The paper presents a study of boundary conditions for the diffusion equation in radiative transfer, focusing on three common boundary conditions: partial-current, extrapolated-boundary, and zero-boundary. The authors use the method of images to examine these conditions and find that the partial-current and extrapolated-boundary conditions have the same dipole and quadrupole moments, leading to nearly identical solutions to the diffusion equation. This suggests that these two boundary conditions can be combined to create a unified approach for extracting optical parameters from frequency-domain photon-migration (FDPM) data. The study shows that the partial-current and extrapolated-boundary conditions yield values for the scattering and absorption coefficients that are equal to within 3%. The zero-boundary condition, on the other hand, leads to significant errors in the optical parameters. The authors also find that the phase and modulation data are highly sensitive to the reflectivity of the tissue surface, and a minimum reflectivity of 98% is needed to mimic an infinite-medium geometry. The paper concludes that noninvasive measurements of optically thick tissue require a rigorous treatment of the tissue boundary. The authors suggest a unified partial-current-extrapolated boundary approach, which is accurate and computationally fast. This approach is tested using FDPM data from tissue phantoms, and the results show that the partial-current and extrapolated-boundary conditions are almost indistinguishable, while the zero-boundary condition yields lower phase and higher modulation data. The study highlights the importance of considering the boundary conditions in radiative transfer applications, particularly in noninvasive optical measurements of biological tissues. The authors emphasize that the partial-current and extrapolated-boundary conditions are more accurate and reliable than the zero-boundary condition, and that the unified approach provides a more efficient and accurate method for extracting optical parameters from FDPM data.The paper presents a study of boundary conditions for the diffusion equation in radiative transfer, focusing on three common boundary conditions: partial-current, extrapolated-boundary, and zero-boundary. The authors use the method of images to examine these conditions and find that the partial-current and extrapolated-boundary conditions have the same dipole and quadrupole moments, leading to nearly identical solutions to the diffusion equation. This suggests that these two boundary conditions can be combined to create a unified approach for extracting optical parameters from frequency-domain photon-migration (FDPM) data. The study shows that the partial-current and extrapolated-boundary conditions yield values for the scattering and absorption coefficients that are equal to within 3%. The zero-boundary condition, on the other hand, leads to significant errors in the optical parameters. The authors also find that the phase and modulation data are highly sensitive to the reflectivity of the tissue surface, and a minimum reflectivity of 98% is needed to mimic an infinite-medium geometry. The paper concludes that noninvasive measurements of optically thick tissue require a rigorous treatment of the tissue boundary. The authors suggest a unified partial-current-extrapolated boundary approach, which is accurate and computationally fast. This approach is tested using FDPM data from tissue phantoms, and the results show that the partial-current and extrapolated-boundary conditions are almost indistinguishable, while the zero-boundary condition yields lower phase and higher modulation data. The study highlights the importance of considering the boundary conditions in radiative transfer applications, particularly in noninvasive optical measurements of biological tissues. The authors emphasize that the partial-current and extrapolated-boundary conditions are more accurate and reliable than the zero-boundary condition, and that the unified approach provides a more efficient and accurate method for extracting optical parameters from FDPM data.