Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford, Oxford, United Kingdom.
Ultrasound Med Biol. 2010 Feb;36(2):234-49. doi: 10.1016/j.ultrasmedbio.2009.10.001.
This paper describes the application of signal processing techniques to improve the robustness of ultrasound feedback for displaying changes in temperature distribution in treatment using high-intensity focused ultrasound (HIFU), especially at the low signal-to-noise ratios that might be expected in in vivo abdominal treatment. Temperature estimation is based on the local displacements in ultrasound images taken during HIFU treatment, and a method to improve robustness to outliers is introduced. The main contribution of the paper is in the application of a Kalman filter, a statistical signal processing technique, which uses a simple analytical temperature model of heat dispersion to improve the temperature estimation from the ultrasound measurements during and after HIFU exposure. To reduce the sensitivity of the method to previous assumptions on the material homogeneity and signal-to-noise ratio, an adaptive form is introduced. The method is illustrated using data from HIFU exposure of ex vivo bovine liver. A particular advantage of the stability it introduces is that the temperature can be visualized not only in the intervals between HIFU exposure but also, for some configurations, during the exposure itself.
本文介绍了信号处理技术的应用，以提高超声反馈在显示高强度聚焦超声（HIFU）治疗中温度分布变化的稳健性，特别是在体内腹部治疗中可能出现的低信噪比情况下。温度估计基于 HIFU 治疗期间采集的超声图像中的局部位移，并引入了一种提高抗异常值能力的方法。本文的主要贡献在于应用了卡尔曼滤波器，这是一种统计信号处理技术，它使用简单的热扩散分析温度模型，从 HIFU 暴露期间和之后的超声测量中提高温度估计。为了降低该方法对先前关于材料均匀性和信噪比假设的敏感性，引入了自适应形式。该方法使用离体牛肝 HIFU 暴露的数据进行说明。该方法引入的稳定性的一个特别优点是，不仅可以在 HIFU 暴露之间的间隔期间，而且对于某些配置，还可以在暴露期间本身可视化温度。
