Detection/Segmentation of lung cancer and more with deep learning
Principal Investigator
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-376
Initial CDAS Request Approval
Nov 17, 2017
Title
Detection/Segmentation of lung cancer and more with deep learning
Summary
In this project, we're trying to detect lung cancer in lung CT, and perform segmentation of the nodule region. NLST datasets will be used as training/testset for our detection model. The second step will be to detect other diseases such as emphysema, PAH, etc. by annotating those anatomical structures, and training them for the model.
Aims
1. Determine if deep learning algorithm detects nodules accurately in low dose lung CT
2. Determine if our segmentation model performs segmentation well enough on nodules in low dose lung CT
3. Determine if deep learning algorithm classifies disease such as emphysema, and pulmonary arterial hypertension accurately.
Collaborators
Jay Lee, Medipixel
Hyungkyu Kim, Medipixel