Advertisement

Deep learning model for diagnosing gastric mucosal lesions using endoscopic images: development, validation, and method comparison

  • Author Footnotes
    ∗ Drs Nam, H. J. Chung, K. S. Choi, and H. Lee contributed equally to this article as co–first authors.
    Joon Yeul Nam
    Footnotes
    ∗ Drs Nam, H. J. Chung, K. S. Choi, and H. Lee contributed equally to this article as co–first authors.
    Affiliations
    Department of Internal Medicine and Liver Research Institute, Seoul National University College of Medicine, Seoul, Korea
    Search for articles by this author
  • Author Footnotes
    ∗ Drs Nam, H. J. Chung, K. S. Choi, and H. Lee contributed equally to this article as co–first authors.
    Hyung Jin Chung
    Footnotes
    ∗ Drs Nam, H. J. Chung, K. S. Choi, and H. Lee contributed equally to this article as co–first authors.
    Affiliations
    Bio Imaging and Signal Processing Laboratory, Department of Bio and Brain Engineering, Korea Advanced Institute for Science and Technology, Daejeon, Korea
    Search for articles by this author
  • Author Footnotes
    ∗ Drs Nam, H. J. Chung, K. S. Choi, and H. Lee contributed equally to this article as co–first authors.
    Kyu Sung Choi
    Footnotes
    ∗ Drs Nam, H. J. Chung, K. S. Choi, and H. Lee contributed equally to this article as co–first authors.
    Affiliations
    Department of Radiology, Seoul National University College of Medicine, Seoul, Korea
    Search for articles by this author
  • Author Footnotes
    ∗ Drs Nam, H. J. Chung, K. S. Choi, and H. Lee contributed equally to this article as co–first authors.
    Hyuk Lee
    Footnotes
    ∗ Drs Nam, H. J. Chung, K. S. Choi, and H. Lee contributed equally to this article as co–first authors.
    Affiliations
    Department of Internal Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea
    Search for articles by this author
  • Tae Jun Kim
    Affiliations
    Department of Internal Medicine, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea
    Search for articles by this author
  • Hosim Soh
    Affiliations
    Department of Internal Medicine and Liver Research Institute, Seoul National University College of Medicine, Seoul, Korea
    Search for articles by this author
  • Eun Ae Kang
    Affiliations
    Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Korea
    Search for articles by this author
  • Soo-Jeong Cho
    Affiliations
    Department of Internal Medicine and Liver Research Institute, Seoul National University College of Medicine, Seoul, Korea
    Search for articles by this author
  • Jong Chul Ye
    Affiliations
    Bio Imaging and Signal Processing Laboratory, Department of Bio and Brain Engineering, Korea Advanced Institute for Science and Technology, Daejeon, Korea
    Search for articles by this author
  • Jong Pil Im
    Affiliations
    Department of Internal Medicine and Liver Research Institute, Seoul National University College of Medicine, Seoul, Korea
    Search for articles by this author
  • Sang Gyun Kim
    Affiliations
    Department of Internal Medicine and Liver Research Institute, Seoul National University College of Medicine, Seoul, Korea
    Search for articles by this author
  • Joo Sung Kim
    Affiliations
    Department of Internal Medicine and Liver Research Institute, Seoul National University College of Medicine, Seoul, Korea
    Search for articles by this author
  • Author Footnotes
    † Drs H. Chung and J.-H. Lee contributed equally to this article as co–senior authors.
    Hyunsoo Chung
    Footnotes
    † Drs H. Chung and J.-H. Lee contributed equally to this article as co–senior authors.
    Affiliations
    Department of Internal Medicine and Liver Research Institute, Seoul National University College of Medicine, Seoul, Korea
    Search for articles by this author
  • Author Footnotes
    † Drs H. Chung and J.-H. Lee contributed equally to this article as co–senior authors.
    Jeong-Hoon Lee
    Correspondence
    Reprint requests: Jeong-Hoon Lee, MD, PhD, Department of Internal Medicine, Seoul National University Hospital, 101, Daehak-ro, Jongno-gu, Seoul 03080, Republic of Korea.
    Footnotes
    † Drs H. Chung and J.-H. Lee contributed equally to this article as co–senior authors.
    Affiliations
    Department of Internal Medicine and Liver Research Institute, Seoul National University College of Medicine, Seoul, Korea
    Search for articles by this author
  • Author Footnotes
    ∗ Drs Nam, H. J. Chung, K. S. Choi, and H. Lee contributed equally to this article as co–first authors.
    † Drs H. Chung and J.-H. Lee contributed equally to this article as co–senior authors.
Published:September 04, 2021DOI:https://doi.org/10.1016/j.gie.2021.08.022

