Related Experiment Video
Updated: Aug 21, 2026
Identifying Coronary Artery Calcification on Non-gated Computed Tomography Scans
Published on: August 28, 2018
Opportunistic Screening of Carotid Atherosclerosis and Cardiovascular Mortality Risk Using Chest Radiographs:
Azka Rehman1, Jaewon Kim1, Hyeokjong Lee1
1Department of Biomedical Sciences Seoul National University Graduate School Seoul South Korea.
Insights
Foundation models can detect carotid atherosclerosis from chest X-rays, providing a new biomarker for cardiovascular disease (CVD) death risk. This AI-driven approach aids in risk stratification and early detection of CVD.
Area of Science:
- Artificial Intelligence in Medical Imaging
- Cardiovascular Disease Research
- Radiology and Diagnostic Imaging
Background:
- Cardiovascular disease (CVD) is a leading cause of mortality globally.
- Population-level screening for atherosclerosis often requires specialized, unavailable imaging techniques.
- Chest radiographs offer a potential, widely accessible imaging source for opportunistic atherosclerosis detection.
Purpose of the Study:
- To evaluate foundation models for detecting carotid atherosclerosis from chest radiographs.
- To assess the prognostic value of an AI-derived atherosclerosis biomarker for CVD death.
- To examine the explainability and clinical relevance of the AI model.
Main Methods:
- Compared four foundation models (Rad-DINO, DINOv2, OpenCLIP, CheXagent) for atherosclerosis detection on chest radiographs.
- Utilized linear probing, selective fine-tuning, and low-rank adaptation for model comparison.
- Assessed model performance using area under the ROC curve and explainability via gradient-weighted class activation mapping and a clinical relevance index.
- Validated the developed Deep-Learning Chest X-Ray Atherosclerosis Score in a large cohort for association with CVD death.
Main Results:
- Rad-DINO demonstrated the highest performance (AUC 0.74) with selective fine-tuning and low-rank adaptation.
- The model achieved high anatomic relevance (clinical relevance index 0.31) under low-rank adaptation.
- Higher scores from the Deep-Learning Chest X-Ray Atherosclerosis Score were strongly associated with increased risk of CVD death (adjusted HR 9.36).
Conclusions:
- Foundation models can opportunistically detect carotid atherosclerosis from routine chest radiographs.
- The AI-derived biomarker provides significant prognostic information for CVD mortality.
- Explainability analyses confirm clinical relevance, supporting scalable risk stratification for cardiovascular disease.
Background:
Cardiovascular disease (CVD) remains a leading cause of death, but population-level screening for atherosclerosis often depends on specialized imaging that is not routinely available. We evaluated foundation models for opportunistic detection of carotid atherosclerosis from chest radiographs, assessed the prognostic value of the resulting biomarker for CVD death, and examined model explainability.
Methods:
Among 5785 participants with carotid sonography annotations, we compared Rad-DINO, DINOv2, OpenCLIP, and CheXagent for chest radiograph-based atherosclerosis detection using linear probing, selective fine-tuning, and low-rank adaptation. Performance was assessed by area under the receiver operating characteristic curve with 95% CIs. Explainability was evaluated using gradient-weighted class activation mapping and a clinical relevance index, which quantified attention within cardiovascular anatomic masks. The selected model generated a Deep-Learning Chest X-Ray Atherosclerosis Score, which was tested for association with CVD death in 32 524 participants using Framingham Risk Score-adjusted Cox models.
Results:
Rad-DINO achieved the highest performance with selective fine-tuning (area under the receiver operating characteristic curve, 0.74 [95% CI, 0.71-0.77]) and low-rank adaptation (area under the receiver operating characteristic curve, 0.74 [95% CI, 0.70-0.76]), with the highest anatomic relevance under low-rank adaptation (clinical relevance index, 0.31). Higher Deep-Learning Chest X-Ray Atherosclerosis Score tertiles were strongly associated with CVD death (adjusted hazard ratio, 9.36 [95% CI, 3.35-26.15] for high versus low; P-trend<0.005). The model concordance index was 0.73 (95% CI, 0.67-0.75).
Conclusions:
Foundation models enable opportunistic detection of carotid atherosclerosis from chest radiographs and yield a biomarker that provides prognostic information for CVD death. Explainability analyses demonstrated alignment with CVD anatomic regions, supporting clinical relevance and scalable risk stratification.
Related Concept Videos
01:25Imaging Studies for Cardiovascular System VI: Calcium -Scoring CT
01:20Imaging Studies for Cardiovascular System III: X-Ray
Definition and Purpose
An X-ray, or radiograph, is a non-invasive method that uses ionizing radiation to take images of internal structures. It is mainly used in cardiac imaging to examine the heart, lungs, and major blood vessels, aiming to identify abnormalities in the heart's size, shape, and position, such as heart failure, congenital defects, and vascular...
01:28Imaging Studies for Cardiovascular System V: CT
01:30Radiological Investigation I: X-ray and CT
01:27Atherosclerosis II: Clinical Manifestations and Diagnostic Tests
01:21Aneurysm II: Clinical Manifestations and Diagnostic Studies