Search Results (31)

Recent advances in cardiovascular imaging have revolutionized the assessment and management of abdominal aortic aneurysm (AAA) through the integration of sophisticated haemodynamic biomarkers. This comprehensive review evaluates the clinical utility and mechanistic significance of multiple biomarkers in AAA pathogenesis, progression, and treatment outcomes. Advanced cardiac imaging modalities, including four-dimensional magnetic resonance imaging (4D MRI), computational fluid dynamics (CFD), and specialized echocardiography, enable precise quantification of critical haemodynamic parameters. Wall shear stress (WSS) emerges as a fundamental biomarker, with values below 0.4 Pa indicating pathological conditions and increased risk for aneurysm progression. Time-averaged wall shear stress (TAWSS), typically maintaining values above 1.5 Pa in healthy arterial segments, provides crucial information about sustained haemodynamic forces affecting the vessel wall. The oscillatory shear index (OSI), ranging from 0 (unidirectional flow) to 0.5 (purely oscillatory flow), quantifies directional changes in WSS during cardiac cycles. In AAA, elevated OSI values between 0.3 and 0.4 correlate with disturbed flow patterns and accelerated disease progression. The relative residence time (RRT), combining TAWSS and OSI, identifies regions prone to thrombosis, with values exceeding 2–3 Pa⁻¹ indicating increased risk. The endothelial cell activation potential (ECAP), calculated as OSI/TAWSS, serves as an integrated metric for endothelial dysfunction risk, with values above 0.2–0.3 Pa⁻¹ suggesting increased inflammatory activity. Additional biomarkers include the volumetric perivascular characterization index (VPCI), which assesses vessel wall inflammation through perivascular tissue analysis, and pulse wave velocity (PWV), measuring arterial stiffness. Central aortic systolic pressure and the aortic augmentation index provide essential information about cardiovascular load and arterial compliance. Novel parameters such as particle residence time, flow stagnation, and recirculation zones offer detailed insights into local haemodynamics and potential complications. Implementation challenges include the need for specialized equipment, standardized protocols, and expertise in data interpretation. However, the potential for improved patient outcomes through more precise risk stratification and personalized treatment planning justifies continued development and validation of these advanced assessment tools. Full article

Background/Objectives: Thyroid disorders such as Hashimoto’s thyroiditis and Graves’ disease represent major endocrine challenges worldwide, often requiring long-term management. Recently, nutritional supplementation with selenium and myo-inositol has been proposed as a supportive strategy. This review aims to summarize the current evidence regarding their efficacy in improving thyroid function, reducing thyroid autoantibodies in Hashimoto’s disease, and restoring biochemical euthyroidism in Graves’ disease. Methods: A narrative review of the available literature was undertaken, concentrating on randomized controlled trials and observational studies evaluating selenium and myo-inositol, alone or in combination (MYO+Se), in patients with autoimmune thyroid disorders and benign thyroid nodules. Search Strategy and Study Selection: We searched MEDLINE/PubMed, Embase, Cochrane CENTRAL, and Scopus from inception to 31 July 2025. The search used Boolean operators to combine the following keywords: (“selenium” OR “selenomethionine”) AND (“myo-inositol” OR “inositol”) AND (thyroid OR Hashimoto* OR Graves’ OR hyperthyroid* OR hypothyroid* OR nodule* OR goiter OR orbitopathy). We included human studies in English. Inclusion criteria: Research designs include RCTs, quasi-experimental studies, cohort/case-control studies, and big case series (n ≥ 30). Exclusion criteria: Animal-only or in vitro studies (unless mechanistic), pediatric case reports, and editorials/commentaries. Study selection and data extraction: Two reviewers screened independently; discrepancies were settled through consensus. The data retrieved included the population, baseline iodine/selenium status (if reported), dose/formulation, treatment duration, outcomes (TSH, FT4, FT3, TPOAb, TgAb, TRAb, nodule metrics), and adverse events. Quality assessment: The risk of bias was assessed using the RoB-2 for RCTs and the Newcastle-Ottawa Scale or JBI checklists for observational studies. A qualitative synthesis emphasized study quality, consistency, directness, and accuracy. Results: Clinical research indicate that selenium supplementation may reduce thyroid peroxidase antibody (TPOAb) levels in Hashimoto’s disease, thereby attenuating autoimmune activity. Myo-inositol, particularly when combined with selenium, has been proven to improve thyroid hormone profiles while also lowering nodule size or growth. In Graves’ disease, supplementation has been linked to the restoration of biochemical euthyroidism in certain patients, albeit data are limited. Despite these encouraging results, diversity in trial design, treatment length, and dosages restrict the robustness of existing recommendations. Conclusions: Selenium and myo-inositol supplementation have shown promise as adjuvant treatments for autoimmune thyroid diseases and benign thyroid nodules. However, further large-scale, well-designed clinical trials are needed to determine the appropriate dosages, treatment duration, and patient selection criteria. Personalized supplementation solutions may improve medication efficacy and help with more comprehensive thyroid disease care. Full article

