Distinguishing Adenocarcinomas from Granulomas in the CT scan of the chest: performance degradation evaluation in the automatic segmentation framework

Abstract Objective The most common histopathologic malignant and benign nodules are Adenocarcinoma Granuloma, respectively, which have different standards of care. In this paper, we propose an automatic framework for the diagnosis Adenocarcinomas Granulomas in CT scans chest from a private dataset. We use radiomic features attached vessel tortuosity diagnosis. dataset includes 22 CTs each nodule type, i.e., adenocarcinoma granuloma. contains non-smoker patients who between 30 60 years old. To automatically segment delineated area vessels area, apply morphological-based approach. For distinguishing malignancy segmented nodule, two texture curvature Mean number extracted. Results compare our with state-of-the-art feature selection methods differentiating Granulomas. These employ only shape or torsion along nodule. accuracy is improved by considering four selected features.