Computer-aided detection-assisted colonoscopy: classification and relevance of false positives

      Background and Aims

      False positive (FP) results by computer-aided detection (CADe) hamper the efficiency of colonoscopy by extending examination time. Our aim was to develop a classification of the causes and clinical relevance of CADe FPs, and to assess the relative distribution of FPs in a real-life setting.

      Methods

      In a post-hoc analysis of a randomized trial comparing colonoscopy with and without CADe (NCT: 04079478), we extracted 40 CADe colonoscopy videos. Using a modified Delphi process, 4 expert endoscopists identified the main domains for the reasons and clinical relevance of FPs. Then, 2 expert endoscopists manually examined each FP and classified it according to the proposed domains. The analysis was limited to the withdrawal phase.

      Results

      The 2 main domains for the causes of CADe FPs were identified as artifacts due to either the mucosal wall or bowel content, and clinical relevance was defined as the time spent on FPs and the FP rate per minute. The mean number of FPs per colonoscopy was 27.3 ± 13.1, of which 24 ± 12 (88%) and 3.2 ± 2.6 (12%) were due to artifacts in the bowel wall and bowel content, respectively. Of the 27.3 FPs per colonoscopy, 1.6 (5.7%) required additional exploration time of 4.8 ± 6.2 seconds per FP (ie, 0.7% of the mean withdrawal time). In detail, 15 (24.2%), 33 (53.2%), and 14 (22.6%) FPs were classified as being of mild, moderate, or severe clinical relevance. The rate of FPs per minute of withdrawal time was 2.4 ± 1.2, and was higher for FPs due to artifacts from the bowel wall than for those from bowel content (2.4 ± 0.6 vs 0.3 ± 0.2, P < .001).

      Conclusions

      FPs by CADe are primarily due to artifacts from the bowel wall. Despite a high frequency, FPs result in a negligible 1% increase in the total withdrawal time because most of them are immediately discarded by the endoscopist.

      Abbreviations:

      CADe (computer-aided detection), CI (confidence interval), CRF (case report form), FP (false positive)
      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

        • Hassan C.
        • Wallace M.B.
        • Sharma P.
        • et al.
        New artificial intelligence system: first validation study versus experienced endoscopists for colorectal polyp detection.
        Gut. 2020; 69: 799-800
        • Hoogenboom S.A.
        • Bagci U.
        • Wallace M.B.
        AI in gastroenterology. The current state of play and the potential. How will it affect our practice and when?.
        Techniques in Gastrointest Endosc. 2019; : 150634
        • Lui T.K.
        • Guo C.-G.
        • Leung W.K.
        Accuracy of artificial intelligence on histology prediction and detection of colorectal polyps: a systematic review and meta-analysis.
        Gastrointest Endosc. 2020; 92: 11-22.e6
        • Greenhill A.T.
        • Edmunds B.R.
        A primer of artificial intelligence in medicine.
        Techn Gastrointest Endosc. 2020; 22: 85-89
        • Vinsard D.G.
        • Mori Y.
        • Misawa M.
        • et al.
        Quality assurance of computer-aided detection and diagnosis in colonoscopy.
        Gastrointest Endosc. 2019; 90: 55-63
        • Kudo S.-E.
        • Mori Y.
        • Misawa M.
        • et al.
        Artificial intelligence and colonoscopy: current status and future perspectives.
        Dig Endosc. 2019; 31: 363-371
        • Yamada M.
        • Saito Y.
        • Imaoka H.
        • et al.
        Development of a real-time endoscopic image diagnosis support system using deep learning technology in colonoscopy.
        Sci Rep. 2019; 9 (8): 14465
        • Urban G.
        • Tripathi P.
        • Alkayali T.
        • et al.
        Deep learning localizes and identifies polyps in real time with 96% accuracy in screening colonoscopy.
        Gastroenterology. 2018; 155: 1069-1078.e8
        • Mori Y.
        • Kudo S.-E.
        • Misawa M.
        • et al.
        Real-time use of artificial intelligence in identification of diminutive polyps during colonoscopy: a prospective study.
        Ann Intern Med. 2018; 169 (18): 357-366
        • Repici A.
        • Badalamenti M.
        • Maselli R.
        • et al.
        Efficacy of real-time computer-aided detection of colorectal neoplasia in a randomized trial.
        Gastroenterology. Epub 2020 May 1;
        • Custer R.L.
        • Scarcella J.A.
        • Stewart B.R.
        The modified Delphi technique - a rotational modification.
        J Career Tech Educ. 1999; 15: 50-58