Macao Employment Survey of Macao, Monthly 2007|就业调查数据集|澳门数据集
收藏PlantVillage
在这个数据集中,39 种不同类别的植物叶子和背景图像可用。包含 61,486 张图像的数据集。我们使用了六种不同的增强技术来增加数据集的大小。这些技术是图像翻转、伽玛校正、噪声注入、PCA 颜色增强、旋转和缩放。
OpenDataLab 收录
中国社会事件数据库(CSED)
中国社会事件数据库(CSED)是一个基于时间线的事件汇总和分析工具,旨在记录每日社会动态与网络舆情。
github 收录
Hospital Deterioration Dataset
这是一个高保真模拟医院队列数据集,包含10,000个模拟医院入院记录,每个记录跟踪最多72小时。数据集提供每小时的生命体征(心率、血压、呼吸频率等)和实验室数值(白细胞计数、乳酸、肌酐等),以及患者人口统计学信息和多种恶化结果标签。专门设计用于构建和基准测试早期预警系统和临床恶化风险模型的机器学习应用。
github 收录
Supplementary Material for: Endoscopic Laryngopharyngeal Surgery Combined with Endoscopic Submucosal Dissection as a Treatment Option for Superficial Hypopharyngeal Cancer with Suspected Synchronous Lymph Node Metastasis
We report 4 cases of hypopharyngeal cancer preoperatively suspected with synchronous lymph node metastases. Pathologic lymph node metastasis was confirmed in three of the four cases. All 4 cases underwent endoscopic laryngopharyngeal surgery (ELPS) combined with endoscopic submucosal dissection (ESD) and subsequent lymph node dissection as an optional treatment rather than the standard treatment. Peroral resection for primary site was selected because of the expected decline in quality of life (QoL) after radical surgery. Among 4 patients, one developed local recurrence; however, the other three remained recurrence-free and survived without any additional treatment. Furthermore, the patient who developed local recurrence had a recurrence-free survival for more than 5 years, with additional chemoradiation therapy. No disorders in speech, swallowing, or breathing was observed during the follow-up period. ELPS combined with ESD is generally indicated for laryngopharyngeal cancer without synchronous lymph node metastasis. However, this can be a treatment option for patients may wish to preserve a greater QoL after treatment. In the future, when more data on the results and long-term prognosis of this treatment are accumulated, it may be possible to discuss its validity further.
DataCite Commons 收录
中国逐日格点降水数据集V2(1960–2024,0.1°)
CHM_PRE V2数据集是一套高精度的中国大陆逐日格点降水数据集。该数据集基于1960年至今共3476个观测站的长期日降水观测数据,并纳入11个降水相关变量,用于表征降水的相关性。数据集采用改进的反距离加权方法,并结合基于机器学习的LGBM算法构建。CHM_PRE V2与现有的格点降水数据集(包括CHM_PRE V1、GSMaP、IMERG、PERSIANN-CDR和GLDAS)表现出良好的时空一致性。数据集基于63,397个高密度自动雨量站2015–2019年的观测数据进行验证,发现该数据集显著提高了降水测量精度,降低了降水事件的高估,为水文建模和气候评估提供了可靠的基础。CHM_PRE V2 数据集提供分辨率为0.1°的逐日降水数据,覆盖整个中国大陆(18°N–54°N,72°E–136°E)。该数据集涵盖1960–2024年,并将每年持续更新。日值数据以NetCDF格式提供,为了方便用户,我们还提供NetCDF和GeoTIFF格式的年度和月度总降水数据。
国家青藏高原科学数据中心 收录
