Skip to main content


Fig. 5 | BMC Bioinformatics

Fig. 5

From: ODTbrain: a Python library for full-view, dense diffraction tomography

Fig. 5

Dependence of reconstruction quality on the size of the cell. a Normalized root mean square (RMS) error in dependence of the cell diameter for two- and 3D reconstructions with the Born, Radon, and Rytov approximations. The extent of the simulation volume is 30 λ. b Same as in (a), except that the extent of the simulation volume is 172 λ. Due to computational limitations, a corresponding 3D simulation was not feasible with the FDTD method. c Normalized total variation (TV) error in dependence of the cell diameter corresponding to the values in (a). d Normalized TV error for the 2D series as described in (b). At the position indicated by the arrow we observed a sudden break down of the one-dimensional phase unwrapping algorithm. e, f, g 2D refractive index reconstructions with the Rytov approximation. The cell diameters are 7 λ (e), 52 λ (f), and 97 λ (g). The refractive index values of the cell phantom and the total number of projections (200) are unchanged for all simulations

Back to article page