Multiple sclerosis (MS) is a chronic neurological disease often resulting in motor and autonomic dysfunction. This case report investigates the acute and subacute effects of the EXOPULSE Mollii Suit (EMS), a wearable device capable of delivering transcutaneous electrical nerve stimulation to multiple anatomical regions, in a 43-year-old woman with MS. The patient underwent a clinical evaluation before the EMS treatment, during which central nervous system (CNS) and autonomic nervous system (ANS) responses were monitored using electroencephalography (EEG), heart rate variability (HRV), and infrared thermography (IRT). Immediately after the first EMS application, the clinical evaluation was repeated. The intervention continued at home for one month, followed by a post-treatment evaluation similar to the pre-intervention assessment. Functional evaluations showed improvements in sit-to-stand performance (from 8 s to 6 s), muscle tone (MAS scale for the right side from 3 to 2 and for the left side from 2 to 1), clonus, and spasticity (from 3 to 2). EEG results revealed decreased θ-band power (on average, from 0.394 to 0.253) and microstates’ reorganization. ANS activity modifications were highlighted by both HRV (e.g., RMSSD from 0.118 to 0.0837) and IRT metrics (e.g., nose tip temperature sample entropy from 0.090 to 0.239). This study provides the first integrated analysis of CNS and ANS responses to EMS in an MS patient, combining functional scales with multimodal instrumental measurements, emphasizing the possible advantages EMS for MS treatment. Although preliminary, these results demonstrated the potentiality of the EMS to deliver effective and personalized rehabilitative interventions for MS patients. Full article

This study explores the application of federated learning (FL) in security camera surveillance systems to overcome the structural limitations inherent in traditional centralized artificial intelligence (AI) training approaches, while simultaneously enhancing operational efficiency and data security. Conventional centralized AI models require the transmission of raw surveillance data from individual security camera units to a central server for model training, which poses significant challenges, including network congestion, a heightened risk of personal data leakage, and inadequate adaptation to localized environmental characteristics. These limitations are particularly critical in high-security environments such as military bases and government facilities, where reliability and real-time processing are paramount. In contrast, FL enables decentralized training by retaining data on local devices and sharing only model parameters with a central aggregator, thereby improving privacy preservation, reducing communication overhead, and facilitating adaptive, context-aware learning. This paper does not present a new federated learning algorithm or original experiment. Instead, it synthesizes existing research findings and applies the Analytic Hierarchy Process (AHP) to evaluate and prioritize critical factors for deploying FL in surveillance systems. By combining literature-based evidence with structured expert judgment, this study provides practical guidelines for real-world application. This paper identifies four key performance metrics—detection accuracy, false alarm rate, response time, and network load—and conducts a comparative analysis of FL and centralized AI-based approaches in the recent literature. In addition, the AHP is employed to evaluate expert survey data, quantitatively prioritizing eight critical factors for effective FL implementation. The results highlight detection accuracy and data security as the most significant concerns, indicating that FL presents a promising solution for future smart surveillance infrastructures. This research contributes to the advancement of AI-powered surveillance systems that are both high-performing and resilient under stringent privacy and operational constraints. Full article

