New segmentation method improves lung cancer diagnosis


May 13, 2013  A novel approach to lung cancer imaging, using a highly automatic, accurate and reproducible lung tumor delineation algorithm, improves diagnosis and prognosis assessments, according to a recent study.1

Once imaging data is acquired from computed tomography (CT), positron emission tomography (PET) or magnetic resonance imaging (MRI) diagnostic technologies, the method uses a single click ensemble segmentation (SCES) approach based on an existing "Click&Grow" algorithm to help segment and extract features of a tumor. The SCES approach requires only one operator selected seed point as compared with multiple operator inputs, which are typically needed.1 

In a joint collaboration, researchers from Moffitt Cancer Center and the University of South Florida, and medical institutions in China, the United Kingdom, the Netherlands, and Germany, used SCES to evaluate a set of 129 CT lung tumor images using a similarity index (SI).


SCES incorporates the original seed point to define an area within which multiple seed points are automatically generated. Ensemble segmentation can then be obtained from the multiple regions. The measurement can be used to determine if the tumor is increasing or decreasing in size, as well as describe features such as shape and texture.

The researchers concluded that the newly developed automatic lung lesion segmentation algorithm is stable, accurate and automated.

“A big advantage with single click ensemble segmentation is that it only requires one human interaction – the manual seed input. This is when the radiologist or radiation oncologist places the seed points in the tumor area,” said senior author Robert J. Gillies, PhD, chair of the Department of Cancer Imaging and Metabolism at Moffitt. “With SCES, lesion delineation was accurate and consistent, and the lung segmentations workload was greatly reduced.”


Gu Y, Kumar V, Hall LO, et al. Automated Delineation of Lung Tumors from CT Images Using a Single Click Ensemble Segmentation Approach. Pattern Recognit. 2013;46:692-702.

For more information:, and


Back To Top

May 13, 2013

Add A Comment


Allowed HTML: <b>, <i>, <u>, <a>


Copyright © Anderson Publishing 2015