中文摘要：

平均散射元间距（MSS）已被众多研究者用于分析准规则组织的微结构特征，平均骨小梁间距（MTBS）是诊断骨质疏松症的重要参数之一。本研究提出一种基于简易反向滤波跟踪（SIFT）的算法，用于估计平均骨小梁间距。对计算机仿真信号和离体牛胫骨松质骨的超声背散射信号分别进行分析处理。实验结果表明，SIFT算法能简单而有效地减小测量系统和传播介质特性的干扰，并且在较强噪声存在的情况下，SIFT算法比AR倒谱法的估计结果更为精确。因此，SIFT算法可用于估计平均骨小梁间距，从而反映松质骨的微结构特征。

英文摘要：

Mean scatterer spacing （MSS） has been used by many researchers to characterize the quasi-periodic tissue microstructures and mean trabocular bone spacing （MTBS） is an important parameter to diagnose osteoporosis. This paper proposed a new technique based on simplified inverse filter tracking （SIFT） algorithm to estimate MTBS. Both simulated backscattered signals and ultrasonic backscattered signals from bovine cancellous bone in vitro were processed. The results indicated that SIFT algorithm minimized the influence of measurement system and transmission medium property. It also provided a more reliable estimation in the presence of strong diffuse scatterers and noise. Thus the SIFT method is expected to be useful for providing information about the quasi-periodic tissue microstructures.