Browse Articles
Discover research articles across all indexed journals
Pulmonary carcinogenic effects of subchronic inhalation exposure with long-term follow-up to polyhexamethylene guanidine phosphate in rats
The role of type VI secretion system in resistance and pathogenicity of Acinetobacter baumannii
Synthesis, characterization, and catalytic performance of cobalt and nickel-substituted polyoxometalates organic-inorganic nanohybrids
Green synthesis of copper nanoparticles using Satureja khuzestanica plant and investigation of its antifungal properties: Candida albicans and Candida tropicalis, an in vitro study
Optimal control strategies and parameter estimation with a time delay dengue model using Penang Hospital data
Exploration of the efficacy and multivariate analysis of 5G remote rehabilitation intervention based on peer education model for patients after coronary intervention
Abstract This study aims to investigate the effects of a peer education-based model utilizing 5G remote rehabilitation guidance with wearable smart devices on patients with coronary heart disease undergoing percutaneous coronary intervention (PCI), along with the influence of various cross-factors. A total of 174 patients who underwent PCI between July 2024 and July 2025 were randomly assigned to a control group (n = 87) and a 5G group (n = 87), both receiving 12 months of follow-up. The control group received phase I and II cardiac rehabilitation, personalized exercise plans, and routine follow-up. In addition to this, the 5G group received remote rehabilitation guided by wearable devices. Clinical indicators, cardiopulmonary exercise capacity, psychological status, and the impact of different variables on the risk of adverse cardiovascular events were compared between the two groups. Overall adherence in this study was 87.88% from baseline to the end of the 12-month follow-up. Thecontrol group showed improvements in both LDL-C and LVEF levels post-intervention (P K 0.05). Furthermore, patients receiving 5G intervention demonstrated significant improvements in TC andLVEF ( P < 0.01). Changes in VO2 peak and METs were also significantly greater in the 5G groupcompared to the control group (VO2 peak: d = 0.61, P < 0.001; METs: d = 0.78, P < 0.001). Additionally, the improvement in VO2AT after 5G intervention was significantly superior to that ofconventional intervention (d = 0.22, P = 0.035). Both groups exhibited a significant decrease inGAD-7 scores post-intervention (control group: A = -2.69, P < 0.001; 5G group: A = − 2.99,P K 0.001), with the 5G group showing a greater degree of GAD-7 improvement. For PSQl, the controlgroup showed no significant change post-intervention (A = − 0.96, P = 0.176), while the 5G groupexhibited a significant reduction in PSQl scores (A = − 3.58, P < 0.001). The 5G group also revealedassociations between age, urban resident basic medical insurance, and cardiac function classificationwith the incidence of adverse events ( P < 0.01). Negative binomial regression analysis indicated thatyounger patients and those with better cardiac function had a lower risk of adverse events, and thisprotective effect was more pronounced in the 5G intervention group. Low body weight significantlyincreased the risk of adverse events, and male sex in the 5G intervention group, male gender and smoking were significantly associated with the occurrence of adverse events ( P < 0.05). The 5G remote personalized exercise rehabilitation intervention based on peer education effectively improves cardiopulmonary function and quality of life in patients post-PCI and helps prevent adverse events. This study provides a theoretical basis for the rehabilitation intervention and management of coronary heart disease patients after PCI.
Correction: Multiple objectives escaping bird search optimization and its application in stock market prediction based on transformer model
Earliest evidence of behavioural handedness in the Ediacaran motile bilaterian Spriggina floundersi
Biomimetic Catalytic Atroposelective Ring-Opening Aminolysis of <i>N</i> -Sulfonylated Biaryl Lactams
Periodic Molecular-Level Asymmetric Channels for Synergistical Purification of Iodide Wastewater and Osmotic Power Generation
Mechanism and development of CO2 pre–pad fracturing stimulation for tight conglomerate low–permeability condensate gas reservoirs
Non-canonical ATR signaling mediates direct and bystander cold atmospheric plasma-induced stress responses in A549 lung adenocarcinoma cells
Role of tyrosinase in ocular growth and myopia development in a guinea pig model
Lightweight soil replacement for reducing the collapsibility of loess in permeable sponge-city roads: insights from multi-stratum centrifuge model tests
Stereoretentive Norrish–Yang Photocyclization Mediated by Hydrogen Bonding
A novel method of using neural networks to predict wine composition
A Type II CDK6 Degrader Enables Cellular Targeting beyond the Limits of Type II Inhibition
An explainable AfroXLMR approach for multi-label emotion classification of Amharic social media text with dataset release
Abstract Emotion detection from social media is crucial for understanding human emotions across languages. However, for low-resourced languages such as Amharic, the lack of annotated data makes this task challenging. Additionally, most current models use black-box methods that obscure whether predictions rely on linguistically meaningful cues. To address these gaps, this study proposes a multi-label emotion classification model for Amharic by fine-tuning AfroXLMR. To enhance transparency, we integrate explainable artificial intelligence (XAI) into the framework. We compiled and annotated a new dataset of 22,000 unique social media comments across eight emotion categories for training, validation, and testing. The data was split into 80% for training, 10% for validation, and 10% for testing. The proposed model achieved a recall of 87% and a Hamming loss of 0.08. To interpret its predictions, we applied Local Interpretable Model-agnostic Explanations (LIME). We also evaluated the model against several state-of-the-art baselines, including XLM-R base, mBART, BiLSTM, LSTM, CNN, and AfriBERTa. The results show that our approach outperformed each baseline, achieving F1-score improvements of 5% over XLM-R base, 3% over mBART, 5% over BiLSTM, 7% over LSTM, 9% over CNN, and 2% over AfriBERTa. Bootstrapped statistical significance testing confirms that these improvements are robust and not attributable to random variation. In conclusion, the fine-tuned AfroXLMR model demonstrates promising performance in Amharic multi-label emotion classification. Building on this success, next steps could involve exploring more advanced fine-tuning strategies and expanding our datasets to strengthen both performance and the model’s ability to generalize across diverse Amharic contexts.