World Congress Thoracic Imaging June 18-21, 2017, Hynes Convention Center, Boston, Massachusetts June 18-21, 2017, Hynes Convention Center, Boston, Massachusetts

Sponsoring Societies:


Fleischner Society Society of Thoracic Radiology European Society of Thoracic Imaging Japanese Society of Thoracic Radiology Korean Society of Thoracic Radiology
WCTI Home Congress Information Final Program

Back to 2017 Hot Topics


Lepidic Predominant Pulmonary Lesions (LPL): CT-based Distinction From More Invasive Adenocarcinoma Using 3D Volumetric Density and Histogram Analysis
J. B. ALPERT, H. RUSINEK, J. P. KO, H. I. PASS, B. K. CRAWFORD, A. RAPKIEWICZ, D. P. NAIDICH;
NYU LANGONE MEDICAL CENTER, NEW YORK CITY, NY.

Purpose: To evaluate surgically excised stage I adenocarcinomas, using CT-based differentiation of pathologically defined lepidic predominant lesions (LPL) such as minimally invasive adenocarcinoma (MIA) and lepidic predominant adenocarcinoma (LPA) from more invasive adenocarcinomas (INV).
Materials and Methods: The study was IRB approved and HIPAA compliant. A total of 64 cases of pathologically proven stage 1 adenocarcinoma, including 43 LPL and 21 INV, were evaluated using high resolution CT. Automated computer 3D image analyses included measurements of nodule volume, percent solid volume, and histogram features of skewness, kurtosis, entropy and mean nodule attenuation within each of 4 quartiles. A binomial logistic regression model was used to identify the best set of parameters distinguishing LPL from INV nodules.
Results: The accuracy of a logistic (binomial) model to discriminate LPL from INV nodules based on size and % solid volume had overall accuracy of 85.9%. With optimized probability cutoff, the model achieves 81% sensitivity and 76.7% specificity and an AUC of 0.897 (95%CI, 0.821-0.973). Similar accuracy was achieved with a model based on third quartile of nodule attenuation (Hu_Q3) histogram, with AUC of 0.877 (95%CI, 0.790-0.964).
Conclusions: Both CT-based 3D density calculations and histogram analysis of subsolid stage 1 lung adenocarcinomas allow differentiation of invasive carcinomas from lepidic predominant nodules.


Back to 2017 Hot Topics
Home | About | Contact