Machine Learning for Lung Nodule Detection and Diagnosis Classification
Principal Investigator
Name
Mehmet Kaya
Degrees
Ph.D., Biomedical Engineering, RutgersUniversity
Institution
Florida Institute of Technology
Position Title
Assistant Professor, Biomedical Engineering, Biomedical and Chemical Engineering and Sciences
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-777
Initial CDAS Request Approval
Mar 30, 2021
Title
Machine Learning for Lung Nodule Detection and Diagnosis Classification
Summary
Lung cancer is the leading cause of cancer deaths worldwide. The lung nodule, a small abnormal area found during a chest CT scan, could be the effective marker for diagnosing lung cancer. Lung cancer is more likely to be successfully treated when diagnosed at an earlier stage before spreading. The promising development of machine learning applications may yield potential benefits for medical diagnosis purposes. In this research, machine learning techniques are used for automatic lung nodule detection and diagnosis classification. Furthermore, a deep learning-based survival prediction model is generated for assessing the mortality of lung cancer patients.
Aims
1. Creating a machine learning pipeline for lung region segmentation and bronchioli segmentation directed from CT scans.
2. Convolutional neural networks for lung nodule detection and diagnosis classification.
3. Feature extraction from the lung nodule region for patient survival analysis.
Collaborators
Guochang Ye, Florida Institute of Technology