Optimizing myocardial infarction detection: a hybrid CNN-GRU deep learning approach
Bibliographic record
Abstract
BACKGROUND: Myocardial infarction (MI) is a life-threatening condition caused by sudden interruption of blood supply to the heart. Electrocardiogram (ECG) is the primary tool for MI diagnosis, but interpretation challenges exist. This study aimed to optimize MI detection by developing a hybrid CNN-GRU Deep Learning model (DLM) based on ECG as a diagnostic support tool. METHODS: This retrospective diagnostic study included a total of 56,354 ECGs, comprising 41,871 from patients diagnosed with (MI) and 14,474 from healthy patients. Each ECG record consisted of a 20-second 15-lead recording per individual, sampled at 1000 Hz. The CNN-GRU model was trained on 85% of these ECGs and validated on the remaining 15%. The CNN-GRU model was executed on the pre-processed data using the Pan-Tompkins algorithm obtained from the PhysioNet website (PTB Diagnostic ECG Database), and all recordings were labelled by expert cardiologists. We examined a new model for classifying ECG heartbeats and found that it can compete with advanced models. The performance of the DLM was evaluated using the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, Macro Average, and Weighted Average. RESULTS: The area under the receiver operating characteristic curve (AUC) of the CNN-GRU model for MI detection was close to one. The CNN-GRU model achieved excellent performance with 15 leads (ACC = 99.43%, sensitivity = 99.71%, specificity = 98.59%). Using lead II alone, performance improved slightly (ACC = 99.73%, sensitivity = 99.75%, specificity = 99.66%). The high AUC and other metrics indicate strong diagnostic ability. Based on the reported results, the CNN-GRU model using lead II was the best model. CONCLUSIONS: The findings suggest that the proposed model can support clinical decision-making and guide future research in cardiovascular medicine.
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 | 0.001 | 0.001 |
| 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 | 0.001 |
| 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".