光子学报, 2016, 45 (3): 0317001, 网络出版: 2016-04-01
基于漫反射模型的生物组织光学特性参量迭代反演
Iterative Inversion of Optical Property Parametersof Biological Tissue Based on Diffuse Reflectance Model
生物组织 漫反射模型 迭代反演 光学特性参量 相对误差 Biological tissue Diffuse reflectance model Iterative inversion Optical property parameters Relative error
摘要
提出一种基于迭代反演的输运平均自由程估计及光源-检测器最小距离确定方法.该方法利用迭代估计思想,自适应地改变光源-检测器最小距离,提高光学特性参量的反演准确度.对29组仿真数据的研究结果表明:迭代反演对输运平均自由程初始值不敏感,具有较高的鲁棒性,可在一定程度上提高吸收系数和有效散射系数的反演准确度.在无噪声的条件下,吸收系数反演的平均相对误差为7.17%,有效散射系数反演的平均相对误差为5.73%,与传统方法给定的光源-检测器最小距离下反演的最佳结果相比,分别降低了1.73%和1.14%.在加入信噪比为40~80 dB噪声的情况下,该方法仍然能获得较高的光学特性参量反演准确度,吸收系数误差的变动范围为8.46%~10.05%;有效散射系数误差的变动范围为6.79%~8.76%.
Abstract
A method based on the iterative inversion estimation of transport mean free path and the minimum source-detector distance determination was proposed. The method utilizes the idea of iterative estimation and adaptively changes the minimum source-detector distance to improve the inversion accuracy of optical property parameters. The study of 29 sets simulation data indicates that the iterative inversion is not sensitive to the initial value of the transport mean free path, and has a higher robustness. In a certain extent, the inversion accuracy of the two parameters can be improved. Under the condition of no noise, the average relative error of the absorption coefficient is 7.17%, and the average relative error of the reduced scattering coefficient is 5.73%. Compared with the best result of traditional empirical method for a given minimum source-detector distance, average relative error decreases 1.73% for the absorption coefficient and 1.14% for the reduced scattering coefficient. In the case of 40~80 dB for signal-to-noise ratio, the method can still obtain higher inversion accuracy of optical property parameters. The error ranges of the absorption coefficient and the reduced scattering coefficient are 8.46%~10.05% and 6.79%~8.76%, respectively.
王伟, 朱启兵, 黄敏. 基于漫反射模型的生物组织光学特性参量迭代反演[J]. 光子学报, 2016, 45(3): 0317001. WANG Wei, ZHU Qi-bing, HUANG Min. Iterative Inversion of Optical Property Parametersof Biological Tissue Based on Diffuse Reflectance Model[J]. ACTA PHOTONICA SINICA, 2016, 45(3): 0317001.