Non-invasive multiomics anlaysis based lung cancer diagnosis
Principal Investigator
Name
Mavis Mu
Degrees
Ph.D
Institution
Key Laboratory of Molecular Imaging, Chinese Academy of Science
Position Title
Proffessor
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-1051
Initial CDAS Request Approval
Apr 24, 2023
Title
Non-invasive multiomics anlaysis based lung cancer diagnosis
Summary
In this project, a non-invasive lung nodule diagnosis model will be developed using multimodality data, including CT images, and DNA methylation data from peripheral blood mononuclear cells, which will be further validated in our prospective study. Given the high sensitivity of CT images and high specificity of DNA methylation data, we have the hypothesis that their combination should improve the accuracy of the diagnosis, and reduce the over or under diagnosis. To construct a robust model, international and multi-intuitional data are in urgent need. Therefore, the NLST data is our hope to develop the final model. The objectives of our work include:
Aims
1. Develop and validate a deep learning based CT diagnosis model;
2. Develop and validate a deep learning based methylation diagnosis model;
3. Develop a fusion model for further prospective validation.
Collaborators
Peking University People's Hospital