EUVIMEDEUVIMED entdecken
VergleichsstudieBeobachtungsstudieOpen Access · CC BY-NC 4.0

Comparison of lung ultrasound assisted by artificial intelligence to radiology examination in pneumothorax

Chengdi Yang, Huijing Zhao, Anqi Wang, Jian Li, Jianling Gao iD

Journal of Clinical Ultrasound · 2024-10 · 52 (8) · 1051-1055

DOI: 10.1002/jcu.23756 ↗

Kurz gefasst

Vergleichsstudie in der Zeitschrift Journal of Clinical Ultrasound, online erschienen am 29.06.2024. Verfasst von Chengdi Yang und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Critical Care Medicine, The Fourth Affiliated Hospital of Soochow University, Suzhou, Jiangsu, China. Hauptthemen laut MEDLINE: Artificial Intelligence, Lung, Pneumothorax, Ultrasonography. Die Arbeit wurde bisher 16-mal zitiert (Crossref) und nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Science and Technology Program of Suzhou.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Lung ultrasound can evaluate for pneumothorax but the accuracy of diagnosis depends on experience among physicians. This study aimed to investigate the sensitivity and specificity of intelligent lung ultrasound in comparison with chest x‐ray, employing chest computed tomography (CT) as the gold standard for diagnosis of pneumothorax in critical ill patients.

Methods

This prospective, observational study included 75 dyspnea patients admitted to the Intensive Care Unit of the Fourth Affiliated Hospital of Soochow University from January 2021 to April 2023. Lung ultrasound images were collected using BLUE‐plus protocol and analyzed by artificial intelligence software to identify the pleural line, with CT results serving as the gold standard for diagnosis. Pneumothorax was diagnosed based on either the disappearance of pleural slip sign or identification of lung point. Additionally, chest x‐ray images and diagnostic results were also obtained during the same period for comparison.

Results

The sensitivity and specificity of intelligent lung ultrasound in diagnosing pneumothorax were 79.4% and 85.4%, respectively. The sensitivity and specificity of x‐ray diagnosis were 82.4% and 80.5%. Additionally, the diagnostic time for lung ultrasound was significantly shorter than that for x‐ray examination.

Conclusion

Intelligent lung ultrasound has diagnostic efficiency comparable to that of x‐ray examination but offers advantages in terms of speed.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Artificial IntelligenceLungPneumothoraxUltrasonographyAdultAgedFemaleHumansMaleMiddle AgedProspective StudiesRadiography, ThoracicSensitivity and SpecificityTomography, X-Ray Computed

Stichwörter der Autor:innen

automatic recognitionintelligent lung ultrasoundpleural linepneumothoraxx‐ray

Autor:innen und Institutionen

  • Department of Critical Care Medicine, The Fourth Affiliated Hospital of Soochow University, Suzhou, Jiangsu, China.
  • Department of Anesthesiology, The Fourth Affiliated Hospital of Soochow University, Suzhou, Jiangsu, China.
  • Department of Anesthesiology, the First Affiliated Hospital of Soochow University, Suzhou, Jiangsu, China.

ORCID-Profile: Jianling Gao

Förderung

  • Science and Technology Program of Suzhou

Zitieren

APA

Yang, C., Zhao, H., Wang, A., Li, J., & Gao, J. (2024). Comparison of lung ultrasound assisted by artificial intelligence to radiology examination in pneumothorax. Journal of Clinical Ultrasound, 52(8), 1051-1055. https://doi.org/10.1002/jcu.23756

Vancouver

Yang C, Zhao H, Wang A, Li J, Gao J. Comparison of lung ultrasound assisted by artificial intelligence to radiology examination in pneumothorax. Journal of Clinical Ultrasound. 2024;52(8):1051-1055. doi:10.1002/jcu.23756

BibTeX
@article{Yang2024,
  author = {Yang, Chengdi and Zhao, Huijing and Wang, Anqi and Li, Jian and Gao, Jianling},
  title = {Comparison of lung ultrasound assisted by artificial intelligence to radiology examination in pneumothorax},
  journal = {Journal of Clinical Ultrasound},
  year = {2024},
  volume = {52},
  number = {8},
  pages = {1051--1055},
  doi = {10.1002/jcu.23756}
}
Kostenlos starten →