Deep learning based dual stage model for accurate nasogastric tube positioning in chest radiographs.

Journal: Scientific reports
PMID:

Abstract

Accurate placement of nasogastric tubes (NGTs) is crucial for ensuring patient safety and effective treatment. Traditional methods relying on manual inspection are susceptible to human error, highlighting the need for innovative solutions. This study introduces a deep-learning model that enhances the detection and analysis of NGT positioning in chest radiographs. By integrating advanced segmentation and classification techniques, the model leverages the nnU-Net framework for segmenting critical regions and the ResNet50 architecture, pre-trained with MedCLIP, for classifying NGT placement. Trained on 1799 chest radiographs, the model demonstrates remarkable performance, achieving a Dice Similarity Coefficient of 65.35% for segmentation and an Area Under the Curve of 99.72% for classification. These results underscore its ability to accurately distinguish between correct and incorrect placements, outperforming traditional approaches. This method not only enhances diagnostic precision but also has the potential to streamline clinical workflows and improve patient care. A functional prototype of the model is accessible at https://ngtube.ziovision.ai .

Authors

  • Inseo Park
    Department of Research and Development, ZIOVISION Co. Ltd., Chuncheon, 24341, Kangwon, Republic of Korea.
  • Gwiseong Moon
    Department of Computer Science and Engineering, Kangwon National University, Kangwon-do, Korea.
  • Ji Young Hong
    Division of Pulmonary, Allergy and Critical Care Medicine, Department of Internal Medicine, Chuncheon Sacred Heart Hospital, Hallym University Medical Center, Chuncheon, Gangwon-do, Republic of Korea.
  • Jeongwon Heo
    Department of Internal Medicine, Kangwon National University, Chuncheon, 24341, Republic of Korea.
  • Hongseok Ko
    Department of Radiology, Seoul National College of Medicine, Seoul National University Hospital, 101 Daehak-ro, Jongno-gu, Seoul, 03080, Korea (H.C., S.H.Y., S.J.P., C.M.P., J.H.L., H. Kim, E.J.H., S.J.Y., J.G.N., C.H.L., J.M.G.); CHESS Center, The First Hospital of Lanzhou University, Lanzhou, China (Q.X., J.L.); Department of Radiology, Seoul National University Bundang Hospital, Gyeonggi-do, Korea (K.H.L.); Department of Internal Medicine, Incheon Medical Center, Incheon, Korea (J.Y.K.); Department of Radiology, Seoul Medical Center, Seoul, Korea (Y.K.L.); Department of Radiology, National Medical Center, Seoul, Korea (H. Ko); Department of Radiology, Myongji Hospital, Gyeonggi-do, Korea (K.H.K.); and Department of Radiology, Chonnam National University Hospital, Gwanju, Korea (Y.H.K.).
  • Doohee Lee
    Medical IP Co., Ltd, Seoul, Republic of Korea.
  • Yoon Kim
    Department of Research and Development, ZIOVISION Co. Ltd., Chuncheon, 24341, Kangwon, Republic of Korea.
  • Woo Jin Kim
    The Heart Center of Chonnam National University Hospital, 42 Jaebongro, Dong-gu, Gwangju 501-757, South Korea.
  • Hyun-Soo Choi
    Department of Electrical and Computer Engineering, Seoul National University, room 908 Bldg. 301, 1 Gwanak-ro, Gwanak-gu, Seoul, 08826, Korea.
  • Kyoung Min Moon
    Department of Pulmonary, Allergy, and Critical Care Medicine, Gangneung Asan Hospital, College of Medicine, University of Ulsan, 38, Bangdong-gil, Sacheon-myeon, Gangneung-si, 25440, Gangwon-do, Republic of Korea. pulmogicu@ulsan.ac.kr.