Fig 1.
A schematic representation of the overarching model simulating micro-emboli in the brain post-thrombectomy. Two separate models are developed in this paper: a) a micro-emboli model of the capillary bed (based on geometrically representative networks) and, b) a micro-emboli model of the penetrating arterioles. The grey region in b) represents the homogenized healthy capillary bed to which the terminal vessels of the trees couple. The red planes denote the dividing planes between the 6 layers of the voxel over which the blood flow characteristics will be calculated. c) Is a porous multi-compartment representation of the brain which will depend on the parameters derived from the models in a) and b). The full brain is not simulated in this paper.
Fig 2.
Flowchart representation of the micro-emboli occlusion algorithm.
Fig 3.
a) A scatter plot of the permeability of the 100 voxels at each of the 6 depth layers. The cyan triangle indicates the median permeability of the capillary bed. The mean arteriolar permeability is indicated with a red star at each layer. The arteriolar permeabilities (circles) are shaded differently to aid visualisation. b) A quadratic line of best fit over the median permeabilities at each layer–error bars are interquartile ranges.
Fig 4.
a,b) The fractional drop in permeability against the fraction of vessels blocked where a) is in the top layer of the voxel and b) is the middle layer. c,d) The fractional drop in permeability against the fraction of vessel surface area blocked where c) is in the top layer of the voxel and d) is the middle layer. e,f) The fractional drop in permeability against the fraction of vessel volume blocked where e) is in the top layer of the voxel and f) is the middle layer. The middle layer is at a depth of 1.25–1.67mm. A line of best fit is plotted in black, with the fit equation in the top right corner of each graph, with the bottom left of the graph giving the coefficients of the line of best fit and the R2 goodness-of-fit metric.
Fig 5.
A 2-dimensional surface plot for the ADAPT thrombectomy technique removing a hard clot.
The permeability is plotted against voxel depth and volume fraction of the vessels occluded.
Fig 6.
A comparison of the permeability drop against volume fraction occluded for 3 different voxel geometries simulated with 100 different clot distributions (for the ADAPT technique, hard clot) a) Voxel 1, b) Voxel 4, c) Voxel 31.
Table 1.
Summary of the differences between thrombectomy technique and clot consistency when considering total micro-emboli sampled and those micro-emboli that occlude vessels.
The values are averaged over the 100 voxel simulations. The percentage of sampled clots occluding vessels is the average of the 100 individual percentages.
Fig 7.
Comparison of the permeability drops against volume fraction blocked in the middle layer for 4 different thrombectomy techniques: a) removing hard clots and b) removing soft clots.
Fig 8.
The drops in permeability over 500 statistically accurate capillary networks with a cube length of 375 μm.
Lines of best fit are in black, with the gradient B in the top right corner of each graph. a) The fractional drop in permeability with fraction of vessels occluded (B = −3.2), b) the fractional drop in permeability with vessel volume occluded (B = −4.1).