Temporal Feature Weaving for Neonatal Echocardiographic Viewpoint Video Classification

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: French, Satchel, Zhu, Faith, Jain, Amish, Khan, Naimul
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916554970497024
author French, Satchel
Zhu, Faith
Jain, Amish
Khan, Naimul
author_facet French, Satchel
Zhu, Faith
Jain, Amish
Khan, Naimul
contents 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
format Preprint
id arxiv_https___arxiv_org_abs_2501_03967
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Temporal Feature Weaving for Neonatal Echocardiographic Viewpoint Video Classification
French, Satchel
Zhu, Faith
Jain, Amish
Khan, Naimul
Computer Vision and Pattern Recognition
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
title Temporal Feature Weaving for Neonatal Echocardiographic Viewpoint Video Classification
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2501.03967