Skip to main content
ACG 2023 Annual Meeting Posters Main banner
Back
Rate ePoster

Monday Poster Session

Category: Colorectal Cancer Prevention

P1746 - Artificial Intelligence-Assisted Colonoscopy in Real World Clinical Practice: A Systematic Review and Meta-Analysis

Monday, October 23, 2023
10:30 AM – 4:15 PM PT
Location: Exhibit Hall
Has Audio

    Presenting Author(s)

  • Mike Wei, MD

    Stanford University School of Medicine
    Palo Alto, CA

Mike Wei, MD1, Shmuel Fay, MD2, Diana Yung, MD3, Uri Ladabaum, MD1, Uri Kopylov, MD4
1Stanford University School of Medicine, Palo Alto, CA; 2Sheba Medical Center, Ramat Gan, Tel Aviv, Israel; 3Gold Coast Hospital and Health Service, Gold Coast, Queensland, Australia; 4Gastroenterology Institute, Sheba Medical Center Tel Hashomer, Sackler School of Medicine, Tel Aviv University, Ramat Gan, Tel Aviv, Israel

Introduction: Artificial intelligence (AI) could minimize the operator-dependent variation in colonoscopy quality. Computer-aided detection (CADe) has improved adenoma detection rate (ADR) and adenomas per colonoscopy (APC) in randomized controlled trials (RCTs). There is a need to assess the impact of CADe in real-world settings.

Methods: We searched MEDLINE, EMBASE, and Web of Science for non-randomized, real-world studies of CADe in colonoscopy. Two physicians reviewed titles, abstracts and full texts and performed data extraction independently. Random-effects meta-analyses were performed to examine the effect of CADe on ADR and APC.

Results: Twelve of 1,314 studies met inclusion criteria [Table 1]. Overall, ADR was statistically significantly higher with vs. without CADe (36.3% vs. 35.8%, risk ratio [RR] 1.13, 95% CI 1.01-1.28) [Figure 1A]. This difference remained significant in subgroup analyses evaluating 6 prospective (37.3% vs 35.2%, RR 1.15, 95% CI 1.01-1.32) but not 6 retrospective (35.7% vs 36.2%, RR 1.12, 95% CI 0.92-1.36) studies. Among 6 studies with APC data, the APC rate ratio with vs. without CADe was 1.12 (95% CI 0.95-1.33) [Figure 1B]. In 4 studies with GI Genius (Medtronic), there was no difference in ADR with vs. without CADe (RR 0.96, 95% CI 0.85-1.07).

Discussion: ADR, but not APC, was slightly higher with vs. without CADe among all available real-world studies. This difference was attributed to the results of prospective but not retrospective studies. The discrepancies between these findings vs. those of RCTs call for future research on the true impact of current AI technology on colonoscopy quality, and the subtleties of human-AI interactions.

Figure: A. Pooled ADR risk ratio of all studies including abstracts B. Pooled rate ratio for APC with vs. without CAD

StudyYearLocationStudy DesignControlCADe used (N)Without CADe (N)CADe vs without CADe APC (p-value)CADe vs without CADe ADR (p-value)
Quan2022United StatesMulticenter prospectiveHistoricalEndoVigilant (N=300)3001.35 vs 1.07 (0.099)52 vs 46.3 (0.165)
Koh2022SingaporeSingle center prospectiveHistoricalGI Genius (N=298)NA 30.4 vs 24.3 (0.02)**
Ishiyama2021JapanSingle center prospectiveConcurrentEndoBRAINEYE (N=918)9180.42 vs 0.3, (0.003)26.4 vs 19.9 (0.001)
Shaukat2022United StatesSingle center prospectiveHistoricalSkout (N=83)2831.46 vs 1.01, (0.104)54.2 vs 40.6 (0.028)
Richter2023GermanySingle center retrospective HistoricalCADEye (N=163)140 0.39 vs 0.41 ( >0.05)
Nehme 2021 United States Single center prospectiveHistoricalGI Genius (N=403)6411.27 vs 1.17, (0.45)50.4 vs 53 (0.41)
Ahmad*2021England Single center prospectiveHistorical GI Genius (N=82)86  48.8 vs 46.5 (0.77)
Agazzi*2022ItalySingle center prospectiveHistorical CADEye (N=250)450 46 vs 30.7 (< 0.005)
Wong2022Hong KongSingle center retrospectiveHistorical ENDO-AID (N=119)115 52.9 vs 37.4 (0.017)
Schauer2021 New ZealandSingle center retrospective Historical ENDO-AID (N=213)213 47.9 vs 38.5 (0.03)
Levy 2022 Israel Single center retrospective HistoricalGI Genius (N=1,969)2,175 0.6 vs 0.68 (0.001)30.3 vs 35.2 (0.001)
Ladabaum2022United StatesSingle center retrospectiveHistorical and concurrentGI Genius (N=619)619 0.78 vs 0.89 (0.23)40.1 vs 41.8 (0.44)


Disclosures:
Mike Wei: AgilTX – Consultant. Capsovision – Consultant. Neptune Medical – Consultant.
Shmuel Fay indicated no relevant financial relationships.
Diana Yung indicated no relevant financial relationships.
Uri Ladabaum: Freenome – Consultant. Guardant – Consultant. Kohler Ventures – Consultant. Medial EarlySign – Consultant. Medtronic – Consultant. Neptune Medical – Consultant. Universial Dx – Advisory Committee/Board Member. Vivante – Advisory Committee/Board Member.
Uri Kopylov: Abbvie – Speakers Bureau. CTS – Consultant. Janssen – Grant/Research Support, Speakers Bureau. Takeda – Consultant, Grant/Research Support, Speakers Bureau.


Mike Wei, MD1, Shmuel Fay, MD2, Diana Yung, MD3, Uri Ladabaum, MD1, Uri Kopylov, MD4. P1746 - Artificial Intelligence-Assisted Colonoscopy in Real World Clinical Practice: A Systematic Review and Meta-Analysis, ACG 2023 Annual Scientific Meeting Abstracts. Vancouver, BC, Canada: American College of Gastroenterology.