The behavior of pigs is intricately tied to their health status, highlighting the critical importance of accurately recognizing pig behavior, particularly abnormal behavior, for effective health monitoring and management. This study addresses the challenge of accommodating frequent non-rigid deformations in pig behavior using deformable convolutional networks (DCN) to extract more comprehensive features by incorporating offsets during training. To overcome the inherent limitations of traditional DCN offset weight calculations, the study introduces the multi-path coordinate attention (MPCA) mechanism to enhance the optimization of the DCN offset weight calculation within the designed DCN-MPCA module, further integrated into the cross-scale cross-feature (C2f) module of the backbone network. This optimized C2f-DM module significantly enhances feature extraction capabilities. Additionally, a gather-and-distribute (GD) mechanism is employed in the neck to improve non-adjacent layer feature fusion in the YOLOv8 network. Consequently, the novel DM-GD-YOLO model proposed in this study is evaluated on a self-built dataset comprising 11,999 images obtained from an online monitoring platform focusing on pigs aged between 70 and 150 days. The results show that DM-GD-YOLO can simultaneously recognize four common behaviors and three abnormal behaviors, achieving a precision of 88.2%, recall of 92.2%, and mean average precision (mAP) of 95.3% with 6.0MB Parameters and 10.0G FLOPs. Overall, the model outperforms popular models such as Faster R-CNN, EfficientDet, YOLOv7, and YOLOv8 in monitoring pens with about 30 pigs, providing technical support for the intelligent management and welfare-focused breeding of pigs while advancing the transformation and modernization of the pig industry.
Tomato leaf disease control in the field of smart agriculture urgently requires attention and reinforcement. This paper proposes a method called LAFANet for image-text retrieval, which integrates image and text information for joint analysis of multimodal data, helping agricultural practitioners to provide more comprehensive and in-depth diagnostic evidence to ensure the quality and yield of tomatoes. First, we focus on six common tomato leaf disease images and text descriptions, creating a Tomato Leaf Disease Image-Text Retrieval Dataset (TLDITRD), introducing image-text retrieval into the field of tomato leaf disease retrieval. Then, utilizing ViT and BERT models, we extract detailed image features and sequences of textual features, incorporating contextual information from image-text pairs. To address errors in image-text retrieval caused by complex backgrounds, we propose Learnable Fusion Attention (LFA) to amplify the fusion of textual and image features, thereby extracting substantial semantic insights from both modalities. To delve further into the semantic connections across various modalities, we propose a False Negative Elimination-Adversarial Negative Selection (FNE-ANS) approach. This method aims to identify adversarial negative instances that specifically target false negatives within the triplet function, thereby imposing constraints on the model. To bolster the model's capacity for generalization and precision, we propose Adversarial Regularization (AR). This approach involves incorporating adversarial perturbations during model training, thereby fortifying its resilience and adaptability to slight variations in input data. Experimental results show that, compared with existing ultramodern models, LAFANet outperformed existing models on TLDITRD dataset, with top1, top5, and top10 reaching 83.3% and 90.0%, and top1, top5, and top10 reaching 80.3%, 93.7%, and 96.3%. LAFANet offers fresh technical backing and algorithmic insights for the retrieval of tomato leaf disease through image-text correlation.
Cross-modal retrieval for rice leaf diseases is crucial for prevention, providing agricultural experts with data-driven decision support to address disease threats and safeguard rice production. To overcome the limitations of current crop leaf disease retrieval frameworks, we focused on four common rice leaf diseases and established the first cross-modal rice leaf disease retrieval dataset (CRLDRD). We introduced cross-modal retrieval to the domain of rice leaf disease retrieval and introduced FHTW-Net, a framework for rice leaf disease image-text retrieval. To address the challenge of matching diverse image categories with complex text descriptions during the retrieval process, we initially employed ViT and BERT to extract fine-grained image and text feature sequences enriched with contextual information. Subsequently, two-way mixed self-attention (TMS) was introduced to enhance both image and text feature sequences, with the aim of uncovering important semantic information in both modalities. Then, we developed false-negative elimination-hard negative mining (FNE-HNM) strategy to facilitate in-depth exploration of semantic connections between different modalities. This strategy aids in selecting challenging negative samples for elimination to constrain the model within the triplet loss function. Finally, we introduced warm-up bat algorithm (WBA) for learning rate optimization, which improves the model's convergence speed and accuracy. Experimental results demonstrated that FHTW-Net outperforms state-of-the-art models. In image-to-text retrieval, it achieved R@1, R@5, and R@10 accuracies of 83.5%, 92%, and 94%, respectively, while in text-to-image retrieval, it achieved accuracies of 82.5%, 98%, and 98.5%, respectively. FHTW-Net offers advanced technical support and algorithmic guidance for cross-modal retrieval of rice leaf diseases.
