EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Pilot Evaluation of a Deep Learning Model for Nasogastric Tube Verification on Chest Radiographs: A Single-Center Retrospective Study

Sang Won Park iD, Doohee Lee iD, Jae Eun Song, Yoon Kim iD, Hyun-Soo Choi iD, Seung-Joon Lee iD, Woo Jin Kim iD, Kyoung Min Moon, Oh Beom Kwon iD

Tomography · 2025

DOI: 10.3390/tomography11120140 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Tomography, 2025. Verfasst von Sang Won Park und 8 weiteren Autor:innen. Beteiligte Einrichtung: Department of Next Generation Information Center, Kangwon National University Hospital, Chuncheon 24289, Republic of Korea. Hauptthemen laut MEDLINE: Deep Learning, Intubation, Gastrointestinal, Radiography, Thoracic. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 21 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Science and ICT, Korea Technology R&D Project through the National IT Industry Promotion Agency (NIPA) and the Gangwon Information and Culture Industry Promotion Agency, Ministry of Science and ICT, Republic of Korea.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Accurate confirmation of nasogastric (NG) tubes is essential for patient safety, but delays and variability in interpretation remain common in clinical practice. Deep learning (DL) models have shown potential for assisting in this task, but real-world performance, particularly in detecting malpositioned tubes, remains insufficiently characterized.

Methods

We conducted a pilot evaluation of a previously developed DL model using 135 chest radiographs from Kangwon National University Hospital. Expert physicians established the reference standard. Model performance was assessed and receiver operating characteristic (ROC) curve and precision recall curve (PRC) analyses were performed. Differences between correctly classified and misclassified cases were examined using Wilcoxon rank-sum and Fisher’s exact tests to explore potential clinical or radiographic contributors to model failure.

Results

The model correctly classified 129 of 135 cases. The sensitivity was 96.1% (95% confidence interval (CI): 92.2–98.9%), specificity was 85.7% (95% CI: 42.2–97.7%), positive predictive value (PPV) was 99.2% (95% CI: 96.1–99.9%), negative predictive value (NPV) was 54.5% (95% CI: 25.4–80.8%), balanced accuracy was 90.8%, and F1-score was 0.976. The area under the ROC curve was 0.970 (95% CI: 0.929–1.000) and that under the PRC was 0.727 (95% CI: 0.289–1.000), reflecting substantial uncertainty related to the very small number of incomplete cases (n = 6). No statistically significant differences in clinical or radiographic characteristics were observed between correctly classified and misclassified cases.

Conclusions

The DL model performed well in identifying correctly positioned NG tubes but demonstrated limited and unstable performance for detecting incomplete placements. Given the safety implications of misclassification, the model should be used only as an assistive tool with mandatory physician oversight. Larger, multi-center studies with greater representation of incomplete cases are required to obtain more reliable estimates and support safe clinical implementation.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningIntubation, GastrointestinalRadiography, ThoracicHumansPilot ProjectsRetrospective StudiesFemaleMaleMiddle AgedAgedROC CurveSensitivity and SpecificityAdultAged, 80 and overPredictive Value of Tests

Stichwörter der Autor:innen

deep learning modelnasogastric tubereal-world validation

Autor:innen und Institutionen

  • Department of Next Generation Information Center, Kangwon National University Hospital, Chuncheon 24289, Republic of Korea.
  • Department of Research and Development, ZIOVISION Co., Ltd., Chuncheon 24341, Republic of Korea.
  • Department of Computer Science and Engineering, College of IT, Kangwon National University, Chuncheon 24341, Republic of Korea.
  • Department of Radiology, Kangwon National University Hospital, Chuncheon 24289, Republic of Korea.
  • Department of Computer Science and Engineering, Seoul National University of Science and Technology, Seoul 01811, Republic of Korea.
  • Division of Pulmonology, Department of Internal Medicine, Kangwon National University Hospital, Chuncheon 24289, Republic of Korea.

ORCID-Profile: Sang Won Park, Doohee Lee, Yoon Kim, Hyun-Soo Choi, Seung-Joon Lee, Woo Jin Kim, Oh Beom Kwon

Förderung

  • Ministry of Science and ICT
  • Korea Technology R&D Project through the National IT Industry Promotion Agency (NIPA) and the Gangwon Information and Culture Industry Promotion Agency
  • Ministry of Science and ICT, Republic of Korea

Zitieren

APA

Park, S. W., Lee, D., Song, J. E., Kim, Y., Choi, H. S., Lee, S. J., Kim, W. J., Moon, K. M., & Kwon, O. B. (2025). Pilot Evaluation of a Deep Learning Model for Nasogastric Tube Verification on Chest Radiographs: A Single-Center Retrospective Study. Tomography. https://doi.org/10.3390/tomography11120140

Vancouver

Park SW, Lee D, Song JE, Kim Y, Choi HS, Lee SJ, et al. Pilot Evaluation of a Deep Learning Model for Nasogastric Tube Verification on Chest Radiographs: A Single-Center Retrospective Study. Tomography. 2025. doi:10.3390/tomography11120140

BibTeX
@article{Park2025,
  author = {Park, Sang Won and Lee, Doohee and Song, Jae Eun and Kim, Yoon and Choi, Hyun-Soo and Lee, Seung-Joon and Kim, Woo Jin and Moon, Kyoung Min and Kwon, Oh Beom},
  title = {Pilot Evaluation of a Deep Learning Model for Nasogastric Tube Verification on Chest Radiographs: A Single-Center Retrospective Study},
  journal = {Tomography},
  year = {2025},
  doi = {10.3390/tomography11120140}
}
Kostenlos starten →