合成神经外科数据集
收藏资源简介:
本研究使用的合成神经外科数据集是由GPT-4o生成的,它根据现实世界开放访问的神经外科数据集的统计特性,生成具有高保真度的合成数据。数据集包含139名接受神经外科手术的老年患者的记录,并额外生成了一个放大十倍样本量的数据集。该数据集保留了现实世界数据的单变量和双变量统计特性,没有直接暴露任何真实患者记录,可以用于增强临床数据,训练机器学习模型以预测神经外科结果。
The synthetic neurosurgical dataset utilized in this study was generated by GPT-4o, which produces high-fidelity synthetic data following the statistical characteristics of real-world open-access neurosurgical datasets. This dataset includes records of 139 elderly patients who underwent neurosurgery, alongside an additional dataset with a sample size ten times larger. It retains the univariate and bivariate statistical properties of real-world clinical data, does not directly expose any real patient records, and can be employed to augment clinical datasets and train machine learning models for predicting neurosurgical outcomes.




