Computational Optics & Translational Imaging Lab

GPU-based Photon Transport Model

Monte Carlo based photon simulations

Biophotonics techniques play increasingly important roles in drug discovery, disease diagnosis, medical interventions as well as fundamental science research. The success of many biophotonics techniques is critically dependent on the ability to model complex interactions between photons and biological tissues accurately and efficiently. In 2009, Dr. Fang published one of the first papers using graphics processing units (GPUs) to accelerate 3D photon transport simulations. His code demonstrated a revolutionary 300-1000x acceleration and was considered “game-changing” by Dr. Simon Arridge (UCL), a pioneer in optical tomography. In 2010, Dr. Fang published a new mesh-based Monte Carlo algorithm. This paper became the most downloaded paper for Biomedical Optics Express in 5 consecutive months in 2013! So far, these works have attracted over 550 citations, 1,000 registered users and 15,000 downloads worldwide. Our open-source Monte Carlo simulation packages are actively used by many optical imaging research labs across the world. Our registration data also indicates that over 30 global companies utilize our software in their development, including Canon, Sony and Ricoh. In 2015, we were awarded an NIH R01 grant to continue leading this large user community and the development of novel massively parallel computational techniques.