IMPROVEMENT OF DEPTH SENSITIVITY TO CEREBRAL HEMODYNAMICS W/ TIME GATED SYSTEM

点击翻译按钮获取中文摘要

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. By recording the distribution of times of flight of photons, Time Domain (TD) systems intrinsically provide more information that continuous wave (CW) ones. In particular, they enable depth discrimination at a single source-detector (SD) separation. This is of particular interest for functional brain imaging, where cortical activation is often hidden by superficial systemic response. Our TD device is based on a Ti:Sapphire pulsed laser and an intensified CCD camera (ICCD). We have developed a probe combining depth sensitivity and 2D imaging. It consists in 2 halves  one per hemisphere  each with 4¿4 sources and 3¿3 detectors in a square geometry (SD = 2.5 cm). Each detector consists of 7 fibers of different lengths, for parallel detection at 7 delays. All 126 fibers are imaged in parallel on the ICCD array. The 32 sources are illuminated sequentially, with 4 sets of 8 sources that are turned on during the same CCD frame without causing significant cross-talk. This source time-multiplexing and the parallel detection allow for an imaging frequency of almost 2 Hz for the whole head. After successful demonstration of the systems improved depth penetration in phantoms and humans, we are now developing a linear 3D image reconstruction, using the information at all delay gates and all SD pairs simultaneously to better realize the full potential of this new technology. The forward sensitivity matrix A (size number of voxels ¿ [number of SD pairs ¿ number of delay gates]) relates the changes in the absorption coefficient ¿¿a to the changes in normalized intensity ¿I/I0 in the normalized Born approximation: ¿I/I0 = A . ¿¿a. The reconstructed image is obtained by inversion of the sensitivity matrix, using the covariance matrix in the regularization. This reconstruction enables both depth resolution and better lateral uniformity than CW data. A manuscript on these reconstruction results is being prepared. We will be applying the reconstructions to more human data over the next years.