Promise and challenges of heart chamber segmentation from non-contrast CT scans using contrastive unpaired image translation: a feasibility study
arXiv:2606.23879v1 Announce Type: cross Abstract: Purpose: To evaluate the feasibility and challenges of heart chamber segmentation from non-contrast CT scans using contrastive unpaired image translation and deep learning-based segmentation. Approach: We developed ChameleonNet, a framework utilizing the Contrastive Unpaired Translation (CUT) network with decoupled contrastive learning (DCL) loss to synthesize non-contrast CT from contrast CT scans. Using annotations of four heart chambers (left atrium (LA), left ventricle (LV), right atrium (RA), and right ventricle (RV)) from contrast scans,
The continuous advancements in AI, particularly deep learning and generative models, are enabling increasingly sophisticated medical image analysis techniques that overcome previous data limitations.
Improving diagnostic capabilities from non-contrast CT scans reduces patient risk and cost, potentially democratizing advanced cardiovascular diagnostics by leveraging readily available scan types.
The ability to accurately segment heart chambers from non-contrast CTs using AI tools could reduce the reliance on contrast agents, expanding the applicability of CT for heart disease screening and monitoring.
- · Medical AI companies
- · Hospitals and diagnostic centers
- · Patients with cardiovascular conditions
- · Radiology departments
- · Manufacturers of contrast agents (minor impact)
- · Traditional manual image analysis services
Non-invasive cardiovascular diagnostics become more accessible and safer due to reduced need for contrast agents.
Increased early detection of heart conditions could lead to better patient outcomes and reduced healthcare costs over time.
The methodology could be extended to other organs and pathologies, accelerating AI's integration into routine diagnostic pipelines and potentially influencing medical device regulation.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.AI