Temporal Feature Weaving for Neonatal Echocardiographic Viewpoint Video Classification
This work addresses faster diagnosis and screening in under-resourced clinics by automating echocardiographic viewpoint classification, though it is incremental as it builds on existing video classification methods.
The paper tackled automated viewpoint classification in neonatal echocardiograms by treating it as a video classification problem, achieving a 4.33% accuracy increase over baseline image classification using a CNN-GRU architecture with temporal feature weaving on only four consecutive frames.
Automated viewpoint classification in echocardiograms can help under-resourced clinics and hospitals in providing faster diagnosis and screening when expert technicians may not be available. We propose a novel approach towards echocardiographic viewpoint classification. We show that treating viewpoint classification as video classification rather than image classification yields advantage. We propose a CNN-GRU architecture with a novel temporal feature weaving method, which leverages both spatial and temporal information to yield a 4.33\% increase in accuracy over baseline image classification while using only four consecutive frames. The proposed approach incurs minimal computational overhead. Additionally, we publish the Neonatal Echocardiogram Dataset (NED), a professionally-annotated dataset providing sixteen viewpoints and associated echocardipgraphy videos to encourage future work and development in this field. Code available at: https://github.com/satchelfrench/NED