Junzhuo Song

To address the mismatch between agricultural development and water resource conditions, this study examined 22 provinces and four municipalities in China over the period 2014–2023 and incorporated terrain factors into the analytical framework. An integrated evaluation system was developed for the Agricultural Development Index (ADI) and Water Resource Condition Index (WCI). The entropy weight method, coupling coordination degree model, standard deviation ellipse model, spatial difference coefficient, and two-way fixed-effects model were employed to characterize the spatiotemporal dynamics and coordination between agricultural development and water resource conditions. The results indicate that the ADI increased steadily throughout the study period, whereas the WCI remained relatively stable, reflecting distinct evolutionary trajectories of the two systems. Although the coordination level of the Agriculture–Water Resources System improved overall, pronounced regional disparities persisted. Plain regions exhibited stronger coordination, whereas mountainous and plateau regions showed lower coordination levels because of terrain constraints and limited resource endowments. Spatial mismatches between agricultural and water resource advantage regions were evident, although the overall matching relationship between the two systems gradually strengthened. The proposed evaluation framework provides a scientific basis for optimizing regional agricultural layouts, improving water resource allocation, and promoting coordinated agricultural and water resource development.

Addressing the mismatch between agricultural development and water resource conditions, this study selected 22 provinces and 4 municipalities in China from 2014 to 2023 as the research area and incorporated terrain factors into the analytical framework. An evaluation system for the Agricultural Emphasis Index and Water Resource Condition Index was established. The entropy weight method, coupling coordination degree model, standard deviation ellipse model, spatial difference coefficient, and two-way fixed effects model were employed to investigate the spatiotemporal variations and coordination relationships between agricultural development and water resource conditions. The results showed that the Agricultural Emphasis Index continuously increased during the study period, whereas water resource conditions remained relatively stable, indicating different evolutionary patterns between the two systems. The coordination level of the Agriculture–Water Resources System generally improved, but significant regional disparities remained. Plain regions exhibited stronger coordination foundations, while mountainous and plateau regions showed relatively lower coordination levels due to terrain constraints and resource limitations. Spatial displacement existed between agricultural advantage regions and water resource advantage regions, although the matching degree between the two systems gradually improved. The evaluation framework established in this study provides a scientific basis for optimizing regional agricultural layouts, improving water resource allocation, and promoting coordinated agricultural and water resource development.

Reliable confidence estimation is essential for surgical visual question answering because overconfident errors may compromise clinical decision support in high-risk settings. However, existing methods typically model predictive uncertainty as a single global quantity, which fails to capture the hierarchical uncertainty arising from ambiguous objects, uncertain tool–tissue interactions, and complex scene context. To address this limitation, we propose a scene graph-guided framework that decomposes uncertainty into object-level, relation-level, and scene-level components and adaptively fuses them for confidence-aware surgical visual question answering. The framework further incorporates Dirichlet-based calibration to improve probabilistic quality and support selective answering. Experiments on the Surgical Scene Graph-Question Answering (SSG-QA) benchmark show that the proposed method achieves an overall accuracy of 63.58%, an Expected Calibration Error of 16.97%, and a Risk-Coverage AUC of 0.0861. Our SG-UD framework achieves competitive calibration performance, significantly improving Brier Score and NLL over baselines, while providing more granular uncertainty interpretability. Performance improves across all question types, with the largest gain observed for relation-type questions (+0.92% over MCAN). Ablation studies further show that uncertainty decomposition contributes most to answer accuracy (+0.70%), whereas Dirichlet calibration is critical for improving probabilistic quality. These findings indicate that explicitly modeling uncertainty across semantic levels can enhance the reliability and interpretability of surgical visual question answering and provide fine-grained confidence information for safer artificial intelligence-assisted surgical decision support.

Accurate and rapid identification of bacterial species is essential for public health, clinical diagnostics, and environmental monitoring. Although Raman spectroscopy offers a powerful, non-invasive alternative, reliance solely on spectral data often fails to distinguish species with highly similar signatures, particularly when the discriminating features are subtle. This difficulty is frequently compounded by a lack of integrated biological prior knowledge, which can hinder model performance. To address these challenges, we introduce BactoRamanBioNet, a novel multimodal neural network architecture. Our model employs a synergistic approach that utilizes a ResNet-Transformer architecture to capture complex spectral patterns and a CLIP text encoder to incorporate descriptive biological information, thereby enabling highly accurate multimodal classification of bacterial species. Empirical results demonstrate that BactoRamanBioNet achieves a classification accuracy of 98.2% and an F1-score of 98.0%. This performance surpasses the current state-of-the-art deep learning model, ResNet-1D, by 2.4% in accuracy and 2.0% in F1-score. Moreover, our model outperforms traditional classifiers, such as Support Vector Machine (SVM) and Random Forest (RF), by 9.8% and 7.9% in accuracy, respectively, while also exhibiting significant improvements in precision and recall. By establishing a new benchmark in performance and robustness, BactoRamanBioNet offers a powerful and reliable framework for automated microbiological analysis, paving the way for next-generation diagnostic systems.