Automatic lung segmentation and lung nodule detection through High-Resolution Computed Tomography (HRCT) image is a new and exciting researchin the area of medical image processing and analysis. In this research, two newtechniques for segmentation of lung regions and extraction of nodules on theHRCT image are proposed. An automatic lung segmentation system is proposed foridentifying the lungs in HRCT lung images. First, lung regions are extracted fromthe HRCT images by grey-level thresholding. The lung background information iseliminated by linear scans originating from border pixels. Finally, lung boundariesare smoothed along the mediastinum. The lung nodule extraction from the HRCTimage is processed based on a set of continuous HRCT slices of lung images. In thefirst stage, the abnormal areas are extracted based on nodule pixel collection andcombination. In the final stage, the abnormal area is extracted by comparing thedensity and shape profile. Both of the systems have been tested by processing datasets from 10 continuous image sets (100 images). Lung segmentation results arepresented by comparing our automatic method to manually traced borders.Averaged over all results, the accuracy of lung segmentation is 96.10%. Theproposed nodule detection method has been tested on image sets containing healthyand unhealthy lung images. Statistical analysis has been done and the results showthe overall nodule detection rate is 88.44% along with the false positive rate of 0.18.
展开▼
