中文摘要：

采用基于光子输运模型的扩散层析成像（DOT）实现在近红外脑功能成像的诸多方案具有良好的定量研究潜力,但由于其逆问题固有的病态性,传统的全三维重建存在精度和分辨率低以及计算量大等缺点,因此根据实际问题约束和简化重建过程对改善成像性能具有重要意义.根据人体头部的解剖结构及近红外光在人脑中的传输特性,提出了改善逆问题病态性的双网格半三维图像重建算法,即在正问题中采用密集网格剖分下的全三维有限元计算提高计算精度,在逆问题中采用稀疏网格剖分且在同一脑组织层进行二维重建方法.模拟结果表明,对于单目标重建,双网格半三维重建算法与全三维重建算法相比,量化度提高30%、重建速度快4~12倍;对于双目标重建在信噪比30,d B以上的情况,半三维重建算法在CCS=12.5,mm的空间分辨能力与全三维重建算法CCS=20,mm的空间分辨能力相当.

英文摘要：

Diffuse optical tomography（DOT）based on the photon transport model provides good quantitative potential for near infrared cerebral functional imaging. Due to the inherent ill-posed inverse problem,the disadvantages of the traditional 3D reconstruction algorithm are low-precision,low-resolution and a big burden in calculation time.Therefore,the simplification and constraint of the reconstruction procedure according to the practical situation have significant meaning. Based on the anatomic structure of head and the diffusive nature of near infrared light propagation in the brain,the semi-3D image reconstruction algorithm was proposed to alleviate the ill-posed problem. Namely,the dense mesh of 3D finite element method is used to improve the calculation accuracy in forward problem,and the sparse mesh of 2D image reconstruction algorithm is used in inverse problem by assuming that the optical properties in grey matter are invariable along the depth. Simulation results show that,for the single target sample,the quantitativeness ratio from the semi-3D reconstruction algorithm is 30% higher than that from the 3D reconstruction algorithm,and the reconstruction speed by using the semi-3D reconstruction algorithm is 4—12,times faster than that by using the 3D reconstruction algorithm. Additionally,for the two targets sample,the spatial resolution of the semi-3D reconstruction algorithm at CCS of 12.5,mm is almost the same as that of the 3D reconstruction algorithm at CCS of 20,mm with the value of SNR≥30.