ECGNextFormer: A Lightweight, Scalable, and Explainable Architecture for Estimating Cardiovascular Well-Being from Electrocardiograms

Keyphrases

• Well-being 100%
• Scalable Architecture 100%
• Electrocardiogram 100%
• Lightweight Structures 100%
• Explainable Architectures 100%
• Age Estimation 57%
• Neural Network 28%
• Multi-scale Pooling 28%
• Healthcare 14%
• Disease Diagnosis 14%
• Cardiac Disease 14%
• Non-invasive Test 14%
• Feature Extraction 14%
• Superior Performance 14%
• Chronological Age 14%
• Receptive Field 14%
• Feature Map 14%
• Human-Interpretable 14%
• Transformer Architecture 14%
• Hybrid Transformer 14%
• Self-attention 14%
• Self-Attention Layer 14%
• Heatmap 14%
• Global Receptive Field 14%
• Low-parameter 14%
• CNN Architecture 14%
• Spatial Details 14%
• Scoring Scheme 14%
• Guided Backpropagation 14%
• Current Network 14%
• Novel Leads 14%
• Hybrid CNN 14%
• Derived Electrocardiograms 14%
• Lead Scoring 14%

Computer Science

• Neural Network 100%
• Receptive Field 100%
• Self-Attention 100%
• Feature Extraction 50%
• Superior Performance 50%
• Convolutional Neural Network 50%
• Feature Map 50%
• Attention (Machine Learning) 50%
• Chronological Age 50%
• Spatial Detail 50%
• Vertebral Column 50%