      Background and Aims

      Endoscopic differential diagnoses of gastric mucosal lesions (benign gastric ulcer, early gastric cancer [EGC], and advanced gastric cancer) remain challenging. We aimed to develop and validate convolutional neural network–based artificial intelligence (AI) models: lesion detection, differential diagnosis (AI-DDx), and invasion depth (AI-ID; pT1a vs pT1b among EGC) models.

      Methods

      This study included 1366 consecutive patients with gastric mucosal lesions from 2 referral centers in Korea. One representative endoscopic image from each patient was used. Histologic diagnoses were set as the criterion standard. Performance of the AI-DDx (training/internal/external validation set, 1009/112/245) and AI-ID (training/internal/external validation set, 620/68/155) was compared with visual diagnoses by independent endoscopists (stratified by novice [<1 year of experience], intermediate [2-3 years of experience], and expert [>5 years of experience]) and EUS results, respectively.

      Results

      The AI-DDx showed good diagnostic performance for both internal (area under the receiver operating characteristic curve [AUROC] = .86) and external validation (AUROC = .86). The performance of the AI-DDx was better than that of novice (AUROC = .82, P = .01) and intermediate endoscopists (AUROC = .84, P = .02) but was comparable with experts (AUROC = .89, P = .12) in the external validation set. The AI-ID showed a fair performance in both internal (AUROC = .78) and external validation sets (AUROC = .73), which were significantly better than EUS results performed by experts (internal validation, AUROC = .62; external validation, AUROC = .56; both P < .001).

      Conclusions

      The AI-DDx was comparable with experts and outperformed novice and intermediate endoscopists for the differential diagnosis of gastric mucosal lesions. The AI-ID performed better than EUS for evaluation of invasion depth.

      Abbreviations:

      AGC (advanced gastric cancer), AI (artificial intelligence), AI-DDx (artificial intelligence differential diagnosis), AI-ID (artificial intelligence invasion depth), AI-LD (artificial intelligence lesion detection), AUROC (area under the receiver operating characteristic curve), BGU (benign gastric ulcer), CNN (convolutional neural network), EGC (early gastric cancer), Grad-AM (gradient-weighted class activation mapping), ROI (region of interest)
      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