Objectives: This study aimed to evaluate the safety and efficacy of the eight-chop technique in cataract surgery in patients with pseudoexfoliation (PEX) syndrome and assess the intraoperative parameters, changes in corneal endothelial cells, intraocular pressure (IOP), and intraoperative complications. Methods: This technique was applied in patients with and without PEX syndrome. Preoperative and postoperative assessments were conducted on best-corrected visual acuity, IOP, corneal endothelial cell density (CECD), coefficient of variation, percentage of hexagonal cells, and central corneal thickness. Intraoperative recordings included operative time, phaco time, aspiration time, cumulative dissipated energy (CDE), and fluid of volume used. Results: We analyzed 150 eyes from 150 patients (mean age, 75.5 ± 5.7 years; 59 men, 91 women). In the PEX group, operative time, phaco time, aspiration time, CDE, and volume of fluid used were 6.7 min, 17.4 s, 85.2 s, 6.91 µJ, and 33.4 mL, respectively, demonstrating favorable surgical metrics. On the other hand, in the control group, operative time, phaco time, aspiration time, CDE, and volume of fluid used were 4.5 min, 14.3 s, 64.0 s, 5.83 µJ, and 25.5 mL, respectively. In addition, CECD losses were 3.7% at week 7 and 2.7% at week 19 in the PEX group and 2.7% and 1.6%, respectively, in the control group. Significant decreases were observed at 7 and 19 weeks postoperatively in the PEX and control groups. No eye in the PEX group required a capsular tension ring due to zonular dialysis. Conclusions: The eight-chop technique in cataract surgery demonstrates excellent intraoperative parameters in patients with PEX, is effective against zonular weakness, and does not require the use of a capsular tension ring. This technique will aid in establishing personalized treatment strategies and improve cataract management and treatment. Full article

Genome-wide association studies (GWASs) play a central role in precision medicine, powering a range of clinical applications from pharmacogenomics to disease risk prediction. A critical component of GWASs is genotype imputation, a computational method used to infer untyped genetic variants. While imputation increases variant coverage by estimating genotypes at untyped loci, this expanded coverage can enhance the ability to detect genetic associations in some cases. However, imputation also introduces biases, particularly for rare variants and underrepresented populations, which may compromise clinical accuracy. This review examines the challenges and clinical implications of genotype imputation errors, including their impact on therapeutic decisions and predictive models, like polygenic risk scores (PRSs). In particular, the sources of imputation errors have been deeply explored, emphasizing the disparities in performance across ancestral populations and downstream effects on healthcare equity and addressing ethical considerations surrounding the access to equitable genomic resources. Based on the above, we propose evidence-based best practices for clinical GWAS implementation, including the direct genotyping of clinically actionable variants, the cross-population validation of imputation models, the transparent reporting of imputation quality metrics, and the use of ancestry-matched reference panels. As genomic data becomes increasingly adopted in healthcare systems worldwide, ensuring the accuracy and inclusivity of GWAS-derived insights is paramount. Here, we suggest a framework for the responsible clinical integration of imputed genetic data, paving the way for more reliable and equitable personalized medicine. Full article

Background and Objective: Dysregulated tyrosine kinase signaling is a central driver of tumorigenesis, metastasis, and therapeutic resistance. While tyrosine kinase inhibitors (TKIs) have revolutionized targeted cancer treatment, identifying compounds with optimal bioactivity remains a critical bottleneck. This study presents a robust machine learning framework—leveraging deep artificial neural networks (dANNs), convolutional neural networks (CNNs), and structural molecular fingerprints—to accurately predict TKI bioactivity, ultimately accelerating the preclinical phase of drug development. Methods: A curated dataset of 28,314 small molecules from the ChEMBL database targeting 11 tyrosine kinases was analyzed. Using Morgan fingerprints and physicochemical descriptors (e.g., molecular weight, LogP, hydrogen bonding), ten supervised models, including dANN, SVM, CatBoost, and CNN, were trained and optimized through a randomized hyperparameter search. Model performance was evaluated using F1-score, ROC–AUC, precision–recall curves, and log loss. Results: SVM achieved the highest F1-score (87.9%) and accuracy (85.1%), while dANNs yielded the lowest log loss (0.25096), indicating superior probabilistic reliability. CatBoost excelled in ROC–AUC and precision–recall metrics. The integration of Morgan fingerprints significantly improved bioactivity prediction across all models by enhancing structural feature recognition. Conclusions: This work highlights the transformative role of machine learning—particularly dANNs and SVM—in rational drug discovery. By enabling accurate bioactivity prediction, our model pipeline can effectively reduce experimental burden, optimize compound selection, and support personalized cancer treatment design. The proposed framework advances kinase inhibitor screening pipelines and provides a scalable foundation for translational applications in precision oncology. By enabling early identification of bioactive compounds with favorable pharmacological profiles, the results of this study may support more efficient candidate selection for clinical drug development, particularly in regards to cancer therapy and kinase-associated disorders. Full article

