Output details
11 - Computer Science and Informatics
Kingston University
Segmentation of pulmonary nodules in thoracic CT scans: a region growing approach
<28> This paper presents an efficient 3D Computer Aided Segmentation method for different types of pulmonary nodules (with high/low contrast, with vasculature attachment or in close vicinity of the lung wall/diaphragm). Tested on 815 nodules the segmentation results were visually inspected by a radiologist with a success rate above 95%. The results show the proposed method enhances radiologists’ performance and is a very cost effective diagnostic tool. It received approvals by FDA (USA) and CE (EU) and a high citation rate. The underlying algorithm is integrated into commercial applications used by companies such as Vital Images, Viatronix and Terarecon. -.-