Southwest Minzu University

Purpose In an era of intensifying climate change, geopolitical conflicts, and economic volatility, ensuring the stability of food supply chains is essential for food security. This study investigates a critical yet underexplored question: When subjected to diverse external shocks, do global food trade networks possess the necessary resilience and adaptability to maintain functional import channels for nations? Design/methodology/approach Utilizing trade data from 2000 to 2022, we construct and analyze four distinct global food trade networks: wheat and meslin, maize, rice, and a composite total food network. We employ a novel, multi-dimensional analytical framework that integrates the structural evolution of networks, node-level resilience, and structural resilience. The robustness of these networks is rigorously tested through simulations of both random and deliberate strategies, targeted attacks on core nodes. Findings Our analysis reveals that while global food trade networks are expanding and becoming more interconnected, their resilience is unevenly distributed. Countries like Japan, South Korea, Iran, and China are reliant on imported food and exhibit trade vulnerability due to concentrated import sources. Major food traders, including Canada, Mexico, China, and South Korea, serve as hub nodes of the network, alongside smaller trading nations like South Africa, Denmark, and Belgium. Among the four types of food trade networks, network connectivity is strongest for wheat and meslin, whereas the rice trade network displays the highest structural stability. Deliberate attacks are more detrimental to network resilience than random attacks, with the failure of core nodes such as France and Germany significantly exposing a critical systemic vulnerability. Originality/value Moving beyond prevailing research that often focuses on single shock types or aggregate networks, this paper provides a nuanced, multi-network and multi-strategy analysis of trade resilience. It introduces an integrated node-structure framework for assessing systemic risk. The findings offer actionable insights for policymaker, emphasizing the urgency of diversifying import sources, strengthening regional trade cooperation, and developing differentiated contingency plans based on specific network roles and shock characteristics.

With the development of society, the incidence of cardiovascular diseases has continued to rise. As an important non-invasive diagnostic tool, electrocardiography (ECG) has been widely used in clinical screening and auxiliary diagnosis. In recent years, deep learning-based automated ECG analysis methods have attracted considerable attention; however, most existing approaches rely on single time-domain features, making it difficult to fully exploit the discriminative information of ECG signals in both time and frequency domains. Therefore, this study proposes a Frequence-Enhance ECG Network (FEENet), which performs multi-scale feature extraction on both temporal ECG signals and their frequency-domain representations. A Transformer-based causal encoder (TCE) is then employed to model the relationships among features at different scales. Subsequently, a time–frequency cross-attention (TFCA) module is introduced to enable bidirectional interaction and deep fusion between temporal and frequency-domain features. Finally, a classification layer is used to produce accurate cardiovascular disease classification results. The proposed method achieves an accuracy of 89.6% and an F1-score of 77.4% on the PTB-XL dataset, demonstrating its effectiveness and strong classification capability in complex ECG classification tasks.

Freeze–thaw cycles on the Qinghai–Tibetan Plateau inhibit microbial activity and challenge silage preservation. This paper aimed to elucidate how an indigenous, freeze–thaw-resistant Lactobacillus plantarum strain (LP160) improves oat silage quality under such stress. Oats were ensiled for 60 days under constant 20 °C (t) or freeze–thaw cycles (12 h at 20 °C/−5 °C; s) with or without LP160 inoculation. Samples after ensiling and 5-day aerobic exposure were analyzed for fermentation parameters, nutrients, microbiome, and non-targeted metabolomics using liquid chromatography–tandem mass spectrometry (LC-MS/MS). LP160 inoculation improved silage quality, as shown by the lower pH, ammoniacal nitrogen, neutral detergent fiber, acid detergent fiber contents as well as the greater amount of lactic acid. Key findings demonstrated that LP160 inoculation significantly enhanced Lactobacillus dominance, effectively curbed the growth of detrimental bacteria like Mucor, and regulated the microbial structure. During the aerobic exposure phase, the microbial community structures and successions varied under different temperature treatments. When inoculated under freeze–thaw conditions, the genus Bacillus increased, while Paenibacillus was not impeded. A total of 943 metabolites were identified, predominantly comprising amino acids, fatty acids, and the like. The expressions of metabolites with antioxidant and antibacterial properties were upregulated with LP160 inoculation. This led to the inhibition of protein hydrolysis and a reduction in ammonia–nitrogen production. The results of correlation analysis indicated that inoculating LP160 suppressed the proliferation of Mucor and enhanced the abundance of Torulaspora; meanwhile, the expression of L-palmitoylcarnitine involved in the fatty acid degradation pathway and fatty acid metabolism pathway was inhibited along with the generation of ammonia–nitrogen. Consequently, the degradation of fatty acids and proteins was restrained. The results of this paper provided new insights into the silage under freeze–thaw conditions.