Background/Objectives: The brain serves as the central command center for the nervous system in the human body and is made up of nerve cells known as neurons. When these nerve cells grow rapidly and abnormally, it can lead to the development of a brain tumor. Brain tumors are severe conditions that can significantly reduce a person’s lifespan. Failure to detect or delayed diagnosis of brain tumors can have fatal consequences. Accurately identifying and classifying brain tumors poses a considerable challenge for medical professionals, especially in terms of diagnosing and treating them using medical imaging analysis. Errors in diagnosing brain tumors can significantly impact a person’s life expectancy. Magnetic Resonance Imaging (MRI) is highly effective in early detection, diagnosis, and classification of brain cancers due to its advanced imaging abilities for soft tissues. However, manual examination of brain MRI scans is prone to errors and heavily depends on radiologists’ experience and fatigue levels. Swift detection of brain tumors is crucial for ensuring patient safety. Methods: In recent years, computer-aided diagnosis (CAD) systems incorporating deep learning (DL) and machine learning (ML) technologies have gained popularity as they offer precise predictive outcomes based on MRI images using advanced computer vision techniques. This article introduces a novel hybrid CAD approach named ViT-PCA-RF, which integrates Vision Transformer (ViT) and Principal Component Analysis (PCA) with Random Forest (RF) for brain tumor classification, providing a new method in the field. ViT was employed for feature extraction, PCA for feature dimension reduction, and RF for brain tumor classification. The proposed ViT-PCA-RF model helps detect early brain tumors, enabling timely intervention, better patient outcomes, and streamlining the diagnostic process, reducing patient time and costs. Our research trained and tested on the Brain Tumor MRI (BTM) dataset for multi-classification of brain tumors. The BTM dataset was preprocessed using resizing and normalization methods to ensure consistent input. Subsequently, our innovative model was compared against traditional classifiers, showcasing impressive performance metrics. Results: It exhibited outstanding accuracy, specificity, precision, recall, and F1 score with rates of 99%, 99.4%, 98.1%, 98.1%, and 98.1%, respectively. Conclusions: Our innovative classifier’s evaluation underlined our model’s potential, which leverages ViT, PCA, and RF techniques, showing promise in the precise and effective detection of brain tumors. Full article

The complexity of large-scale group decision-making (LSGDM) in the digital society is becoming increasingly prominent. How to achieve efficient consensus through social networks (SNs) has become a core challenge in improving the decision quality. First, conventional clustering methods often rely on a single-distance metric, neglecting both numerical assessments and preference rankings. Second, ensuring the decision authenticity requires considering diverse behaviors, such as trust propagations, risk preferences, and minority opinion expressions, for scientific decision-making in SNs. To address these challenges, a consensus-reaching process (CRP) method based on an enhanced K-means++ clustering is proposed. The above method not only focuses on minority opinion handling (MOH), but also incorporates decision indicator sets (DISs) to analyze the degree of opinion divergences within groups. First, the Hamacher aggregation operator with a decay factor completes trust matrices, improving the trust representation. Second, a personalized distance metric that combines cardinal distances with ordinal distances is incorporated into the enhanced K-means++ clustering, enabling more precise clustering. Third, weights for decision-makers (DMs) and subgroups are determined based on trust levels and degree centrality indices. Fourth, minority opinions are appropriately handled via considering the diverse backgrounds and expertise of DMs, leveraging a difference-oriented DIS to detect and adjust these opinions via weight modifications until a consensus is reached. Fifth, the alternative ranking is objectively generated via DIS scores derived from multigranulation rough approximations. Finally, the feasibility of the proposed method is validated via a case study on unmanned aerial vehicle (UAV) selection using online reviews, supported by a sensitivity analysis and comparative experiments demonstrating superior performances over existing methods. The result shows that the proposed model can enhance clustering accuracies with hybrid distances, objectively measure the consensus via DISs, handle minority opinions effectively, and improve LSGDM’s overall efficiencies. Full article

