Retinal image-based deep learning for mild cognitive impairment detection in coronary artery disease population
Bibliographic record
Abstract
BACKGROUND: Coronary artery disease (CAD) is linked to an increased risk of mild cognitive impairment (MCI). Effective and convenient screening methods for identifying MCI from the CAD population are still lacking. This study aims to develop a deep learning model using fundus images to optimise MCI diagnosis in the CAD population, achieving early intervention and improving prognosis. METHODS: Patients with CAD (at least one ≥50% stenosis) from July 2021 to July 2023 at Beijing Anzhen Hospital were included in the single-centre cross-sectional study. Eligible patients from July 2021 to May 2023 were randomly assigned in an 8:2 ratio for training and internal testing of the model. Patients enrolled from June 2023 to July 2023 were included in the external validation group. Four different convolutional neural network architectures were used to train the subjects' fundus images. The reference standards were a Mini-Mental State Examination (MMSE) score of <27 and a Montreal Cognitive Assessment (MoCA) score of <26, respectively. A comprehensive visual model of MCI detection was established through model integration. RESULTS: A total of 9009 eligible images from 4357 patients with CAD were collected. The artificial intelligence (AI) algorithm based on the MMSE achieved an area under the curve (AUC) of 0.832 (95% CI 0.800 to 0.863) in the test group and 0.776 (95% CI 0.730 to 0.821) in the validation group. The AI algorithm based on the MoCA achieved an AUC of 0.764 (95% CI 0.742 to 0.785) in the test group and 0.725 (95% CI 0.701 to 0.750) in the validation group. The calibration curves of the internal test sets of the two models exhibited a good calibration effect. The results of decision curves revealed extensive clinical application value. 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. TRIAL REGISTRATION NUMBER: NCT06102226.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".