Calculation of hemoglobin saturation from in vivo human brain tissues using a modified diffusion theory model

During deep brain stimulation, a neurosurgical procedure to relieve tremors, a thin electrode is inserted into deep brain regions to provide stimulation. Accurate electrode placement is crucial to provide tremor suppression without damaging adjacent optical and motor regions. A portable, real-time display fiber optic reflectance probe is used to obtain reflected signals from living, human brain tissues. The optical results are compared to pre-operative MRI scans to confirm anatomical structures and verify electrode placement. In addition to reflectance, tissue oxygen saturation may assist brain tissue identification. A modified diffusion model has been developed and tested using Monte Carlo data for small source-detector separations between 500 to 600 nm for hemoglobin oxygen saturation levels of 100% and 0%. In turn, curve fitting between the spectral curves and the model was performed, producing errors of 5.24% and 5.43%, respectively, for cases of 100% and 0% hemoglobin saturation. Additional validation was performed through in vitro studies. Following validation, the model was then used to extract oxygen saturation values from in vivo reflectance spectra. This method permits calculations for hemoglobin saturation and absorption and scattering coefficients from in vivo tissues. Those parameters may be used as clinical markers for neurological localization in the future.