Detection Of Lung Cancer Nodule Using Digital Image Processing

Abstract

Lung nodule detection and segmentation is impor- tant for clinical diagnosis. Characteristics of pneumonic nodules always indicate the nature  of  lung  disease.  This  paper  uses  CT images for the lung nodule detection and segmentation.The identification stage includes pattern matching and confirmation to increase accuracy, performed by otsus algorithm, support vector machine, segmentation. The categorization stage includes matching characteristics (like texture, shape and density) of the detected nodules. Firstly, lung CT images are put through otsus algorithm for nodule detection. Secondly, lung nodules are de- tected inside the lung area using image processing techniques.And finally segmentation is applied to highlight the characteristics.