Diagnosis of Lung Nodule Using IndependentComponent Analysis in Computerized Tomography Images

摘要

This paper analyzes the application of Independent Component Analysis to the characterization of lung nodules as malignant or benign in computerized tomography images. The characterization method is based on a process that verifies which combination of measures, from the proposed measures, has been best able to discriminate between the benign and malignant nodules using Support Vector Machine. In order to verify this application we also describe tests that were carried out using a sample of 38 nodules: 29 benign and 9 malignant. The methodology reaches 100% of Specificity, 98.34% of Sensitivity and 96.66% of accuracy. Thus, preliminary results of this approach are very promising in contributing to pulmonary nodules diagnosis, but it will be necessary to test it in larger series and to make associations with other quantitative imaging methods in order to improve global performance.