Detection Algorithms Of Pulmonary Nodules In CT Images

Lung cancer is the most ordinary malignant tumor nowadays, and the early diagnosis and treatment can significantly increase the chance of survival for patients. CT scanning has been the most important method of lung cancer examination, which is excellent diagnosis tool for early lung cancer. Lung nodules are the major indications of lung cancer on CT images. Since the complexity of the lung structure and the multiplicity of nodule's shape, even the seasoned doctor cannot easily discover all of the possible disease in time. Furthermore, the huge amounts of CT scanning pictures, especially the high resolution pictures which could produce more than 300 slices from only one patient, also will bring very large work burden and difficulty for the doctors. With the rapid development of computer and its correlative technology, it is possible to carry on the computer aided detection (CAD), which can help the doctor to understand and judge pictures precisely and further improve the detection accuracy.Lung nodules are some approximate circular sickness region whose diameter is less than 3cm. It always shows as compact regions on CT images. For effectively implementing the CAD of lung nodules, an integrated automatic detection algorithm of lung nodules is proposed in this paper. The detection scheme includes four steps. Based on the morphological analysis, an effective image enhancement method is presented on the characteristic of lung CT images, which can obtain preferable filtering result. For getting the whole lungs parenchyma, the improved GMM is employed to segment the images preliminarily, which combined with the region growing and morphology method. According to the basic image characteristics such as intensity and contrast, the plump seed regions can be extracted from the region of interest (ROI). On the basis of these seeds, the whole ROI can be extracted by the proposed adaptive three-dimensional region growth algorithm. Owing to a great deal of false positives lie in ROI, the SVM classifier is designed to distinguish nodules from normal areas with good detected result after the effective features extraction.The experimental results show that the proposed detection algorithms can obtain good detection result. It is believed that these detection algorithms will be of extensive application prospect.