Visualization of mucosal vasculature with narrow band imaging: A theoretical study

Narrow band imaging (NBI) is a spectrally-selective reflectance imaging technique that is used as an adjunctive approach to endoscopic detection of mucosal abnormalities such as neoplastic lesions. While numerous clinical studies in tissue sites such as the esophagus, oral cavity and lung indicate the efficacy of this approach, it is not well theoretically understood. In this study, we performed Monte Carlo simulations to elucidate the factors that affect NBI device performance. The model geometry involved a two-layer turbid medium based on mucosal tissue optical properties and embedded cylindrical, blood-filled vessels at varying diameters and depths. Specifically, we studied the effect of bandpass filters (415±15 nm, 540±10 nm versus white light), blood vessel diameter (20-400 μm) and depth (30-450 μm), wavelength, and bandwidth on vessel contrast. Our results provide a quantitative evaluation of the two mechanisms that are commonly believed to be the primary components of NBI: (i) the increased contrast provided by high hemoglobin absorption and (ii) increase in the penetration depth produced by the decrease in scattering with increasing wavelength. Our MC model can provide novel, quantitative insight into NBI, may lead to improvements in its performance.