Wavelength-resolved morphology-related fluctuations in tissue scattering coefficient can be masked by corresponding spectral variations in tissue anisotropy

A long-standing challenge in tissue optics is extracting biophysical information about tissue structures (cells, organelles, collagen fibers) from backscattered diffuse or sub-diffuse light. Diagnostic instruments typically measure the tissue reduced scattering coefficient μ_s’, equal to the scattering coefficient μ_s multiplied by (1 - g), where g is the anisotropy. Since μ_s and g are both wavelength-dependent but vary with wavelength in different ways, biophysically-important information related to spectral variations in μ_s and g is lost when μ_s and g are combined into μ_s’. We performed a computational study to quantitatively illustrate the magnitude of this phenomenon. Wavelength-resolved μ_s and g spectra were modeled using a publicly available Mie Scattering Calculator for different biologically-relevant scatterer sizes. Substantial differences were observed between the μ_s(λ) and μ_s’(λ) spectra due to the wavelength dependence of g(λ). These results emphasize the continued need for diagnostic technologies to accurately measure, model, and interpret both μ_s(λ) and g(λ) to better quantify signatures of tissue disease.