As urbanization accelerates, a host of negative ecological impacts have become increasingly prominent. Green roofs, as a sustainable solution, can effectively mitigate the urban heat island effect and reduce carbon footprints. However, the lack of datasets on plant species suitable for green roofs in China has hindered the advancement of relevant research and practical applications. Therefore, this study constructed a diversified dataset of plant species for green roofs in China, using data sources from the China National Knowledge Infrastructure (CNKI) and Web of Science (WoS). Generated by integrating the Kimi large language model (Kimi LLM) API with knowledge graph technology, the dataset contains 2248 plant records. It specifically includes a statistical CSV file with detailed plant information, a CSV file of species combinations, a CSV file linking plant combinations to cities, and original plant data extracted from research papers. Technical experiments have validated the accuracy and efficiency of this dataset in acquiring plant species. Suitable for plant selection in green roof projects, this dataset will provide strong support for in-depth research and wider applications in the field of urban sustainability.

Accurate kidney tumor segmentation from abdominal CT is essential for quantitative assessment and treatment planning. However, indistinct tumor boundaries and substantial inter-patient shape variability render traditional hand-crafted feature-based methods unreliable for precise delineation. Although deep learning has advanced this task, these methods still struggle with multi-scale tumor characteristics, complex morphological variations, and background noise in medical images. To address these challenges, we propose MDA-Net, an end-to-end segmentation method based on enhanced multi-scale feature extraction and attention refinement. Specifically, we introduce a Multi-Scale Feature Extraction (MSFE) module into encoder–decoder skip connections to aggregate dilated features across multiple receptive fields and learn branch-wise weights for adaptive refinement and fusion, thereby enhancing boundary details and semantic cues to reduce tumor-tissue ambiguity. At the bottleneck, a Deformable Pyramid Feature Refinement (DPFR) module combines deformable sampling with pyramid contextual modeling, thereby improving adaptability to variations in tumor shape and scale while preserving feature resolution. Moreover, a Channel and Spatial Attention (CASA) module is embedded in the decoder to suppress background interference and enhance boundary-sensitive structures during upsampling via coordinated channel and spatial reweighting, thereby improving the reconstruction of fine-grained tumor morphology and contours. Experiments on both KiTS19 and KiTS21 show that MDA-Net consistently improves tumor boundary delineation, lesion localization, and mask reconstruction, demonstrating stronger robustness and cross-dataset generalizability than representative baseline methods. Ablation studies further confirm the complementary effects of MSFE, DPFR, and CASA. In addition, Grad-CAM visualizations improve the clinical transparency and interpretability of the model. Overall, this method advances deep learning for medical image analysis and supports precise diagnosis and treatment of renal tumors.