Background: Elderly stroke survivors are encouraged to receive appropriate health information to prevent recurrences. After discharge, older patients seek health information in everyday contexts, examining aspects that facilitate or impair healthy behavior. Objectives: To explore the experiences of older stroke patients when searching for health information, focusing on search methods, identification of health information, and difficulties faced during the search process. Methods: Using the qualitative descriptive methodology, semi-structured interviews were conducted with fifteen participants. Results: Participants associated the health information they sought with concerns about future life prospects triggered by perceived intrusive changes in their living conditions. Based on the participants' descriptions, four themes were refined: participants' motivation to engage in health information acquisition behavior, basic patterns of health information search, source preferences for health information, and difficulties and obstacles in health information search, and two search motivation subthemes, two search pattern subthemes, four search pathway subthemes, and four search difficulty subthemes were further refined. Conclusion: Older stroke patients face significant challenges in searching for health information online. Healthcare professionals should assess survivors' health information-seeking skills, develop training programs, provide multichannel online access to health resources, and promote secondary prevention for patients by improving survivors' health behaviors and self-efficacy.
Escherichia coli (E. coli) mutant strains have been reported to extend the life span of Caenorhabditis elegans (C. elegans). However, the specific mechanisms through which the genes and pathways affect aging are not yet clear. In this study, we fed Drosophila melanogaster (fruit fly) various E. coli single-gene knockout strains to screen mutant strains with an extended lifespan. The results showed that D. melanogaster fed with E. coli purE had the longest mean lifespan, which was verified by C. elegans. We conducted RNA-sequencing and analysis of C. elegans fed with E. coli purE (a single-gene knockout mutant) to further explore the underlying molecular mechanism. We used differential gene expression (DGE) analysis, enrichment analysis, and gene set enrichment analysis (GSEA) to screen vital genes and modules with significant changes in overall expression. Our results suggest that E. coli mutant strains may affect the host lifespan by regulating the protein synthesis rate (cfz-2) and ATP level (catp-4). To conclude, our study could provide new insights into the genetic influences of the microbiota on the life span of a host and a basis for developing anti-aging probiotics and drugs.
Recent studies have shown that gut microorganisms can modulate host lifespan and activities, including sleep quality and motor performance. However, the role of gut microbial genetic variation in regulating host phenotypes remains unclear. In this study, we investigated the links between gut microbial genetic variation and host phenotypes using Saccharomyces cerevisiae and Drosophila melanogaster as research models. Our result suggested a novel role for peroxisome-related genes in yeast in regulating host lifespan and activities by modulating gut oxidative stress. Specifically, we found that deficiency in catalase A (CTA1) in yeast reduced both the sleep duration and lifespan of fruit flies significantly. Furthermore, our research also expanded our understanding of the relationship between sleep and longevity. Using a large sample size and excluding individual genetic background differences, we found that lifespan is associated with sleep duration, but not sleep fragmentation or motor performance. Overall, our study provides novel insights into the role of gut microbial genetic variation in regulating host phenotypes and offers potential new avenues for improving health and longevity.
Gastrointestinal Microbiome , Longevity , Animals , Longevity/genetics , Drosophila melanogaster/genetics , Saccharomyces cerevisiae , Genetic Variation
[This corrects the article DOI: 10.1371/journal.pone.0241047.].
Qinling-Daba Mountains (QDM), which are located in central China, are considered as a significant climatic boundary delimiting north and south. However, the influence of complex topographic and climatic features makes it challenging to identify the exact location of the boundary, and different scholars delimit the boundary with significant differences. In addition, there is a gradual transition between climate zones, and no real dividing line exists. To explore the climate regionalization of the QDM, we focused on the identification of the transition zone rather than the exact location of the boundary between subtropical and temperate zones. Thus, we proposed a new workflow for climate regionalization based on the Geodetector-SVM model (a combination of Geodetector and support vector machines). First, we selected the spatial distribution data of six vegetation types (including typical subtropical and temperate vegetation) to represent the spatial distribution of climatic zones. Environmental factors (such as topography, temperature, precipitation, and soil) were used as explanatory variables for the spatial distribution of vegetation. Second, using the Geodetector-SVM model, the distribution characteristics and suitable environment of typical vegetation in different climatic zones are comprehensively explored. By analyzing the multiple boundaries between subtropical and temperate vegetation, the location of the transition zone of the QDM was identified. The results revealed the following: (1) The new workflow for climate regionalization based on the Geodetector-SVM model is powerful for the identification of the transition zone. The q-statistics are generally greater than 0.35, indicating that the transition zone between subtropical and temperate zones can highly reflect the character of the QDM; (2) From west to east, the transition zone mainly passes through the cities of Heishui County, Kang County, Liuba County, and Yichuan County and is approximately 30 km wide.
Climate , Geography , China , Cities , Climate Change , Environment , Soil , Support Vector Machine , Temperature
