Predict false positive rate for lung cancer screening tests in the NLST data
Principal Investigator
Name
ping hu
Degrees
ScD
Institution
NCI
Position Title
Mathematical Statistician
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-45
Initial CDAS Request Approval
Jan 9, 2014
Title
Predict false positive rate for lung cancer screening tests in the NLST data
Summary
Cancer screening is a very commonly used tool in clinical and public health setting. Screening tests, however, are not error free and might produce false-positive results. Patients with false-positive results will be faced with unnecessary follow-up assessment, sometimes including invasive diagnostic procedures. There are several papers that study the false positive rates of screening tests. First, most of them focus on breast cancer. Second the false-positive rate for lung cancer screening has not been well investigated. More importantly there is no prior paper in which the false positive rate for lung cancer screening test has been studied using 3D voxel-wise image data.
Our primary goal is to use NLST 3D voxel-wise image data (in DICOM files) and machine learning models developed by Peng Huang, etc (2013) for image texture analysis to build prediction model that identify subjects who are at high risk to develop lung cancer. More specifically, we draw image motifs from selected region of interest and apply random forest and support vector machine to build classifiers. We will further develop risk scores using both image and non-image covariates. False positive rate of this image prediction will be estimated and compared with conventional manual image film reading method.
We propose to study 200 lung cancer cases and 200 control participants. Among them, 100 lung cancer cases and 100 control participants will be used as training set, and another 100 lung cancer cases and 100 control participants will be used as test set.
Our primary goal is to use NLST 3D voxel-wise image data (in DICOM files) and machine learning models developed by Peng Huang, etc (2013) for image texture analysis to build prediction model that identify subjects who are at high risk to develop lung cancer. More specifically, we draw image motifs from selected region of interest and apply random forest and support vector machine to build classifiers. We will further develop risk scores using both image and non-image covariates. False positive rate of this image prediction will be estimated and compared with conventional manual image film reading method.
We propose to study 200 lung cancer cases and 200 control participants. Among them, 100 lung cancer cases and 100 control participants will be used as training set, and another 100 lung cancer cases and 100 control participants will be used as test set.
Aims
CT texture analysis (CTTA) is a method of quantifying lesion heterogeneity based on distribution of pixel intensities within a region of interest. The aims of this study is to investigate the ability of CTTA to distinguish different lung lesions, and develop a predictive model utilizing CTTA parameters to estimate the false positive rate of Chest x-ray and Low-dose CT for the NLST.
Collaborators
Dr. Peng Huang, Associate Professor of Oncology Biostatistics, Johns Hopkins University
Yifei Sun, Doctoral student, Johns Hopkins University