BOLD效应fMRI图像的自适应阈值小波去噪方法.pdfVIP

170 2017 ，53（8） Computer Engineering and Applications 计算机工程与应用 BOLD 效应fMRI 图像的自适应阈值小波去噪方法 1 1 2 王泉德 ，肖继来 ，谢 晟 1 1 2 WANG Quande , XIAO Jilai , XIE Sheng 1. 武汉大学 电子信息学院，武汉 430070 2.北京中日友好医院，北京 100029 1.School of Electronic Information, Wuhan University, Wuhan 430070, China 2.China-Japan Friendship Hospital, Beijing 100029, China WANG Quande, XIAO Jilai, XIE Sheng. Wavelet denoising algorithm with adaptive threshold for fMRI images based on BOLD effect. Computer Engineering and Applications, 2017, 53 （8 ）：170-173. Abstract ：OEF can be quantitatively calculated using fMRI images based on BOLD effect and a two-component model, and it may be helpful in clinical prognosing and diagnosing of cerebrovascular disease. However, the Signal-Noise-Ratio （SNR）of the image is so poor that it ’s urgent to research and design efficient denoising algorithm in order to improve accuracy of OEF computing. In this paper, a wavelet denoising algorithm based on Bayesian estimating the adaptive threshold is designed and applied to analyse and denoise fMRI images based on BOLD effect, and result images are pro- vided to the post processing and computing OEF. Experimental result shows that the algorithm in this paper can improve accuracy of OEF computing effectively. Key words ：oxygen extraction fraction; image denoise; wavelet analysis; Bayesian estimation 摘 要：利用基于BOLD （Blood Oxygenation Level Dependent ）效应的fMRI 图像和两室模型可以定量计算脑氧摄 取分数（Oxygen Extraction Fraction ，OEF ），在脑血管病的预测和诊断上有较大的临床应用价值。但由于BOLD 效 应fMRI 图像的信噪比较低，研究并设计有效的BOLD 效应fMRI 图像去噪算法，从而提高OEF 计算结果的准确性是 急需解决的问题。因此，设计了基于贝叶斯估计的自适应阈值小波去噪方法对BOLD 效应fMRI 图像进行分析和去 噪，并将结果图像应用于OEF 值的计算。实验结果表明该方法能有效提高OEF 计算结果的准确性。 关键词：脑氧摄取分数；图像去噪；小波分析；贝叶斯估计 文献标志码：A 中图分类号：TP391 doi ：10.3778/j.issn. 1002-8331.1510-0069 1 引言 序列，完成了BOLD（