Research on CT Influence Omics Method of Lung Cancer Based on Deep Transfer Learning
Principal Investigator
Name
Xianwei Zhao
Degrees
pH.d.
Institution
NPIC
Position Title
Radiation Therapy Physicist
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-1091
Initial CDAS Request Approval
Jul 3, 2023
Title
Research on CT Influence Omics Method of Lung Cancer Based on Deep Transfer Learning
Summary
The radiomics research on lung cancer is becoming more and more extensive, which will greatly help reduce the intelligence and automation level of lung cancer diagnosis, and effectively enhance the accuracy of lung cancer diagnosis. This project intends to combine deep learning and transfer learning technology to realize automatic identification based on medical images of lung cancer, to achieve end-to-end feature extraction by means of feature extraction and deep network construction, and to improve the generalization performance of the model pioneeringly.
The NLST public dataset is very useful and comprehensive, which contains information on many research subjects, such as outcomes, diagnostic procedures, lung cancer, etc. Therefore, we would like to apply for full access to the dataset in good faith and hope to be approved.
The NLST public dataset is very useful and comprehensive, which contains information on many research subjects, such as outcomes, diagnostic procedures, lung cancer, etc. Therefore, we would like to apply for full access to the dataset in good faith and hope to be approved.
Aims
1. Lung cancer medical image division based on 3D Slicer
3. Lung cancer feature extraction based on deep learning
2. Feature mining of lung cancer based on manual feature extraction
4. Lung cancer pathological type recognition based on deep transfer learning
5, etc.
Collaborators
No