Severe acute respiratory symptom coronavirus 2 (SARS-CoV-2) infection occurs via the attachment of the spike (S) protein’s receptor binding domain (RBD) to human ACE2 (hACE2). Natural polymorphisms in hACE2, particularly at the interface, may alter RBD–hACE2 interactions, potentially affecting viral infectivity across populations. This study identified the effects of six naturally occurring hACE2 polymorphisms with high allele frequency in the African population (S19P, K26R, M82I, K341R, N546D and D597Q) on the interaction with the S protein RBD of the BA.4/5 Omicron sub-lineage through post-molecular dynamics (MD), inter-protein interaction and dynamic residue network (DRN) analyses. Inter-protein interaction analysis suggested that the K26R variation, with the highest interactions, aligns with reports of enhanced RBD binding and increased SARS-CoV-2 susceptibility. Conversely, S19P, showing the fewest interactions and largest inter-protein distances, agrees with studies indicating it hinders RBD binding. The hACE2 M82I substitution destabilized RBD–hACE2 interactions, reducing contact frequency from 92 (WT) to 27. The K341R hACE2 variant, located distally, had allosteric effects that increased RBD–hACE2 contacts compared to WThACE2. This polymorphism has been linked to enhanced affinity for Alpha, Beta and Delta lineages. DRN analyses revealed that hACE2 polymorphisms may alter the interaction networks, especially in key residues involved in enzyme activity and RBD binding. Notably, S19P may weaken hACE2–RBD interactions, while M82I showed reduced centrality of zinc and chloride-coordinating residues, hinting at impaired communication pathways. Overall, our findings show that hACE2 polymorphisms affect S BA.4/5 RBD stability and modulate spike RBD–hACE2 interactions, potentially influencing SARS-CoV-2 infectivity—key insights for vaccine and therapeutic development. Full article

Federated learning (FL) has emerged as a transformative framework for collaborative learning, offering robust model training across institutions while ensuring data privacy. In the context of making a COVID-19 diagnosis using lung imaging, FL enables institutions to collaboratively train a global model without sharing sensitive patient data. A central manager aggregates local model updates to compute global updates, ensuring secure and effective integration. The global model’s generalization capability is evaluated using centralized testing data before dissemination to participating nodes, where local assessments facilitate personalized adaptations tailored to diverse datasets. Addressing data heterogeneity, a critical challenge in medical imaging, is essential for improving both global performance and local personalization in FL systems. This study emphasizes the importance of recognizing real-world data variability before proposing solutions to tackle non-independent and non-identically distributed (non-IID) data. We investigate the impact of data heterogeneity on FL performance in COVID-19 lung imaging across seven distinct heterogeneity settings. By comprehensively evaluating models using generalization and personalization metrics, we highlight challenges and opportunities for optimizing FL frameworks. The findings provide valuable insights that can guide future research toward achieving a balance between global generalization and local adaptation, ultimately enhancing diagnostic accuracy and patient outcomes in COVID-19 lung imaging. Full article

Objectives: This study investigates disruptions in functional brain networks in Parkinson’s Disease (PD), using advanced modeling and machine learning. Functional networks were constructed using the Nonlinear Autoregressive Distributed Lag (NARDL) model, which captures nonlinear and asymmetric dependencies between regions of interest (ROIs). Key network metrics and information-theoretic measures were extracted to classify PD patients and healthy controls (HC), using deep learning models, with explainability methods employed to identify influential features. Methods: Resting-state fMRI data from the Parkinson’s Progression Markers Initiative (PPMI) dataset were used to construct NARDL-based networks. Metrics, such as Degree, Closeness, Betweenness, and Eigenvector Centrality, along with Network Entropy and Complexity, were analyzed. Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Long Short-Term Memory (LSTM) models, classified PD and HC groups. Explainability techniques, including SHAP and LIME, identified significant features driving the classifications. Results: PD patients showed reduced Closeness (22%) and Betweenness Centrality (18%). CNN achieved 91% accuracy, with Network Entropy and Eigenvector Centrality identified as key features. Increased Network Entropy indicated heightened randomness in PD brain networks. Conclusions: NARDL-based analysis with interpretable deep learning effectively distinguishes PD from HC, offering insights into neural disruptions and potential personalized treatments for PD. Full article

