Fujii M, Nakayama K
Faculty of Science and Technology, Sophia University, 7-1 Kioicho, Chiyoda, Tokyo 102-8554, Japan.
Phys Med Biol. 2009 Mar 21;54(6):1419-33. doi: 10.1088/0031-9155/54/6/003. Epub 2009 Feb 13.
A compact filtered backprojection algorithm that suppresses the undesirable effects of skin circulation for near-infrared diffuse optical topography is proposed. Our approach centers around a depth-selective filtering algorithm that uses an inverse problem technique and extracts target signals from observation data contaminated by noise from a shallow region. The filtering algorithm is reduced to a compact matrix and is therefore easily incorporated into a real-time system. To demonstrate the validity of this method, we developed a demonstration prototype for depth-selective diffuse optical topography and performed both computer simulations and phantom experiments. The results show that the proposed method significantly suppresses the noise from the shallow region with a minimal degradation of the target signal.
提出了一种紧凑的滤波反投影算法，用于抑制近红外漫射光学断层成像中皮肤循环的不良影响。我们的方法围绕一种深度选择滤波算法展开，该算法使用反问题技术，从受浅区域噪声污染的观测数据中提取目标信号。该滤波算法简化为一个紧凑矩阵，因此易于集成到实时系统中。为了证明该方法的有效性，我们开发了一个深度选择漫射光学断层成像演示原型，并进行了计算机模拟和体模实验。结果表明，所提出的方法能显著抑制浅区域的噪声，同时目标信号的退化最小。
