项目名称: HIFU引导超声图像的斑点噪声抑制和测温方法研究
项目编号: No.61002044
项目类型: 青年科学基金项目
立项/批准年度: 2011
项目学科: 数理科学和化学
项目作者: 闫晟
作者单位: 中国科学院声学研究所
项目金额: 7万元
中文摘要: 治疗中的图像引导方式和无损测温技术是当前HIFU 研究中的热点。经济准确的图像引导治疗,结合无损测温技术,可以在临床治疗中有效控制HIFU 的投放位置和剂量,使得HIFU 在临床中的应用更为安全有效。本项目开展对HIFU 引导B 模式超声图像中斑点噪声抑制方法的研究,利用小波和Contourlet 变换在多尺度对图像中包含的信号和噪声进行分析。通过对信号和噪声在分解系数中的先验信息以及系数在尺度内和尺度间相关性的研究,实现有效的信号和噪声的分离,使得在对超声图像进行降噪处理时,有效的保留图像中表示温度改变的灰度变化。为实现对HIFU 靶区的无损测温,本课题对超声图像的分布模型进行研究,利用模型参数值和期望值的改变,对靶区在治疗中的灰度变化进行精确定量描述,并实现基于超声图像模型参数值和期望值靶区的HIFU 无损测温方法。
中文关键词: 高强度聚焦超声;超声图像;噪声抑制;无损测温
英文摘要: Image guidance and noninvasive temperature estimation methods are research focuses in HIFU techniques currently. Inexpensive image guidance and accurate noninvasive temperature estimation methods that monitor and control the delivered HIFU dose and positions can make the HIFU therapies more effective and securer. This project proposes to apply wavelet and contourlet transforms in the multiresolution analysis of the signal and noise to extend speckle noise reduction methods of the guided ultrasound image during HIFU therapies. In order to preserve the temperature information of gray level change in ultrasound images in the speckle reduction processing, the prior probabilities of the decomposition coefficients of the signal and noise and the spatial correlations in scales and across scales are researched to achieve effective separation of the signal and noise to avoid destroying temperature information. To achieve noninvasive temperature estimation in HIFU target zone, this project apply several distribution models to represent the guided ultrasound image, and use the model parameters and ROI expectation value change to quantitatively and accurately describe gray level change caused by temperature rising during HIFU therapies.
英文关键词: HIFU; Ultrasound Image; Noise Reduction; Noninvasive Temperature Estimation