Improving risk stratification of uncertain pulmonary nodules based on deep learning method
Principal Investigator
Name
yulu Wu
Degrees
M.Eng
Institution
zhongyuan university of technology
Position Title
student
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-1219
Initial CDAS Request Approval
Mar 25, 2024
Title
Improving risk stratification of uncertain pulmonary nodules based on deep learning method
Summary
With the progress of screening technology, especially the wide application of computed tomography (CT), the detection rate of uncertain pulmonary nodules (IPNs) has increased dramatically. This phenomenon has brought great challenges to the risk stratification in clinical practice, especially to the judgment of medium risk category. At the same time, due to the unexpected discovery of IPNs and the cost and risk considerations of medical management, its treatment has become particularly critical. In order to manage these situations more effectively, we put forward an innovative method, that is, to analyze CT images through deep learning technology to improve the diagnostic efficiency of IPN. This method is expected to provide highly accurate diagnosis results independently, and it is expected to realize the initial accurate diagnosis of lung cancer without relying on manual interpretation by radiologists.
Aims
1. Develop and verify the deep learning model, integrate CT image data and analyze it with clinical data to improve the diagnostic accuracy of uncertain pulmonary nodules (IPN).
2. Explore the method of diagnosing lung cancer by using the whole CT scan image instead of only using the nodule area, so as to make full use of the background information in the image.
3. Compared with the traditional risk prediction model, the advantages of the deep learning model in diagnosing uncertain pulmonary nodules are verified.
4. The deep learning model is verified by external verification data set to ensure its good generalization ability and reliability.
Collaborators
Doctor, Zhengzhou University, Jess Zhang