Quantitative optimization of source and detector configurations in diffuse optical tomography system

In diffuse optical tomography system, optimized source and detector configurations can enhance the capacity of noise resistance and improve the sensitivity of signal to localized changes in imaging domain, thus can improve reconstructed image quality. In this report, we introduce a rigorous and computational efficient method based on Cramer-Rao lower bound to quantitatively and simultaneously evaluate the precision limits of the reconstructed value and depth of a single target without solving the inverse problem. By choosing the sets of source and detector with low precision limits, high possibilities of achieving better image reconstruction results are expected. To test the effectiveness of our method, simulations were conducted for a given number of sources and detectors on a planar probe surface, with different signal noise levels and different perturbation depths in the reflection geometry. It was demonstrated that the source and detector sets selected by this method afforded better reconstructed images.