Skip to main content
. Author manuscript; available in PMC: 2016 Oct 18.
Published in final edited form as: Phys Rev E. 2016 Oct 4;94(4):042401. doi: 10.1103/PhysRevE.94.042401

Fig. 2.

Fig. 2

The AF burden varies with time, calculated from the time in AF in a sliding window of 5 × 106 time steps for four different patients. Each simulation begins with an initial fraction of vertical connections of ν = 0.25 and is depleted to ν = 0.13, mimicking progression of stochastic uncoupling over a simulated time period of 4.3 × 107 time steps. (a) Patient A develops paroxysmal AF for ν ≲ 0.188, which eventually develops into persistent AF at ν ≲ 0.138. (b) Patient B develops short-lived episodes of AF for 0.206 ≲ ν ≲ 0.238 and a sudden transition into persistent AF ν ≈ 0.206. This relapses into paroxysmal AF three times before remitting back into persistent AF at ν ≈ 0.166. (c) Patient C shows a phase of relapsing-remitting paroxysmal to persistent AF for a considerable period of time until AF becomes persistent at ν ≈ 0.159. (d) Patient D shows isolated short-lived episodes of AF and a sudden transition into persistent AF at ν ≈ 0.164. Remission back into sinus rhythm occurs twice (ν ≈ 0.156 and ν ≈ 0.145) before AF becomes persistent. The vertical dashed lines are values of fractions of vertical connections at which simulations are rerun without progressive uncoupling, that is, with a fixed fraction of vertical connections (see Fig. 3).