医学超声波图像的边界保护滤波及特征映射补偿

提出的补偿方法不依赖于对组织数据的初始化分类,而是从原始数据中抽取组织特征映射到二维可视化平面.首先对图像进行边界保护滤波,然后归纳出4种基于图像灰度信息的特征映射补偿方法.实验结果表明,通过补偿方法得到的图像可以辅助医生对病变体及其它感兴趣的区域进行定性分析,从而提高临床诊断的准确性.

Boundary Detection of Medical Ultrasound Images and Rendering of Filtered Images Using Direct Feature Mapping

In this paper we propose several rendering methods that are not relied on any classification into different tissue categories. Instead of that, features are extracted from the original data-set and projected onto the view plane. After using boundary detection, we summarize four new rendering methods that based on the intensity of the image. We found that some of these methods may give clinically useful information that is very important for the doctor to make a correct diagnosis.