The automated identification of small nodules in Computed Tomography (CT) lung scans represents an activity of our group. CT has been shown to be the best imaging modality for the detection of small pulmonary nodules, particularly after the introduction of the helical technology. The first Italian Randomized Controlled Trial has recently started (Italung-CT) to study the potential impact of a screening-based low-dose helical CT on the high-risk population. In this framework our group develops some automated approaches for the identification of small pulmonary nodules. Various approaches are developed and tested:
-
- detection of nodule candidates by means of a dotenhancement filter;
- detection of nodule candidates by mean of region growing;
- detection of non-pathological structure by means of artificial life model.
