Corrected near infrared spectroscopy, C-NIRS: An optical system for extracting hemodynamic signatures unique to the brain

We propose a method, dubbed Corrected Near Infrared Spectroscopy (C-NIRS), to isolate absorption trends confined to the lower layer of a two-layer turbid medium, as is desired in near-infrared spectroscopy (NIRS) of cerebral hemodynamics. The theory behind the operation of this method has been developed and discussed. Several two-layer Monte-Carlo simulations of NIRS time series were generated using a physiologically relevant range of optical properties. Initial results show that by measuring absorption trends at two source-detector separations and performing a least-squares fit of one to the other, processed signals strongly resemble the simulated absorption properties unique to the bottom-layer. Through this approach, it has been demonstrated that fitting coefficients can be estimated without any a priori knowledge of the optical properties present in the model. An analytical approximation for the least squares coefficient provides physical insight into the nature of errors and suggests ways to reduce them.;Next, a multi-detector, continuous wave, near infrared spectroscopy system has been developed to examine whether the hemodynamics of the scalp and brain in adults contain significant layer-like hemodynamic trends. NIRS measurements were made using contrasting geometries, one with four detectors equidistant from a source 33 mm away, and one with detectors collinear with the source (5-33 mm away). When NIRS time series were acquired over the prefrontal cortex from resting adults using both geometries, variations among the time series were consistent with a substantially homogeneous two-layer model ( p < 0.001) and inconsistent with one dominated by heterogeneities. Additionally, when time series measured 5 mm from the source were subtracted from corresponding 33 mm signals via a least-squares algorithm, 60% of the hemoglobin changes were on average removed. These results suggest that hemodynamic trends present in the scalp can contribute significantly to NIRS measurements, and that attempts to reduce this noise by subtracting a simultaneous near-channel measurement using a two-layer model are justified.;Finally a full implementation of the C-NIRS method in a functional study was evaluated on a small human subject population. For this study, a multi-channel instrument designed for regional mapping of the head (coverage area of 4X8 cm) was developed and will be discussed in detail. C-NIRS was directly compared with data processed under the standard cw NIRS method through the use of two performance metrics. C-NIRS was consistently able to produce functionally specific hemodynamic responses which were cleaner (in terms of signal-to-noise) and more robust (in terms of statistical significance) than those produced by NIRS in each individual subject. Based on the results of this study, it was noted that the magnitude of improvement varied greatly from subject to subject.;The scope of this thesis was to develop an instrument based on a simple, yet physiologically relevant model, and evaluate its performance. At the conclusion of this thesis, C-NIRS has demonstrated a consistent and substantial improvement in measuring hemodynamic signatures unique to the brain over that offered by established cw NIRS methods. There are many additional avenues where this research can be advanced even further. Some of these future directions are considered.