Improving ultrasound video classification: an evaluation of novel deep learning methods in echocardiography
Journal of Medical Artificial Intelligence · 2020-03 · 3 · 4-4
DOI: 10.21037/jmai.2019.10.03 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Medical Artificial Intelligence, 2020-03. Verfasst von James P Howard und 15 weiteren Autor:innen. Beteiligte Einrichtung: National Heart and Lung Institute, Imperial College London, Hammersmith Hospital, London, UK. Die Arbeit wurde bisher 58-mal zitiert (Crossref). Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: Wellcome Trust, British Heart Foundation, Department of Health.
Abstract (Originalsprache)
Echocardiography is the commonest medical ultrasound examination, but automated interpretation is challenging and hinges on correct recognition of the 'view' (imaging plane and orientation). Current state-of-the-art methods for identifying the view computationally involve 2-dimensional convolutional neural networks (CNNs), but these merely classify individual frames of a video in isolation, and ignore information describing the movement of structures throughout the cardiac cycle. Here we explore the efficacy of novel CNN architectures, including time-distributed networks and two-stream networks, which are inspired by advances in human action recognition. We demonstrate that these new architectures more than halve the error rate of traditional CNNs from 8.1% to 3.9%. These advances in accuracy may be due to these networks' ability to track the movement of specific structures such as heart valves throughout the cardiac cycle. Finally, we show the accuracies of these new state-of-the-art networks are approaching expert agreement (3.6% discordance), with a similar pattern of discordance between views.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- National Heart and Lung Institute, Imperial College London, Hammersmith Hospital, London, UK.
- Department of Computing, Imperial College London, Hammersmith Hospital, London, UK.
- Imperial College Healthcare NHS Trust, London, UK.
Förderung
- Wellcome Trust
- British Heart Foundation
- Department of Health
Zitieren
APA
Howard, J. P., Tan, J., Shun-Shin, M. J., Mahdi, D., Nowbar, A. N., Arnold, A. D., Ahmad, Y., McCartney, P., Zolgharni, M., Linton, N. W. F., Sutaria, N., Rana, B., Mayet, J., Rueckert, D., Cole, G. D., & Francis, D. P. (2020). Improving ultrasound video classification: an evaluation of novel deep learning methods in echocardiography. Journal of Medical Artificial Intelligence, 3, 4-4. https://doi.org/10.21037/jmai.2019.10.03
Vancouver
Howard JP, Tan J, Shun-Shin MJ, Mahdi D, Nowbar AN, Arnold AD, et al. Improving ultrasound video classification: an evaluation of novel deep learning methods in echocardiography. Journal of Medical Artificial Intelligence. 2020;3:4-4. doi:10.21037/jmai.2019.10.03
BibTeX
@article{Howard2020,
author = {Howard, James P and Tan, Jeremy and Shun-Shin, Matthew J and Mahdi, Dina and Nowbar, Alexandra N and Arnold, Ahran D and Ahmad, Yousif and McCartney, Peter and Zolgharni, Massoud and Linton, Nick W F and Sutaria, Nilesh and Rana, Bushra and Mayet, Jamil and Rueckert, Daniel and Cole, Graham D and Francis, Darrel P},
title = {Improving ultrasound video classification: an evaluation of novel deep learning methods in echocardiography},
journal = {Journal of Medical Artificial Intelligence},
year = {2020},
volume = {3},
pages = {4--4},
doi = {10.21037/jmai.2019.10.03}
}
EUVIMED