EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Development of Deep Learning Models for Real-Time Thoracic Ultrasound Image Interpretation

Austin J Ruiz, Sofia I Hernández Torres, Eric J Snider iD

Journal of Imaging · 05.07.2025 · 11 (7) · 222

DOI: 10.3390/jimaging11070222 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Imaging, online erschienen am 05.07.2025. Verfasst von Austin J Ruiz und 2 weiteren Autor:innen. Beteiligte Einrichtung: Organ Support and Automation Technologies Group, U.S. Army Institute of Surgical Research, JBSA Fort Sam Houston, San Antonio, TX 78234, USA. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: United States Department of Defense, Oak Ridge Institute for Science and Education, U.S. Army Medical Research and Development Command.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Thoracic injuries account for a high percentage of combat casualty mortalities, with 80% of preventable deaths resulting from abdominal or thoracic hemorrhage. An effective method for detecting and triaging thoracic injuries is point-of-care ultrasound (POCUS), as it is a cheap and portable noninvasive imaging method. POCUS image interpretation of pneumothorax (PTX) or hemothorax (HTX) injuries requires a skilled radiologist, which will likely not be available in austere situations where injury detection and triage are most critical. With the recent growth in artificial intelligence (AI) for healthcare, the hypothesis for this study is that deep learning (DL) models for classifying images as showing HTX or PTX injury, or being negative for injury can be developed for lowering the skill threshold for POCUS diagnostics on the future battlefield. Three-class deep learning classification AI models were developed using a motion-mode ultrasound dataset captured in animal study experiments from more than 25 swine subjects. Cluster analysis was used to define the “population” based on brightness, contrast, and kurtosis properties. A MobileNetV3 DL model architecture was tuned across a variety of hyperparameters, with the results ultimately being evaluated using images captured in real-time. Different hyperparameter configurations were blind-tested, resulting in models trained on filtered data having a real-time accuracy from 89 to 96%, as opposed to 78–95% when trained without filtering and optimization. The best model achieved a blind accuracy of 85% when inferencing on data collected in real-time, surpassing previous YOLOv8 models by 17%. AI models can be developed that are suitable for high performance in real-time for thoracic injury determination and are suitable for potentially addressing challenges with responding to emergency casualty situations and reducing the skill threshold for using and interpreting POCUS.

Schlagwörter

Stichwörter der Autor:innen

artificial intelligencedeep learningemergency medicinethoracictraumaultrasound imaging

Autor:innen und Institutionen

  • Organ Support and Automation Technologies Group, U.S. Army Institute of Surgical Research, JBSA Fort Sam Houston, San Antonio, TX 78234, USA.

ORCID-Profile: Eric J Snider

Förderung

  • United States Department of Defense
  • Oak Ridge Institute for Science and Education
  • U.S. Army Medical Research and Development Command
  • Science Education Programs at National Institutes of Health (NIH)

Zitieren

APA

Ruiz, A. J., Hernández Torres, S. I., & Snider, E. J. (2025). Development of Deep Learning Models for Real-Time Thoracic Ultrasound Image Interpretation. Journal of Imaging, 11(7), 222. https://doi.org/10.3390/jimaging11070222

Vancouver

Ruiz AJ, Hernández Torres SI, Snider EJ. Development of Deep Learning Models for Real-Time Thoracic Ultrasound Image Interpretation. Journal of Imaging. 2025;11(7):222. doi:10.3390/jimaging11070222

BibTeX
@article{Ruiz2025,
  author = {Ruiz, Austin J and Hernández Torres, Sofia I and Snider, Eric J},
  title = {Development of Deep Learning Models for Real-Time Thoracic Ultrasound Image Interpretation},
  journal = {Journal of Imaging},
  year = {2025},
  volume = {11},
  number = {7},
  pages = {222},
  doi = {10.3390/jimaging11070222}
}
Kostenlos starten →