一种基于CT影像的肺癌病灶检测新方法

 肺癌病灶的检测一直是重要与困难的工作,本文提出了一种基于三维CT影像的肺结节计算机辅助检测新方法.基于自适应阈值等方法分割肺实质区域;由于肺血管是肺结节检测的重要干扰,建立一种形变模型精确分割并过滤肺内血管组织;基于Hessian矩阵特征值构造可选择形状滤波器检测疑似结节,并进一步过滤剩余的细小血管组织;提取多个结节特征,并采用基于规则分类器进行分类.实验结果表明,该方法可以有效帮助医生提高肺癌疾病的诊断准确率.

A New Lung Cancer Lesions Detection Scheme Based on CT Image

Lung cancer lesions detection has been an important and difficult work.A computer-aided detection (CAD) scheme for detecting lung nodules is proposed in three-dimensional CT images in this paper.The lung parenchyma is segmented from the CT data using adaptive threshold method etc.Pulmonary vascular is the main disturbance for nodules detection,building an active contour model to segment and remove pulmonary vascular accurately in the lung region.Suspicious nodules are detected and omitted renal vascular is filtered using a selective shape filter,which is based on the eigenvalues of a Hessian Matrix.Nodule features are extracted and rule-based classifier is used to distinguish true or false positive nodules.Experiment results indicate that this scheme can help physician improve the diagnosis efficiency.