Schizophrenia is a complex neuropsychiatric disorder characterized by disruptions in brain connectivity and cognitive functioning. Continuous monitoring of neural activity is essential, as it allows for the detection of subtle changes in brain connectivity patterns, which could provide early warnings of cognitive decline or symptom exacerbation, ultimately facilitating timely therapeutic interventions. This paper proposes a novel approach for detecting schizophrenia-related abnormalities using deep learning (DL) techniques applied to electroencephalogram (EEG) data. Using an openly available EEG dataset on schizophrenia, the focus is on preprocessed event-related potentials (ERPs) from key electrode sites and applied transfer entropy (TE) analysis to quantify the directional flow of information between brain regions. TE matrices were generated to capture neural connectivity patterns, which were then used as input for a hybrid DL model, combining convolutional neural networks (CNNs) and Bidirectional Long Short-Term Memory (BiLSTM) networks. The model achieved a performant accuracy of 99.94% in classifying schizophrenia-related abnormalities, demonstrating its potential for real-time mental health monitoring. The generated TE matrices revealed significant differences in connectivity between the two groups, particularly in frontal and central brain regions, which are critical for cognitive processing. These findings were further validated by correlating the results with EEG data obtained from the Muse 2 headband, emphasizing the potential for portable, non-invasive monitoring of schizophrenia in real-world settings. The final model, integrated into the NeuroPredict platform, offers a scalable solution for continuous mental health monitoring. By incorporating EEG data, heart rate, sleep patterns, and environmental metrics, NeuroPredict facilitates early detection and personalized interventions for schizophrenia patients. Full article

Modern e-businesses heavily rely on advanced data analytics for product recommendations. However, there are still untapped opportunities to enhance user interfaces. Currently, online stores offer a single-page version to all customers, overlooking individual characteristics. This paper aims to identify the essential components and present a framework for enabling multiple e-commerce user interfaces. It also seeks to address challenges associated with personalized e-commerce user interfaces. The methodology includes detailing the framework for serving diverse e-commerce user interfaces and presenting pilot implementation results. Key components, particularly the role of algorithms in personalizing the user experience, are outlined. The results demonstrate promising outcomes for the implementation of the pilot solution, which caters to various e-commerce user interfaces. User characteristics support multivariant websites, with algorithms facilitating continuous learning. Newly proposed metrics effectively measure changes in user behavior resulting from different interface deployments. This paper underscores the central role of personalized e-commerce user interfaces in optimizing online store efficiency. The framework, supported by machine learning algorithms, showcases the feasibility and benefits of different page versions. The identified components, challenges, and proposed metrics contribute to a comprehensive solution and set the stage for further development of personalized e-commerce interfaces. Full article

Social media platforms allow people to share information, connect, and build networks at an unprecedented scale with positive and negative consequences. Social network analysis (SNA) applies mathematical network and graph theory to visualise information transfer as relational networks of connected nodes. Measuring node connectivity (centrality) permits the identification of ‘influencers’. SNA has been applied to analyse the spread of misinformation on Twitter (1), but to date, no research has examined nutrition networks. Therefore, this study examined the #Nutrition conversations on Twitter utilising SNA and linguistic analyses. English language tweets including ‘#Nutrition’ on 1–21 March 2023 were collected using the SNA tool, NodeXL Pro (Network Overview for Discovery and Exploration in Excel) (2). SNA is a multistep process that calculates graph metrics and develops a network graph to measure the relationships between users. SNA also identifies semantically related words, hashtags, and word pairs and identifies the sentiment of words used, as measured against the Opinion Lexicon (2). The #Nutrition network included 17,129 vertices (users) with 26,809 unique edges (connections); edges with duplicates were merged. The network density was low, suggesting that most users communicate heavily with a small number of users. The average geodesic distance between any two users was 5.26, revealing a dispersed online discussion. SNA identified the top 10 influencers in this network, measured by high betweenness centrality (23,375,543–5,207,998). Influential users were from a mix of accounts including personal, online blogs, and government organisations. High betweenness centrality identified the users with the greatest influence, acting as bridges between network groups and therefore amplifying #Nutrition messages. Sentiment analysis found the discourse was more positive (0.047, 22,218 words) than negative (0.015, 6795 words). Semantic analysis calculated the total words, 468,191, and identified the most frequently used words in the tweets: #nutrition, #health, food, more, nutrition, health, #diet, #healthylifestlye, #fitness, and #food. Social network analysis shows the discourse on Twitter relating to #Nutrition is dispersed without clear polarising views. Semantic analysis showed that ‘health’ was the main topic discussed in relation to nutrition in this network and was most frequently associated with #Nutrition. The narrative was positively framed, as identified through sentiment analysis. Full article