Search papers, labs, and topics across Lattice.
This paper introduces EVL-ECG, a knowledge distillation framework tailored for ECG interpretation that addresses the challenges of transferring knowledge from large models to smaller, edge-deployable architectures. It uses multi-head cross-attention alignment, optimal transport-based visual feature matching, and geometric intra-architecture relation matching to preserve ECG signal characteristics during distillation. Experiments show EVL-ECG improves AUC by up to 2.4% and clinical accuracy by 1.1% compared to existing methods, creating a 2B-parameter model suitable for resource-constrained environments.
You can distill a clinically accurate 2B-parameter ECG interpretation model that runs on edge devices by carefully aligning morphological features, structural relationships, and diagnostic reasoning from larger models.
High-fidelity ECG interpretation is increasingly reliant on massive foundation models, yet their deployment in clinical edge-care remains hindered by extreme computational demands. While knowledge distillation (KD) is a promising solution, traditional methods fail to capture the complex spatio-temporal dependencies of ECG signals when transferring knowledge across heterogeneous architectures. In this paper, we propose EVL-ECG, a framework specifically designed for cross-architecture distillation of cardiac diagnostic logic. EVL-ECG introduces three ECG-aware innovations: (1) Multi-Head Cross-Attention Alignment, which harmonizes architectural discrepancies to preserve fine-grained morphological features; (2) Optimal Transport-based Visual Feature Matching, utilizing optimal transport to maintain global structural relationships across ECG leads despite mismatched token representations; and (3) Geometric Intra-Architecture Relation Matching, which distills the latent diagnostic reasoning of the teacher model. Evaluations across ECG benchmarks demonstrate that EVL-ECG yields improvements of up to 2.4% AUC and 1.1% clinical accuracy over existing baselines. Notably, EVL-ECG establishes an efficient 2B-parameter ECG foundation model, suitable for resource-constrained clinical environments.
