ClarusC64/clinical-ventilation-perfusion-instability-v0.1

该数据集用于评估模型是否能够从短期临床代理轨迹中检测呼吸气体交换的不稳定性。每行数据代表一个类似患者的呼吸场景，观察三个时间点。任务目标是判断系统是否保持控制或朝向通气-灌注不稳定的方向发展。数据集包含氧饱和度轨迹、动脉氧代理轨迹、呼吸频率轨迹、PEEP水平、通气支持响应、心输出量代理、分流代理和响应延迟等变量，以及一些干扰变量。预测目标为标记1表示通气-灌注不稳定，标记0表示稳定或受控的气体交换状态。数据集是Clarus稳定性推理基准的一部分，支持呼吸不稳定检测、气体交换轨迹推理等研究领域。

This dataset evaluates whether models can detect respiratory gas-exchange instability from short clinical proxy trajectories. Each row represents a patient-like respiratory scenario observed across three time points. The task is to determine whether the system remains controlled or is moving toward ventilation–perfusion instability. The dataset includes variables such as SpO2 trajectory, PaO2 proxy trajectory, respiratory-rate trajectory, PEEP level, ventilation support response, cardiac output proxy, shunt proxy, and response delay, along with some decoy variables. The prediction target is label 1 for ventilation–perfusion instability and label 0 for a stable or controlled gas-exchange state. The dataset is part of the Clarus Stability Reasoning Benchmark and supports research into respiratory instability detection, gas-exchange trajectory reasoning, and other related fields.