Automated Identification of Patent Ductus Arteriosus Using a Computer Vision Model
Journal of Ultrasound in Medicine · 2023-12 · 42 (12) · 2707-2713
DOI: 10.1002/jum.16305 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Ultrasound in Medicine, online erschienen am 14.07.2023. Verfasst von Jason Erno und 5 weiteren Autor:innen. Beteiligte Einrichtung: College of Medicine, Medical University of South Carolina, Charleston, South Carolina, USA. Hauptthemen laut MEDLINE: Infant, Premature, Ductus Arteriosus, Patent. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 22 Literaturangaben. Gefördert von: South Carolina Clinical and Translational Research Institute, Medical University of South Carolina.
Abstract (Originalsprache)
Objectives
Patent ductus arteriosus (PDA) is a vascular defect common in preterm infants and often requires treatment to avoid associated long‐term morbidities. Echocardiography is the primary tool used to diagnose and monitor PDA. We trained a deep learning model to identify PDA presence in relevant echocardiographic images.
Methods
Echocardiography video clips (n = 2527) in preterm infants were reviewed by a pediatric cardiologist and those relevant to PDA diagnosis were selected and labeled (PDA present/absent/indeterminate). We trained a convolutional neural network to classify each echocardiography frame of a clip as belonging to clips with or without PDA. A novel attention mechanism that aggregated predictions for all frames in each clip to obtain a clip‐level prediction by weighting relevant frames.
Results
In early model iterations, we discovered training with color Doppler echocardiography clips produced the best performing classifier. For model training and validation, 1145 such clips from 66 patients (661 PDA+ clips, 484 PDA− clips) were used. Our best classifier for clip level performance obtained sensitivity of 0.80 (0.83–0.90), specificity of 0.77 (0.62–0.92) and AUC of 0.86 (0.83–0.90). Study level performance obtained sensitivity of 0.83 (0.72–0.94), specificity of 0.89 (0.79–1.0) and AUC of 0.93 (0.89–0.98).
Conclusions
Our novel deep learning model demonstrated strong performance in classifying echocardiography clips with and without PDA. Further model development and external validation are warranted. Ultimately, integration of such a classifier into auto detection software could streamline PDA imaging workflow. This work is the first step toward semi‐automated, bedside detection of PDA in preterm infants.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- College of Medicine, Medical University of South Carolina, Charleston, South Carolina, USA.
- Department of Electrical and Computer Engineering, Clemson University, Clemson, South Carolina, USA.
- Department of Pediatrics, Medical University of South Carolina, Charleston, South Carolina, USA.
ORCID-Profile: Jason Erno
Förderung
- South Carolina Clinical and Translational Research Institute, Medical University of South Carolina
Zitieren
APA
Erno, J., Gomes, T., Baltimore, C., Lineberger, J. P., Smith, D. H., & Baker, G. H. (2023). Automated Identification of Patent Ductus Arteriosus Using a Computer Vision Model. Journal of Ultrasound in Medicine, 42(12), 2707-2713. https://doi.org/10.1002/jum.16305
Vancouver
Erno J, Gomes T, Baltimore C, Lineberger JP, Smith DH, Baker GH. Automated Identification of Patent Ductus Arteriosus Using a Computer Vision Model. Journal of Ultrasound in Medicine. 2023;42(12):2707-2713. doi:10.1002/jum.16305
BibTeX
@article{Erno2023,
author = {Erno, Jason and Gomes, Thomas and Baltimore, Christopher and Lineberger, John P and Smith, D Hudson and Baker, G Hamilton},
title = {Automated Identification of Patent Ductus Arteriosus Using a Computer Vision Model},
journal = {Journal of Ultrasound in Medicine},
year = {2023},
volume = {42},
number = {12},
pages = {2707--2713},
doi = {10.1002/jum.16305}
}
EUVIMED