Fig 1.
Summary of the combined approach process.
Controlled blood flow experiments are performed to form thrombi in real-time (i). These thrombi are reconstructed into a 3D surface (ii) and time-corrected (iii) to account for the duration it takes to image a thrombus. Time-correcting the thrombi allows for accurate growth rates to be calculated and spatially mapped (iv). Flow experiments (v) are completed to determine the detailed flow properties (vi).
Fig 2.
Time-corrected thrombus reconstruction process.
(a) Image metadata showing microscope stage position over time. (b) Individual slices have original acquisition times assigned from metadata. Acquired confocal images are in black; morphed intermediate images are in blue (c). Slices corresponding to the 149-second timepoint are extracted and stacked vertically forming an instantaneous z-stack which is reconstructed (d) to produce the thrombus field at 149 seconds.
Fig 3.
Comparison between real-time surfaces and time skewed surfaces.
Comparison between ellipsoid reconstructions constructed using (a) real-time corrected ellipse slices and (b) ellipse slices derived from time skewed slices. (c) Comparison between time stack thrombus reconstruction (top) and real-time thrombus reconstruction (bottom) mirrored through the central plane.
Fig 4.
Validation of time-correction technique using two time stacks to reconstruct the intermediate time stack.
(a), (b), (c) Time stacks 29, 30 and 31 are all acquired using confocal microscopy and reconstructed. (d) Time stack 29 and 31 are then used to reconstruct the thrombus surface at the intermediate time stack 30. (e) The deviation between time stack 30 acquired using confocal microscopy and the morphologically reconstructed time stack 30 measured as a spatial distance (Maximum spatial deviation < 10−2 μm). Blue indicates a deviation of 0 μm, while red indicates a deviation of 10−2 μm.
Fig 5.
Determining regions of high growth or consolidation and quantifying the amount of growth between two points in time.
(a) The reconstructed thrombus surface at time 1 (100 seconds into growth phase). (b) The reconstructed thrombus surface at time 2 (135 seconds into growth phase). (c) Developed algorithm indicates which regions have grown (shown in red) and those regions which have consolidated or retracted (shown in blue). Normal vectors are calculated between the surface at time 1 towards the surface at time 2 from which a Euclidean distance is calculated to give the magnitude a particular point on the surface has either grown or consolidated between the two points in time. The normal vector magnitudes are mapped onto the surface at time 1 highlighting regions of high growth or consolidation.
Fig 6.
Thrombus growth and shear rate mapped onto original thrombus surface with shear rate experienced by individual platelets.
(a) Shear rate 1 μm above surface mapped onto original thrombus surface. Spheres represent the path a platelet landing in a high shear rate region follows, coloured by shear rate experienced (b) Original thrombus surface with regions of growth (shown in red) and regions of consolidation or embolism (shown in blue) during the last 1 second of blood flow. (c) Shear rate history over the last 3 seconds prior to adhesion for platelets adhering in very high shear rate (γlocal >10000 s-1), high shear rate (γlocal ~ 4000 s-1) and moderate shear rate (γlocal <2000 s-1) regions.