Development of lung nodule categorization, characterization and quantification on LDCT for lung cancer screening
Principal Investigator
Name
Mannudeep Kalra
Degrees
MD
Institution
Massachusetts General Hospital
Position Title
Associate Professor
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-203
Initial CDAS Request Approval
Apr 4, 2016
Title
Development of lung nodule categorization, characterization and quantification on LDCT for lung cancer screening
Summary
A deep learning based algorithm will be developed to categorize, characterize and quantify the morphology and size of lung nodules seen in LDCT images. These deep learning methods will then be verified for accuracy, and reproducibility in comparison to human observers. The automatically extracted features of lung nodules will also be correlated with the lungRad categories of the lung nodules and the subject outcome data. The algorithm is expected to aid radiologists to avoid errors and inconsistencies in lung nodule categorization, characterization and quantification. The large NLST database will provide an optimal resource for rapid development of such artificial intelligence based capabilities. The project will be performed with partnership of clinical (Massachusetts General Hospital) and credible engineering partners from Rensselaer Polytechnic Institute (RPI) and Fudan University's Artificial Intelligence group).
Aims
1. To develop a deep learning based algorithm for detection and categorization of lung nodule subtype seen on LDCT for lung cancer screening.
2. To develop capabilites for automatic quantification of lung nodule morphology and dimensionality
3. To verify the accuracy and reproducibility of the developed new algorithm on basis of large NLST database.
Collaborators
Subba R. Digumarthy, MD; Milena Petranovic, MD; Shaunagh McDermott, MD; Rodrigo Canellas de Souza, MD; (All from Massachusetts General Hospital)
Ge Wang, PhD (Rensselaer Polytechnic Institute, Troy, New York)
Junping Zhang, PhD (Fudan University, Shanghai, China)