Solitary pulmonary nodules classification based on tumor size and volume of nodules

Among five main type of cancer lung cancer is one of causing health hazards in both men and women all over the world. Advanced techniques of Computed Tomography and medical images play an important role in clinically detection of lung cancer tumors in all TNM stages. Efficient Computer Aided Detection (CADe) systems help the radiologist in early detection and diagnosis of lung cancer. The objective of this paper is to develop efficient CADe system using iterative thresholding method for segmentation and freeman chain code algorithm to repair the boundary of separated lung regions. Region growing algorithm is used to extract tumor region from lung regions. Tumor shape, size, whole tumor volume and solid part tumor volume are important factors. These factors are computed in this research work to determine prognosis of tumor. Developed CADe system is evaluated using CT thoracic lung images from Lung Image Database Consortium and Reference Image (LIDC) and Reference Image Database to Evaluate Response (RIDER).