论文部分内容阅读
肺结节的智能识别对肺癌的诊断至关重要。为了在大量的肺部CT图片中准确智能识别肺结节,我们研究了一个基于规则及多特征跟踪的肺结节计算机辅助检测方法。其中,采用活动轮廓模型的分割方法实现候选肺结节分割,采用基于规则的决策方法以及多特征跟踪方法实现肺结节分类。实验证明,该肺结节的智能检测方法满足肺结节计算机辅助诊断的要求。
Intelligent identification of lung nodules is crucial for the diagnosis of lung cancer. To accurately and intelligently identify lung nodules in a large number of lung CT images, we studied a computer-aided detection of lung nodules based on regular and multi-feature tracking. Among them, segmentation of the active contour model is used to implement the segmentation of the candidate pulmonary nodules, and the classification of pulmonary nodules is implemented by the rule-based decision method and the multi-feature tracking method. Experiments show that the intelligent detection of pulmonary nodules meet the requirements of computer-aided diagnosis of pulmonary nodules.