Fig. 2

The algorithm of landscape dynamic network marker. The flow chart above shows how the algorithm works based on city networks and the historical information of clinic hospitalization. Regarding a point T = t (t > 5) as a candidate tipping point, L-DNM scores can be calculated. If the L-DNM score increases significantly, the candidate T = t is determined as the identified tipping point, and the algorithm ends. Otherwise, if there is no significant change in the score, then T = t is classified as a time point belonging to the normal stage, and the algorithm continues with T = t + 1 being a new candidate tipping point