biological pathways and molecular mechanisms involving these biomarkers. Furthermore, validation using peripheral blood from CHD patients and aortic tissues from the Ldlr−/− atherosclerosis mouse model corroborated these findings. Conclusions: This study provides evidence supporting a mechanistic link between MAM dysfunction and CHD pathogenesis, identifying candidate biomarkers that have the potential to serve as diagnostic tools and therapeutic targets for CHD. While the validated biomarkers offer valuable insights into the molecular pathways underlying disease development, additional studies are needed to confirm their clinical relevance and therapeutic potential in larger, independent cohorts. Full article

Digital telecommunications have become the backbone of modern healthcare, transforming how patients and professionals interact, share information, and deliver treatment. The integration of telecommunications with medicine, biomedical engineering and health services has enabled rapid growth in telemedicine, remote patient monitoring, wearable biomedical devices, and data-driven clinical decision-making. Emerging technologies such as artificial intelligence, big data analytics, virtual and augmented reality and robotic tele-surgery are further expanding the scope of digital health. This review provides a comprehensive overview of the role of telecommunications in medicine and biomedical engineering. We classify key applications, highlight enabling technologies and critically examine the challenges regarding interoperability, data security, latency, and cost. Finally, we discuss future directions, including 5G/6G networks, edge computing, and privacy-preserving medical AI, emphasizing the need for reliable and equitable access to telecommunications-enabled healthcare worldwide. Full article

Background: Transcatheter aortic valve replacement (TAVR) has emerged as a pivotal minimally invasive interventional therapy for aortic valve disease and has seen increasingly widespread clinical adoption in recent years. Despite its overall safety, the adverse events and even deaths in the postoperative period still account for a certain percentage. Accurate identification of high-risk patients is therefore critical for optimizing preoperative decision making, guiding individualized treatment strategies and improving long-term outcomes. However, existing scoring systems and predictive models fail to fully leverage multimodal clinical data from patients, resulting in suboptimal predictive accuracy that falls short of the demands of precision medicine, indicating substantial room for improvement. Methods: In this study, a multimodal deep learning model named MULTINet (multimodal learning for TAVR risk network) was constructed using data from the MIMIC-IV (Medical Information Mart for Intensive Care) cohort. This model achieved unimodal and multimodal modeling through a dual-branch structure, and, by using an attention pooling fusion module, flexibly handled the input that contained missing modalities, to predict the 30-day all-cause mortality in TAVR patients. The area under the receiver operating characteristic curve (AUC), the area under the precision–recall curve (AUPR) and the recall rate were used for prediction evaluation. The calibration degree was evaluated by calibration diagrams and Brier scores, and its clinical practicability was assessed through decision curve analysis (DCA). And the integrated gradient method was used to identify key predictive features to enhance interpretability of the model. Results: In the postoperative 30-day all-cause mortality prediction task, the MULTINet method achieved an AUC value of 0.9153, AUPR value of 0.5708 and Recall value of 0.8051, which was significantly superior to the XGBoost method (AUC 0.8958, AUPR 0.4053 and Recall 0.7793) and the MedFuse method (AUC 0.5571, AUPR 0.2487 and Recall 0.3089). The MULTINet method demonstrated more robust and reliable probability estimation performance, with a Brier score of 0.0269, outperforming XGBoost (0.0343) and MedFuse (0.2496). It achieved a higher net benefit in decision analysis, reflecting its effectiveness in strategy optimization and actual decision-making benefits. The renal function, cardiac function and inflammation-related indicators contributed greatly in the prediction process. Conclusions: The multimodal deep learning model proposed in this study named MULTINet enables adaptive integration of multimodal clinical information for predicting all-cause mortality within 30 days post-TAVR, substantially improving both predictive accuracy and clinical applicability, providing robust support for clinical decision making and boosting TAVR management toward greater precision and intelligence. Full article

Advantages in CMOS MOSFET-based electronics served as a basis for modern ubiquitous computerization. At the same time, theoretical and practical developments in material science, analytical chemistry and molecular biology have presented the possibility of applying Boolean logic and information theory findings on a molecular basis. Molecular computing, both organic and inorganic, has the advantages of high computational density, scalability, energy efficiency and parallel computing. Carbon-based and carbohydrate molecular machines are potentially biocompatible and well-suited for biomedical tasks. Molecular computing-enabled sensors, medication-delivery molecular machines, and diagnostic and therapeutic nanobots are at the cutting edge of medical research. Highly focused diagnostics, precision medicine, and personalized treatment can be achieved with molecular computing tools and machinery. At the same time, traditional electronics and AI advancements create a highly effective computerized environment for analyzing big data, assist in diagnostics with sophisticated pattern recognition and step in as a medical routine aid. The combination of the advantages of MOSFET-based electronics and molecular computing creates an opportunity for next-generation healthcare. Full article

Organs-on-Chips (OoC) technology has begun to be considered a pragmatic tool for drug evaluation, offering researchers an opportunity to move beyond the less physiologically relevant animal models. OoCs are microfluidic structures that imitate the functionalities of individual human organs, serving as mimicry tools for drug response and reproducibility studies. On the one hand, companies producing OoCs find managing and analyzing the large amounts of data generated challenging. This is where artificial intelligence (AI) can be deployed to address such problems. This paper will present the state-of-the-art of current OoC technology and AI, discussing the benefits and threats of combining these approaches. AI can be applied to optimize the process of OoC fabrication and operation, as well as for the big data analysis of OoC devices. By combining these technologies, scientists gain a powerful tool for drug development that is more efficient and accurate. However, processing the vast datasets generated by OoC systems often requires specialized AI expertise and computational resources. Despite the numerous possible benefits of amalgamating OoC technology with AI, several challenges and limitations need to be addressed. The large datasets generated by OoC systems can be difficult to process and analyze, which is a task that may require specialized AI expertise. Additionally, limitations of OoC systems include issues with reproducibility, as the devices are sensitive to perturbations in experimental conditions. Furthermore, the development and implementation of AI algorithms require significant computational resources and expertise, which may not be readily available to all research institutions. To overcome these challenges, interdisciplinary collaboration between biologists, engineers, data scientists, and AI experts is essential. Continued advancements in both OoC technology and AI will likely lead to more robust and versatile platforms for biomedical research and drug development, ultimately contributing to the advancement of personalized medicine and the reduction of reliance on animal testing. Full article