ClarusC64/clinical-immune-cascade-instability-v0.1

该数据集用于评估模型是否能检测由免疫级联放大引起的不稳定性。每行数据代表一个简化的免疫激活场景，包含三个时间点。核心任务是判断免疫系统是否保持稳定或趋向于失控的炎症不稳定性。数据集包含多种信号轨迹和代理变量，如CRP轨迹、温度轨迹、细胞因子代理轨迹、淋巴细胞水平、炎症指数、免疫储备和干预延迟等。预测目标为标签1表示免疫级联不稳定性，标签0表示稳定的免疫调节。数据集还包含干扰变量如实验室噪声和图表噪声。评估指标包括准确率、精确率、召回率、F1分数等。

This dataset evaluates whether models can detect instability caused by immune cascade amplification. Each row represents a simplified immune activation scenario across three time points. The task is to determine whether the immune system remains regulated or is moving toward runaway inflammatory instability. The dataset includes various signal trajectories and proxy variables such as CRP trajectory, temperature trajectory, cytokine proxy trajectory, lymphocyte level, inflammation index, immune reserve, and intervention delay. The prediction target is label 1 for immune cascade instability and label 0 for stable immune regulation. The dataset also includes decoy variables such as lab noise and chart noise. Evaluation metrics include accuracy, precision, recall, F1 score, etc.