Robust and Automated Lung Nodule Diagnosis from CT Images Based on Fuzzy Systems

Lung cancer is the most common fatal malignancy in both men and women. Early detection and treatment of lung cancer can greatly improve the survival rate of patient. The present work describes the design and development of a two stage computer-aided diagnosis (CAD) system that can automatically detect and diagnose histological images such as CT scan of lung with a nodule into cancerous or non-cancerous nodule. In the first stage the input image is pre-processed and the cancerous nodule region is segmented and the second stage involves in diagnosis of the nodal based on fuzzy system based on the area and the grey level of the nodule region. The aim of the proposed work is also to reduce the false positive classifications while maintaining a high degree of true- positive diagnosis. The proposed method attains an accuracy of 90% with high true positive rates and also high detection sensitivity and specificity that can meet basically the requirement of clinical diagnosis.