Skip to Main Content

An official website of the United States government

Government Funding Lapse

Because of a lapse in government funding, the information on this website may not be up to date, transactions submitted via the website may not be processed, and the agency may not be able to respond to inquiries until appropriations are enacted. The NIH Clinical Center (the research hospital of NIH) is open. For more details about its operating status, please visit  cc.nih.gov. Updates regarding government operating status and resumption of normal operations can be found at OPM.gov.

Principal Investigator
Name
Xiaofeng Yang
Degrees
Ph.D
Institution
Emory University
Position Title
Associate Professor
Email
About this CDAS Project
Study
NLST (Learn more about this study)
Project ID
NLST-1249
Initial CDAS Request Approval
May 16, 2024
Title
Using foundation model for cardiovascular disease detection.
Summary
1. Introduction
Cardiovascular diseases (CVDs) are a leading cause of mortality worldwide. Accurate and early diagnosis is crucial for reducing the burden of CVDs and improving patient outcomes. However, standard computed tomography (CT) scans expose patients to high radiation levels. Low-dose CT (LDCT) imaging is increasingly used to minimize radiation exposure, yet it often leads to reduced image quality, affecting diagnostic accuracy. Recent advancements in foundation models for medical image analysis offer a promising approach to overcome these challenges.

2. Objective
The primary objective of this proposal is to develop a foundation model, (e.g., Chat-GPT) for accurately classifying cardiovascular diseases on low-dose CT images. This approach will leverage the transfer learning and representation learning capabilities of foundation models, resulting in accurate and robust classification despite the limitations posed by low-dose CT.
Aims

1. Investigate foundation model in cardiovascular disease classification
2. Find a reliable method to have great performance cardiovascular disease classification -far better than current methods.

Collaborators

Xu Xin, Xiaofeng Yang,