基于个体时序列差异网络探测疾病恶化的预警信号

Detecting the Early Warning Signals of Disease Progression Based on Individual Difference Time Series Network

摘要: 前疾病状态是疾病状态的一个临界期，处于这个状态的患者,只要经过合理有效的治疗，就可以回到正常状态。所以，探测前疾病状态对于医疗工作者以及病人来说有着极其重要的意义。本文开发了一种算法，基于个体单样本建立个体时序列差异网络，并根据所建立的个体时序列差异网络，可以有效地探测疾病恶化的信号。该方法的有效性得到了数值模拟和两个真实数据的检验。

Abstract: Pre-disease state is a critical stage of the disease state. Patients in this state can return to the normal state as long as they receive reasonable and effective treatment. Therefore, it is of great significance for medical workers and patients to detect the pre-disease state. In this paper, an al-gorithm is developed to establish the sampling-specific temporal differential network (SSDN) based on the individual single sample, and according to the sampling-specific temporal differential network, the signal of disease progression can be detected effectively. The validity of the method is verified by numerical simulation and two real data.