|
Abstract:
|
Automated interpretation of 12-lead electrocardiogram
(ECG) images offers critical value as a clinical decision
support tool, enabling rapid and accurate patient triage in fastpaced
emergency environments. However, existing classification
models often rely solely on visual data, ignoring the essential
patient history utilized by human cardiologists. This paper
proposes a novel multimodal deep learning architecture that
integrates raw static ECG images with baseline cardiovascular
risk factors (e.g., age, sex, blood pressure) to optimize diagnostic
precision. the model employs a Contrastive Language-Image Pretraining
(CLIP) backbone, utilizing cardiologist reports as an
auxiliary supervisory signal during training to extract complex
morphological features without requiring manual annotations.
During inference, patient clinical metadata generates an attention
mask that dynamically scales the extracted visual embeddings.
This early-fusion gating mechanism balances information across
modalities, enabling the model to adjust its visual processing
based on each patient’s risk profile. Evaluated on ten highly
imbalanced cardiovascular abnormalities from the MIMIC-IV
dataset using Focal Loss, the proposed model achieves a weighted
average F1-score of 75.8%, overall AUC of 95.23%, Accuracy of
93.50%, Precision of 75.72% and Recall of 76.15%, establishing
a highly competitive benchmark against current state-of-the-art
multi-label classifiers. Additionally, the integration of clinical context
significantly improved the predictive confidence for critical
ischemic events, boosting the detection rate for Acute Myocardial
Infarction (AMI) by over 66% compared to an image-only
baseline. These results demonstrate that patient clinical context is
an indispensable prior, effectively transitioning theoretical ECG
classifiers into robust, safety-first automated triage tools. |
