Abstract:
In this talk, I first review our studies on DNM (Dynamical Network Markers) theory that detects early warning signals of imminent codimension-one local bifurcations in a data-driven manner. Then, I use DNM as DNB(Dynamical Network Biomarkers) for early detection of critical transitions from a healthy state to a disease state at the timing of a pre-disease state that is still asymptomatic. I also validate DNB theory for early medicine on the basis of physiological data. Last, I discuss a possibility of applying DNM theory to critical transitions in different fields.