GPU-accelerated polarization-sensitive Monte Carlo simulations for Diffuse Optical Tomography with polarized light

Hind Oulhaj (1) , Julien Wojak (2) , Ugo Tricoli (1) , Callum M. Macdonald (1) , Vadim A. Markel (1) , Anabela Da Silva (1)

1 DiMABio - DiMABio
2 FRESNEL - Institut FRESNEL

An efficient GPU-based Monte Carlo (MC) simulator for modeling polarized light propagation in scattering media is reported. By using reciprocity relations, the inverse problem for Diffuse Optical Tomography with polarized light is formulated within the first Born approximation. The approach allows a fast computation of the sensitivity kernels.

Mots clés: Polarized light; Monte Carlo simulations; v-RTE; Inverse problem; Diffuse Optical Tomography

Domaines: Optique [physics.optics]; Calcul parallèle, distribué et partagé [cs.DC]

Hind Oulhaj, Julien Wojak, Ugo Tricoli, Callum M. Macdonald, Vadim A. Markel, et al.. GPU-accelerated polarization-sensitive Monte Carlo simulations for Diffuse Optical Tomography with polarized light. SPIE Photonics Europe, Mar 2020, Strasbourg, France. ⟨10.1117/12.2526939⟩. ⟨hal-02951643⟩