KiTS21数据集
包含300名肾癌患者的CT图像和临床数据，标注肾脏和肿瘤，用于生存预测和ISUP分级分类。
基本信息
资源简介
KiTS21数据集
【数据集背景】
本数据集源站标题为《Balasingham-AI-Group/Survival_CTplusClinical》。源站首次发布：2023-04-20。
【数据内容】
源站原始描述：Multimodal Deep Learning for Personalized Renal Cell Carcinoma Prognosis: Integrating CT Imaging and Clinical Data # Survival_CTplusClinical Multimodal Deep Learning for Personalized Renal Cell Carcinoma Prognosis: Integrating CT Imaging and Clinical Data Renal cell carcinoma represents perilous cancer worldwide, characterized by a low survival rate. In this study, we develop a multimodal deep-learning approach for estimating survival probabilities in renal cell carcinoma patients, leveraging CT…
【适用方向】
多模态数据适合跨模态对齐、融合建模与联合表征学习方向的研究。
【获取与许可】
源站页面：https://github.com/Balasingham-AI-Group/Survival_CTplusClinical
使用许可：Apache-2.0。
下载信息
注册下载
需要注册 Kaggle 账号并登录后下载，适合需要跟踪下载记录和使用 API 的用户。
暂未开放公开下载
无需注册即可直接获取公开样本或文档，适合快速预览和评估数据集质量。
免登录有偿下载
公开数据集受托下载与技术交付服务。
提供高速下载与技术交付服务（收技术服务费，非数据销售）
暂未开放当前数据集主要来源为 Kaggle 公开托管，完整影像包建议通过原始链接或 Kaggle API 获取。
使用方式
数据集获取
git clone https://github.com/Balasingham-AI-Group/Survival_CTplusClinical.git
curl -L -o repo.zip https://github.com/Balasingham-AI-Group/Survival_CTplusClinical/archive/refs/heads/main.zip
unzip repo.zip
源站 README 摘录（使用方式）
Survival_CTplusClinical
Multimodal Deep Learning for Personalized Renal Cell Carcinoma Prognosis: Integrating CT Imaging and Clinical Data<br>
Renal cell carcinoma represents perilous cancer worldwide, characterized by a low survival rate. In this study, we develop a multimodal deep-learning approach for estimating survival probabilities in renal cell carcinoma patients, leveraging CT imaging and clinical data. We attempt to facilitate the identification of patients requiring urgent treatment by estimating survival probabilities. Our framework comprises three modules: a 3D image feature extractor, clinical variable selection, and survival prediction. Based on the Efficient-Net B7 3D architecture, the feature extractor module extracts the features of CT images by predicting the ISUP grade (defined by the International Society of Urological Pathology) of renal cell carcinoma tumors, which is linked to mortality rates. In order to boost the feature extractor’'s performance, we utilize a number of preprocessing techniques. In the clinical variable selection module, the selection of the most relevant clinical variables to prognosis for integration with CT images is conducted systematically, utilizing the Spearman score and random forest importance score as the criteria. The survival prediction module is a DL-based network trained with discrete LogisticHazard-based loss. A total of nine distinct experiments are performed, employing varying numbers of clinical variables determined by diverse thresholds of the Spearman and importance scores. Our findings show that the suggested strategy outperforms the state-of-the-art literature in the prognosis of renal cancer based on CT scans and clinical factors. The best experiment yielded a concordance index of 0.84 and an area under the curve value of 0.8 on the test subset.
完整仓库：github.com/Balasingham-AI-Group/Survival_CTplusClinical
数据缺失？
依托陕西、四川两大基地，我们与超过十家三甲医院建立直接合作关系，覆盖合规授权、采集治理、专业标注、数据交付的全流程，为AI医疗团队提供即拿即用的高质量临床数据。
⚡️ 需要数据支持或标注服务？立即联系我们获取专业方案。
