Cross-View Attention Fusion Net: A Prior-Guided Dual-View Representation Learning for Cardiac Output Estimation from Short-Term PPG Signals

Name: Cross-View Attention Fusion Net: A Prior-Guided Dual-View Representation Learning for Cardiac Output Estimation from Short-Term PPG Signals
Uploaded: 2026-10-15T20:22:20.074543+00:00
Duration: 344 s
Description: CVAF-Net is a deep learning model that estimates cardiac output from short PPG signals by using two views: raw time-series data and a structured feature map. It learns to combine these views through cross-view attention, achieving high accuracy in both simulated and real-world tests while being more efficient than Transformer models.

CVAF-Net is a deep learning model that estimates cardiac output from short PPG signals by using two views: raw time-series data and a structured feature map. It learns to combine these views through cross-view attention, achieving high accuracy in both simulated and real-world tests while being more efficient than Transformer models.

Open the full episode