Retinal image-based deep learning for mild cognitive impairment detection in coronary artery disease population
Authors' conclusion
The AI algorithm trained on fundus images in this study exerted promising performance in screening MCI in the CAD population and might be a non-invasive and effective alternative for early diagnosis of the disease.
Study context
- Population
- Patients with coronary artery disease (CAD) (at least one ≥50% stenosis), Beijing Anzhen Hospital, July 2021 to July 2023
- Participants
- 4,357 patients9,009 images
- Comparison
- Fundus-image deep learning model vs Mini-Mental State Examination (MMSE) or Montreal Cognitive Assessment (MoCA) reference standard
- Outcome
- Detection of mild cognitive impairment (AUC)
Methodological detailssingle-centre, cross-sectional
- Single-centre
- Cross-sectional
Performance
As reported in the abstract
Mild cognitive impairment (MCI) detection
| Strategy | AUC, test group | AUC, validation group |
|---|---|---|
| MMSE-based artificial intelligence (AI) | 0.832 (95% CI 0.800 to 0.863) | 0.776 (95% CI 0.730 to 0.821) |
| MoCA-based AI | 0.764 (95% CI 0.742 to 0.785) | 0.725 (95% CI 0.701 to 0.750) |
AUC, test group
MMSE-based artificial intelligence (AI)0.832 (95% CI 0.800 to 0.863)
MoCA-based AI0.764 (95% CI 0.742 to 0.785)
AUC, validation group
MMSE-based artificial intelligence (AI)0.776 (95% CI 0.730 to 0.821)
MoCA-based AI0.725 (95% CI 0.701 to 0.750)
Further findings
- Good calibration in internal test sets of both models
- Decision curves revealed extensive clinical application value
Original publication
Retinal image-based deep learning for mild cognitive impairment detection in coronary artery disease population. Heart. Published 16 May 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →