Abstract

We have developed a rapid multi-spectral endoscopic imaging system for real-time mapping and quantifications of mucosa's blood supply namely total hemoglobin concentration (blood volume) and hemoglobin oxygen saturation (tissue oxygenation index). The system was designed to acquire and analyze 18 optimized spectral bands spanning the 400-760 nm spectral range using six triple-band filters and a standard color CCD camera. Imaging speed of 15 spectral image cubes/second was achieved. Pre-processing algorithm and method is used to calibrate the spectral response of the color CCD camera to perform accurate spectral measurements. Spectral analysis algorithm using linear matrix inversion approach was developed and implemented with graphics processing unit (GPU) programming to produce rapid mapping of the mucosa optical properties. The efficacy and validity of the developed device and method was assessed using in-vivo Endoscopic lung measurements obtained during preliminary clinical trial. The results showed that the device and method can acquire spectral images (~250×250 pixel resolution) and extracts the hemoglobin concentration and the hemoglobin oxygen saturation index maps, in near real-time and with accuracy comparable to that obtained from spectroscopy measurements.

© 2015 Optical Society of America