Research Article

Impact of skin pores on photon transport in near–infrared optical tissue imaging

¹ Department of Biomedical Device Technology, Vocational School, Acibadem Mehmet Ali Aydinlar University, Istanbul 34752, Türkiye
² CASE (Center of Advanced Simulation and Education), Acibadem Mehmet Ali Aydinlar University, Istanbul 34752, Türkiye

^* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Received: 28 January 2026
Accepted: 7 May 2026

Abstract

Near-infrared optical tissue imaging is sensitive to both the optical properties of biological media and their microstructural geometry. While macroscopic tissue characteristics are well studied, the quantitative impact of skin pores on photon propagation remains largely unexplored. Here, we investigate this influence using a cascaded computational framework. Anatomically realistic tissue geometries were reconstructed from segmented MRI data of eight cadaveric heads, consisting scalp, skull, cerebrospinal fluid, and brain layers. Ballistic photon propagation simulations first resolved geometric optical interactions at the skin surface with explicit pore microgeometry, and the resulting photon states initialized Monte Carlo-based diffusive photon transport simulations to produce voxel-wise fluence maps and depth-resolved sensitivity analysis. Results indicate that skin pores increase photon angular divergence during the ballistic phase, but these directional perturbations are rapidly randomized by multiple scattering and do not measurably alter depth sensitivity (DS) profiles or photon path statistics. In contrast, pore-induced reductions in photon weight persist, decreasing the photon budget available for deeper tissue transport. These results indicate that surface microgeometry primarily affects optical coupling efficiency rather than stochastic photon propagation, providing guidance for when pore-scale features can be neglected or should be explicitly incorporated in biomedical optical system design.