基于改进LVQ算法的肺结节识别与分类

Recognition and classification of pulmonary nodules based on improved LVQ algorithm

摘要:

目前，肺癌的是发病率最高的肿瘤，若能在早期发现癌变并进行相应治疗，将极大的提高患者的生存率。肺癌的症状在早期表现为肺结节。以提高肺结节检测识别率并进行良恶性分类为目的，提出了一种改进的LVQ分类器算法。首先使用C-V算法对原始图像进行肺实质分割，再使用最优阈值法进行感兴趣区域提取，并进行特征提取和特征归一化。使用多次聚类算法检测肺结节。使用基于改进的LVQ分类器进行肺结节的良恶性进行分类。利用改进后的LVQ分类器在LIDC数据集上进行实验，得到了对良性结节的确诊率为87.3%，对恶性结节的确诊率为80.8%。实验结果表明，改进后的算法在良恶性结节分类上具有较高的确诊率，有助于提高医生的工作效率，实现肺结节的辅助发现。

Abstract:

Pulmonary nodule is the early manifestation of lung cancer. For the detection and classification of benign and malignant lung nodules, an improved LVQ classifier algorithm is proposed. Firstly, the C-V algorithm is used to do the lung parenchyma segmentation on the original images. Then, the optimal threshold method is used to extract the region of interest. Multiple clustering algorithms are used to detect pulmonary nodules after feature extraction and feature normalization. The classification of benign and malignant pulmonary nodules is carried out by an improved LVQ classifier. The experimental results prove that the improved algorithm has a higher rate of diagnosis on the classification of benign and malignant pulmonary nodules.