Figure 1 | Scientific Reports

Figure 1

From: Continuous learning and inference of individual probability of SARS-CoV-2 infection based on interaction data

Figure 1

The learning and inference scheme of the CLIIP model. The diagram shows continuous backtracking learning while the newly infected people are being recorded. Given n is the maximum day of incubation period, the incubation period requires us to group the patients into their actual contagious time according to the distribution of (a) from time t to range \([t-f(n), t-1]\). The arrangement describes the possible time of the latent infection, which lies in the incubation period of the infected. The inference model of every t is represented by an individual directed graph (IDG) and we use a day as t in our simulation. The red circles denote confirmed infected people, the hollow squares mean exposed people, and a filled square is an individual who stays on the path from one infected person to the others which might be asymptomatic virus carriers. A hollow green diamond is labeled as a healthy person. The arrows denote the possible path of transmission derived from people’s location and staying time. The virus will stay in the same place for a while28 which makes the last person to leave the specific place run a high risk of getting infected. Also, we define each layer as the number of edges between two nodes. For instance B is a center-surround node in the fourth layer of A.

Back to article page