Modeling light scattering from biological cells using a finite-difference time domain method

The effect of cell morphology on light scattering is investigated using computer simulations. A parallel implementation of the finite-difference time domain (FDTD) method has been developed to simulate the scattering and obtain various scattering properties such as the Mueller matrix and anisotropy factor. In addition, a program which produces 3D cell models for the FDTD program has been developed. Scattering from realistic red blood cell (RBC) and B-cell precursor (B-cell) models has been simulated and the results are compared with those of the simpler sphere and coated sphere models. The RBC models are based on models developed from mechanical principles which impose constraints on the volume and surface area of an RBC under pressure. The B-cell models are created by taking confocal microscopy images of cultured NALM6 cells and applying statistical and geometric methods to produce a realistic 3D model. Validation of the FDTD program against Mie theory results are presented along with performance evaluations on three different parallel computing platforms. Simulations of the more realistic cell models show that the sphere models are not suitable for determining most of the scattering properties; however, a more complex ellipsoid model provides a good approximation for some scattering properties. It was also found that the amount of forward scattered light is closely related to the volume of the cell, while light scattered toward the side is more closely related to the refractive index. Simulations using various nucleus models show that the complexity of the shape of the nucleus and its relative refractive index influence the light scattering in various ways. A comparison with experimental results for cultured NALM6 cells is also presented.