ClarusC64/clinical-hemodynamic-trajectory-instability-v0.1

该数据集用于评估模型是否能够从短期的临床轨迹中检测出即将出现的血流动力学不稳定性。每个数据行代表一个患者场景，包含三个时间点的测量数据。任务是通过这些数据判断患者的血流动力学状态是否稳定（标签0表示稳定，标签1表示不稳定/崩溃轨迹）。数据集包含多个变量，如血压、乳酸水平、心率、尿量等，以及一些干扰变量。评估指标包括准确率、精确率、召回率、F1分数和混淆矩阵。

This dataset evaluates whether models can detect emerging hemodynamic instability from short clinical trajectories. Each row represents a patient scenario with measurements across three time points. The task is to determine whether the trajectory indicates a stable state or impending circulatory instability (label 0 for stable, label 1 for instability/collapse trajectory). The dataset includes variables such as blood pressure, lactate levels, heart rate, urine output, and some decoy variables. Evaluation metrics include accuracy, precision, recall, F1 score, and confusion matrix.