      References

        • Faulx A.L.
        • Lightdale J.R.
        • Acosta R.D.
        • et al.
        Guidelines for privileging, credentialing, and proctoring to perform GI endoscopy.
        Gastrointest Endosc. 2017; 85: 273-281
        • Park W.G.
        • Shaheen N.J.
        • Cohen J.
        • et al.
        Quality indicators for EGD.
        Gastrointest Endosc. 2015; 81: 17-30
        • Pimenta-Melo A.R.
        • Monteiro-Soares M.
        • Libânio D.
        • et al.
        Missing rate for gastric cancer during upper gastrointestinal endoscopy: a systematic review and meta-analysis.
        Eur J Gastroenterol Hepatol. 2016; 28: 1041-1049
        • Adler D.G.
        • Bakis G.
        • Coyle W.J.
        • et al.
        Principles of training in GI endoscopy.
        Gastrointest Endosc. 2012; 75: 231-235
        • Veitch A.M.
        • Uedo N.
        • Yao K.
        • et al.
        Optimizing early upper gastrointestinal cancer detection at endoscopy.
        Nat Rev Gastroenterol Hepatol. 2015; 12: 660-667
        • Zhang Q.
        • Chen Z-y
        • Chu-di Chen T.L.
        • et al.
        Training in early gastric cancer diagnosis improves the detection rate of early gastric cancer: an observational study in China.
        Medicine. 2015; 94 (e384)
        • Maple J.T.
        • Dayyeh B.K.A.
        • Chauhan S.S.
        • et al.
        Endoscopic submucosal dissection.
        Gastrointest Endosc. 2015; 81: 1311-1325
        • Yanai H.
        • Matsumoto Y.
        • Harada T.
        • et al.
        Endoscopic ultrasonography and endoscopy for staging depth of invasion in early gastric cancer: a pilot study.
        Gastrointest Endosc. 1997; 46: 212-216
        • Akahoshi K.
        • Chijiiwa Y.
        • Hamada S.
        • et al.
        Pretreatment staging of endoscopically early gastric cancer with a 15 MHz ultrasound catheter probe.
        Gastrointest Endosc. 1998; 48: 470-476
        • Kim J.-H.
        • Song K.S.
        • Youn Y.H.
        • et al.
        Clinicopathologic factors influence accurate endosonographic assessment for early gastric cancer.
        Gastrointest Endosc. 2007; 66: 901-908
        • Jin E.H.
        • Lee D.
        • Bae J.H.
        • et al.
        Improved accuracy in optical diagnosis of colorectal polyps using convolutional neural networks with visual explanations.
        Gastroenterology. 2020; 158: 2169-2179
        • Hirasawa T.
        • Aoyama K.
        • Tanimoto T.
        • et al.
        Application of artificial intelligence using a convolutional neural network for detecting gastric cancer in endoscopic images.
        Gastric Cancer. 2018; 21: 653-660
        • Tsuboi A.
        • Oka S.
        • Aoyama K.
        • et al.
        Artificial intelligence using a convolutional neural network for automatic detection of small-bowel angioectasia in capsule endoscopy images.
        Dig Endosc. 2020; 32: 382-390
        • Nagtegaal I.D.
        • Odze R.D.
        • Klimstra D.
        • et al.
        The 2019 WHO classification of tumours of the digestive system.
        Histopathology. 2020; 76: 182-188
      1. Selvaraju RR, Cogswell M, Das A, et al. Grad-cam: visual explanations from deep networks via gradient-based localization. Proceedings of the IEEE International Conference on Computer Vision. IEEE Xplore Digital Library. 2017:618-26.

        • Jung K.-W.
        • Won Y.-J.
        • Kong H.-J.
        • et al.
        Cancer statistics in Korea: incidence, mortality, survival, and prevalence in 2015.
        Cancer Res Treat. 2018; 50: 303
        • Sharma P.
        • Pante A.
        • Gross S.A.
        Artificial intelligence in endoscopy.
        Gastrointest Endosc. 2020; 91: 925-931
        • Yoon H.J.
        • Kim S.
        • Kim J.-H.
        • et al.
        A lesion-based convolutional neural network improves endoscopic detection and depth prediction of early gastric cancer.
        J Clin Med. 2019; 8: 1310
        • Zhu Y.
        • Wang Q.-C.
        • Xu M.-D.
        • et al.
        Application of convolutional neural network in the diagnosis of the invasion depth of gastric cancer based on conventional endoscopy.
        Gastrointest Endosc. 2019; 89: 806-815
        • Han Y.
        • Sun S.
        • Guo J.
        • et al.
        Is endoscopic ultrasonography useful for endoscopic submucosal dissection?.
        Endosc Ultrasound. 2016; 5: 284
        • Kim J.
        • Kim S.G.
        • Chung H.
        • et al.
        Clinical efficacy of endoscopic ultrasonography for decision of treatment strategy of gastric cancer.
        Surg Endosc. 2018; 32: 3789-3797
        • Feng F.
        • Liu J.
        • Wang F.
        • et al.
        Prognostic value of differentiation status in gastric cancer.
        BMC Cancer. 2018; 18: 865
        • jp JGCAjkk-ma
        Japanese gastric cancer treatment guidelines 2018.
        Gastric Cancer. 2020; 24: 1-21