Accession Number : AD1025969

Title :   Noninvasive Characterization of Indeterminate Pulmonary Nodules Detected on Chest High-Resolution Computed Tomography

Descriptive Note : Technical Report,30 Sep 2015,26 Sep 2016

Corporate Author : Vanderbilt University Nashville United States

Personal Author(s) : Maldonado,Fabien

Full Text :

Report Date : 01 Oct 2016

Pagination or Media Count : 28

Abstract : Purpose: Lung cancer is the most common cause of cancer-related deaths in the US. Results from the National Lung Screening Trial (NLST), a large randomized controlled trial, suggest that screening with annual low-dose, high-resolution computed tomography of the chest (HRCT) reduces lung cancer specific mortality by 20 . The major challenge for the implementation of lung cancer screening is the high false positive rate. In the NLST, 40 of the participants in the HRCT were found to have lung nodules, 95 of which proved benign. This high false positive rate limits the applicability of this strategy at the population level, and could result in increased patient anxiety, radiation exposure, health care costs, and procedural morbidity and mortality. Our work aims at identifying HRCT-based and clinical variables to derive a model which will non-invasively distinguish benign from malignant nodules.

Descriptors :   LUNG DISEASES , xray computed tomography , discriminant analysis , predictive modeling , cancer screening , identification , carcinoma , neoplasms , high resolution , computer vision , supervised machine learning , NONINTRUSIVE INSPECTION , biological detection

Distribution Statement : APPROVED FOR PUBLIC RELEASE