Automated Detection of Lung Cancer From LDCT
Principal Investigator
Name
Warren Clarida
Degrees
Ph.D.
Institution
IDx LLC
Position Title
R&D Director
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-385
Initial CDAS Request Approval
Jan 8, 2018
Title
Automated Detection of Lung Cancer From LDCT
Summary
Lung cancer is by far the leading cause of cancer death among men and women, outpacing colon, breast, and prostate cancer deaths combined. Early detection is vital to initiate treatment and prevent metastasis, though most symptoms of lung cancer do not appear until the disease is at an advanced stage. This motivates the need for routine screening of high-risk patients; the American Cancer Society recommends patients in the high-risk pool to have a LDCT every year until age 74. In the NLST trial, 1 in 4 CT scans found abnormalities that turned out not to be cancer.
The adoption of machine learning based algorithms for automated image analysis and diagnosis can simultaneously improve patient care and lower cost. The goal of this study is to determine the ability of machine learning algorithms to detect cancerous lung nodules in LDCT and X-Ray. The final evaluation will consider the ability of the algorithm to detect the lung nodules and correctly classify the lung nodules as cancerous.
The adoption of machine learning based algorithms for automated image analysis and diagnosis can simultaneously improve patient care and lower cost. The goal of this study is to determine the ability of machine learning algorithms to detect cancerous lung nodules in LDCT and X-Ray. The final evaluation will consider the ability of the algorithm to detect the lung nodules and correctly classify the lung nodules as cancerous.
Aims
Aim 1: Develop and validate a deep learning based system for the detection and classification of cancerous nodules in the lung detected with CT.
Aim 2: Develop and validate a deep learning based system for the detection and classification of cancerous nodules in the lung detected with chest X-Ray.
Collaborators
Ryan Amelon, PhD
IDx LLC
Abhay Shah, PhD
IDx LLC