Advertisement

Development and validation of a convolutional neural network model for diagnosing Helicobacter pylori infections with endoscopic images – A multicenter study.

  • Author Footnotes
    ∗ co-first authors
    Ji Yeon Seo
    Footnotes
    ∗ co-first authors
    Affiliations
    Department of Internal Medicine, Healthcare Research Institute, Healthcare System Gangnam Center, Seoul National University Hospital, Seoul, Republic of Korea
    Search for articles by this author
  • Author Footnotes
    ∗ co-first authors
    Hotak Hong
    Footnotes
    ∗ co-first authors
    Affiliations
    Artificial Intelligence Research Center, JLK Inc., Seoul, Republic of Korea
    Search for articles by this author
  • Wi-Sun Ryu
    Affiliations
    Artificial Intelligence Research Center, JLK Inc., Seoul, Republic of Korea
    Search for articles by this author
  • Dongmin Kim
    Affiliations
    Artificial Intelligence Research Center, JLK Inc., Seoul, Republic of Korea
    Search for articles by this author
  • Author Footnotes
    ∗∗ co-corresponding authors
    Jaeyoung Chun
    Correspondence
    Corresponding author 2: Jaeyoung Chun MD. PhD., Division of Gastroenterology, Department of Internal Medicine, Gangnam Severance Hospital, Yonsei University College of Medicine, 20, Eonju-ro 63-gil, Gangnam-gu, Seoul, Republic of Korea 06229. , Tel: +82-2-2019-3310, Fax: +82-2-3463-3882
    Footnotes
    ∗∗ co-corresponding authors
    Affiliations
    Department of Internal Medicine, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, Republic of Korea
    Search for articles by this author
  • Author Footnotes
    ∗∗ co-corresponding authors
    Min-sun Kwak
    Correspondence
    Corresponding author 1: Min-Sun Kwak, MD. PhD., Department of Internal Medicine, Healthcare Research Institute, Healthcare System Gangnam Center, Seoul National University Hospital, 39FL., Gangnam Finance Center 737, Yeoksam-Dong, Gangnam-Gu, Seoul 06236, Korea. , Tel: +82-2-2112-5690, Fax: +82-2-2112-5635
    Footnotes
    ∗∗ co-corresponding authors
    Affiliations
    Department of Internal Medicine, Healthcare Research Institute, Healthcare System Gangnam Center, Seoul National University Hospital, Seoul, Republic of Korea
    Search for articles by this author
  • Author Footnotes
    ∗ co-first authors
    ∗∗ co-corresponding authors
Published:January 11, 2023DOI:https://doi.org/10.1016/j.gie.2023.01.007
      This paper is only available as a PDF. To read, Please Download here.

      Abstract

      Background and Aims

      Insufficient validation limits the generalizability of deep learning in diagnosing Helicobacter pylori (H. pylori) infection with endoscopic images. The aim of this study was to develop a deep learning model for the diagnosis of H. pylori infection using endoscopic images and validate the model with internal and external datasets.

      Methods

      A convolutional neural network (CNN) model was developed based on a training dataset comprising 13,403 endoscopic images from 952 patients who underwent endoscopy at Seoul National University Hospital Gangnam Center. Internal validation was performed using a separate dataset comprising the images of 411 individuals of Korean descent and 131 of non-Korean descent. External validation was performed with the images of 160 patients in Gangnam Severance Hospital. Gradient-weighted class activation mapping (Grad-CAM) was performed to visually explain the model.

      Results

      In predicting H. pylori ever-infected status, the sensitivity, specificity and accuracy of internal validation for people of Korean descent were 0.96 (95% CI 0.93–0.98), 0.90 (95% CI 0.85–0.95), and 0.94 (95% CI, 0.91-0.96), respectively. In the internal validation for people of non-Korean descent, the sensitivity, specificity and accuracy in predicting H. pylori ever-infected status were 0.92 (95% CI, 0.86-0.98), 0.79 (95% CI, 0.67-0.91) and 0.88 (95% CI, 0.82-0.93), respectively. In the external validation cohort, they were 0.86 (95% CI, 0.80-0.93), 0.88 (95% CI, 0.79-0.96), and 0.87 (95% CI, 0.82-0.92), respectively, when performing two-group categorization. The Grad-CAM showed that the CNN model captured the characteristic findings of each group.

      Conclusions

      This CNN model for diagnosing H. pylori infection showed good overall performance in internal and external validation datasets, particularly in categorizing patients into the never- versus ever-infected groups.

      Keywords

      To read this article in full you will need to make a payment

      Purchase one-time access:

      Academic & Personal: 24 hour online accessCorporate R&D Professionals: 24 hour online access
      One-time access price info
      • For academic or personal research use, select 'Academic and Personal'
      • For corporate R&D use, select 'Corporate R&D Professionals'

      Subscribe:

      Subscribe to Gastrointestinal Endoscopy
      Already a print subscriber? Claim online access
      Already an online subscriber? Sign in
      Institutional Access: Sign in to ScienceDirect