基于3维图像增强的肺结节识别

Recognition of pulmonary nodules based on 3D image enhancement

孤立性肺结节的检测是肺癌早期诊断的关键。针对传统点增强滤波器虽然对结节增强具有很好的敏感性,但是却产生很多假阳性区域的问题,提出一种通过计算3维增强密度指数和判别规则来识别肺结节的方法。首先采用自适应双边滤波器对CT图像序列进行降噪和平滑处理;然后计算对应的Hessian矩阵及其特征值得到预增强系数,并获得感兴趣体区域,通过对预增强系数的分析来构造3维增强密度指数;最后应用判别规则对感兴趣体进行识别。针对两个肺部CT图像数据集对该方法进行了测试,结果表明,在识别孤立性肺结节方面该方法是有效的。

The detection of solitary pulmonary nodules （SPN） is proven to be of critical importance in early-stage lung cancer diagnosis. Aiming at reducing the false positive regions caused by the dot enhancement filter which is sensitive to the lung nodules, a new recognition method based on calculation of three-dimensional （3D） enhancement density index and decision rule is proposed. An adaptive bilateral filter is applied to reduce the noisy and smooth CT image sequences. Then, the pre-enhancement coefficients and volume of interest （VOI） are obtained by computing the Hessian matrix and corresponding eigenvalues. After analyzing the distribution of pre-enhaneement coefficients, 3D enhancement density index is constructed. Finally, a decision rule is adopted to identify nodule candidates. The proposed method is tested on two lung CT image sets. The experimental results illustrate the efficiency of the proposed algorithm.