Validation of a near-infrared probe for detection of thin intracranial white matter structures

Object. The authors have developed an intracranial near-infrared (NIR) probe that analyzes the scattering of light emitted from its tip to measure the optical properties of cerebral tissue. Despite its success in distinguishing gray matter from white matter in humans during stereotactic surgery, the limits of this instrument's resolution remain unclear. In this study, the authors determined the spatial resolution of this new probe by using a rodent model supplemented with phantom measurements and computer simulation. Methods. A phantom consisting of Intralipid and gelatin was constructed to resemble a layer of white matter overlying a layer of gray matter. Near-infrared measurements were obtained as the probe was inserted through the gray-white matter transition. A computer simulation of NIR measurements through a gray-white matter transition was also performed using Monte Carlo techniques. The NIR probe was then used to study 19 tracks from the cortical surface through the corpus callosum in an in vivo rodent preparation. The animals were killed and histological sections through the tracks were obtained. Data from the phantom models and computer simulations showed that the NIR probe samples a volume of tissue extending 1 to 1.5 mm in front of the probe tip (this distance is termed the "lookthrough" distance). Measurements obtained from an NIR probe passing through a thin layer of white matter consisted of an initial segment of increasing values, a maximum (peak) value, and a trailing segment of decreasing values. The length of the initial segment is the look-through distance, the position of the peak indicates the location of the superficial white matter boundary, and the length of the trailing segment is the thickness of the layer. These considerations were confirmed in experiments with rodents. All tracks passed through the corpus callosum, which was demonstrated as a broad peak on each NIR graph. The position of the dorsal boundary of the corpus callosum and its width (based on histological measurements) correlated well with the peak of the NIR curve and its trailing segment, respectively. The initial segments correlated well with estimates of the lookthrough distance. Five of the tracks transected the smaller anterior commissure (diameter 0.2 mm), producing a narrow NIR peak at the correct depth. Conclusions. Data in this study confirm that the NIR probe can reliably detect and measure the thickness of layers of white matter as thin as 0.2 mm. Such resolution should be adequate to detect larger structures of interest encountered during stereotactic surgery in humans.