Can lung nodule scoring systems do better?

Sunday, November 29 | 11:05 a.m.-11:15 a.m. | SSA04-03 | Room S404CD
The rise of CT lung cancer screening has been accompanied by a few published lung nodule scoring systems, including Lung-RADS and the McWilliams model. But could the models be tweaked to perform better?

Researchers from the Netherlands will explore the possibilities in this Sunday session.

"Lung cancer screening is an important topic and has gained a lot of interest," presenter Dr. Sarah van Riel, from Radboud University Medical Center in Nijmegen, the Netherlands, told AuntMinnie.com. "In the past couple of years, various scoring systems for screen-detected nodules have been published. In our study, we were interested in comparing the performance of two well-described scoring systems to each other."

The researchers reviewed 60 cancers from the Danish Lung Cancer Screening Trial, along with a set of 120 randomly selected benign nodules of various sizes, all from different screening subjects. Seven readers scored morphological features for each nodule that referred to density distribution, lesion margins, lesion surroundings, and lesion architecture.

They found significant differences related to inhomogeneous density distribution and pleural/fissure retraction among the readers. Several morphological features are significantly associated with nodule malignancy but are not included in published risk-prediction models, the team concluded.

Page 1 of 654
Next Page