Device-tissue interface geometry influences both the intensity of detected fluorescence and the extent of tissue sampled. Previous modelling studies have often investigated fluorescent light propagation using generalised tissue and illumination-collection geometries. However, the implementation of approaches that incorporate a greater degree of realism may provide more accurate estimates of light propagation. In this study, Monte Carlo modelling was performed to predict how illumination-collection parameters affect signal detection in multilayer tissue. Using the geometry and optical properties of normal and atherosclerotic aortas, results for realistic probe designs and a semi-infinite source-detection scheme were generated and compared. As illumination-collection parameters, including single-fibre probe diameter and fibre separation distance in multifibre probes, were varied, the signal origin deviated significantly from that predicted using the semi-infinite geometry. The semi-infinite case under-predicted the fraction of fluorescence originating from the superficial layer by up to 23% for a 0.2 mm diameter single-fibre probe and over-predicted by 10% for a multifibre probe. These results demonstrate the importance of specifying realistic illumination-collection parameters in theoretical studies and indicate that targeting of specific tissue regions may be achievable through customisation of the illumination-collection interface. The device- and tissue-specific approach presented has the potential to facilitate the optimisation of minimally invasive optical systems for a wide variety of applications.
Fibre-optic probeFluorescence spectroscopyLight-tissue interactionMonte Carlo modellingOptical diagnostics