Abstract Transcranial photobiomodulation (tPBM) shows promising therapeutic potential in mouse models of Alzheimer's disease (AD); however, its clinical efficacy in humans remains limited. This translational gap is primarily driven by underlying interspecies differences in head anatomy and light‐propagation mechanisms that profoundly affect treatment responses. To investigate this translational barrier, we developed a 3D computational framework to compare light propagation in anatomically realistic mouse and human head models, aiming to address the critical “from animal to human” challenge in translational tPBM. The Monte Carlo for Voxelized Media was validated using multiple approaches. Head models were constructed using mouse magnetic resonance imaging and the Visible Chinese Human dataset. We quantified light fluence and photon absorption as key indicators of therapeutic effects in AD models. Light propagated more uniformly and with higher penetration efficiency in mouse heads. In contrast, fluence was markedly attenuated in deeper regions of the human brain. Photon absorption was predominantly superficial in humans but demonstrated a homogeneous profile in mouse models. Considering safety constraints, a 4.0 cm Gaussian beam at 810 nm was identified as optimal, providing the highest hippocampal coverage among the tested settings. These quantitative results systematically reveal the magnitude of cross‐species disparities in photon propagation, demonstrating that specific beam optimization is essential for deep‐brain targeting. Consequently, cross‐species differences must be carefully considered in the clinical translation of phototherapy. This study provides a computational framework to guide the design of human tPBM protocols for personalized AD treatment.
Yang et al. (2026) studied this question.