Lung cancer detection based on helical CT images using curved surface morphology analysis

摘要: Lung cancer is known as one of the most difficult cancers to cure. The detection lung in its early stage can be helpful for medical treatment limit danger. A conventional technique that assists uses helical CT, which provides information 3D cross sectional images lung. We expect proposed will increase diagnostic confidence. However, mass screening based on CT leads a considerable number diagnosis, this time-consuming fact makes it used clinic. To efficiency process, we had computer-aided diagnosis (CAD). In paper, describe Images using curved surface morphology analysis. Firstly, extract area from original image. Secondly, compute shape index value area. Thirdly, ROI (Region Of Interest) computed value. Finally, apply rule neural network and detect suspicious regions. show here result our algorithm applied 390 patients.