Large Scale X-Ray and CT Representation Learning from Multi-Source

Principal Investigator

Name
Chuang Niu

Degrees
Ph.D.

Institution
Rensselaer Polytechnic Institute

Position Title
Postdoc

Email
niuc@rpi.edu

About this CDAS Project

Study
PLCO (Learn more about this study)

Project ID
PLCOI-955

Initial CDAS Request Approval
Mar 30, 2022

Title
Large Scale X-Ray and CT Representation Learning from Multi-Source

Summary
X-ray and CT are the most frequently used imaging tools for medical diagnosis. Artificial intelligence techniques, especially deep learning techniques, have shown great potential for improving medical diagnosis. Extracting clinically meaningful features is the core component of AI algorithms. Here we propose to leverage large-scale X-ray and CT datasets and associated multiple other source data to train a large-scale representation learning model for improving the diagnosis performance.

Aims

1. Build a large-scale CT representation learning model.
2. Build a large-scale Chest X-ray representation learning model.
3. Improve the diagnosis performance for lung nodule detection and other lung diseases.

Collaborators

Ge Wang, AI-based X-ray Imaging System (AXIS) Lab Department of Biomedical Engineering, Rensselaer Polytechnic Institute, 110 8th Street, Troy, New York 12180, USA
Giridhar Dasegowda, Department of Radiology Massachusetts General Hospital, Harvard Medical School White 270-E, 55 Fruit St, Boston, MA 02114, USA
Mannudeep K. Kalra, Department of Radiology Massachusetts General Hospital, Harvard Medical School White 270-E, 55 Fruit St, Boston, MA 02114, USA