Lung Nodule Detection from Chest X-Ray Images Using Interval Type-2 Fuzzy Logic System

Lung nodule detection is a crucial task in lung cancer examination since early detection may lead to more successful treatment. In this work, a novel lung nodule detection algorithm based upon the interval type-2 fuzzy logic system is proposed. The method utilizes four features consisting of D-descriptors, the average intensity of the inside boundary, the circularity ratio, and HH diagonal component from the wavelet transform. The proposed method can promisingly detect the probable locations of nodules. The system produces 0.82 of true positive rate with 13.11 false positives per image.