基于局部特征分析的LDCT增强算法

Novel Algorithm for Low-dose CT Image Enhancement Based on Local Feature Analysis

针对CT影像灰度动态范围宽、对比度差的问题,提出一种基于局部特征分析的低剂量CT(Low-dose Computed Tomography,LDCT)影像增强算法.算法通过对影像全局和局部统计特征的分析,构造非线性变换函数,实现对活动子块所对应区域的局部动态范围拉伸.选用两组不同采集协议CT影像进行的对比实验表明,算法实现简捷,可有效增强影像中的细部解剖结构,并可较好地抑制LDCT影像中的线性伪影.算法已被用于肺癌计算机辅助诊断系统的预处理过程.

In order to improve the intensity and contrast qualities of CT （Computed Tomography） images in clinic application, this paper proposed a fast and smart low-dose CT images contrast enhancement algorithm based on local feature analysis. The method based on creating active sub-blocks with analyzing the local static feature for each local region. Then, perform a local modified contrast stretching according to an adaptive transfer function within the image region corresponds with the sub-block. The experimental results of two sets CT images with different acquisition protocol （one with 200-250mA tube current, and another with 80mA） show that the proposed algorithm provides a flexible and efficient way for low-dose CT image enhancement, enhances the detail anatomic structure effectively, and constrains the linear artifacts in low-dose CT images preferably than general method. Now, the method is used as a pre-processing procedure in a Lung Cancer CAD system.