检索问题
AI辅助结肠镜筛查的腺瘤检出率提升、长期临床获益、成本效果和医保支付价值证据
AI辅助结肠镜筛查在腺瘤检出率提升、长期临床获益、成本效果和医保支付价值方面展现出多方面的证据，但同时也存在一些局限性和挑战。以下将对这些方面进行详细阐述。
腺瘤检出率提升
多项研究和系统评价一致表明，人工智能（AI）辅助的计算机辅助检测（CADe）系统能够显著提高结肠镜检查中的腺瘤检出率（ADR）和平均每次结肠镜检查的腺瘤数量（APC）。
-
ADR的显著提高：
- 一项包含44项随机对照试验（RCTs）和36,201例病例的系统评价和荟萃分析显示，CADe辅助结肠镜检查的ADR更高（44.7% vs 36.7%，比率比 [RR] = 1.21）。
- 另一项纳入5项RCTs（4354名患者）的荟萃分析发现，CADe组的ADR显著高于对照组（36.6% vs 25.2%，RR = 1.44）。
- 一项包含33项试验（27,404名患者）的系统评价和荟萃分析也观察到AI辅助结肠镜检查的ADR显著增加（RR, 1.242）。
- 在亚洲人群进行的一项多中心RCT中，AI辅助结肠镜检查显著提高了总体ADR（39.9% vs 32.4%），甚至包括晚期腺瘤的ADR（6.6% vs 4.9%）。
-
APC的增加：
- CADe辅助结肠镜检查的平均APC也更高（0.98 vs 0.78，发生率差异 [IRD] = 0.22）。
- 在包含5项RCTs的荟萃分析中，CADe组的APC高于对照组（0.58 vs 0.36，RR = 1.70）。
- 另一项研究显示，AI辅助结肠镜检查每次结肠镜检查检测到的息肉数量（PPC）和APC都有显著增加，分别平均多出0.271个PPC和0.202个APC。
-
对不同类型腺瘤的检测效果：
- CADe系统在检测不同大小、形态和位置的腺瘤方面都表现出更高的APC，包括≤5毫米、6-9毫米、≥10毫米的腺瘤，以及近端、远端、扁平状和息肉状的腺瘤。
- 在组织学方面，CADe也导致每次结肠镜检查检测到的锯齿状病变数量增加（RR, 1.52）。
-
减少腺瘤漏诊率（AMR）：
- AI辅助结肠镜检查显著降低了息肉漏诊率（PMR）（RR, 0.475）和AMR（RR, 0.495），这意味着AI系统能够有效减少内镜医师的漏诊。
- CADe相对降低了55%的漏诊率。
-
对不同经验内镜医师的影响：
- AI辅助结肠镜检查对有经验和经验不足的内镜医师都能提高ADR，表明AI工具可以弥补操作者的经验差异，提升整体质量。一项研究发现，AI辅助结肠镜检查在经验不足的检查者中提高了ADR，并且检查者的经验水平在多变量分析中似乎对ADR差异影响较小。
-
对高级结直肠肿瘤（ACN）的影响：
- CADe辅助结肠镜检查的平均ACN数量相似（0.16 vs 0.15，IRD = 0.01），但ACN检出率（ACN DR）略有提高（12.7% vs 11.5%，RR = 1.16）。
- 然而，也有研究指出CADe虽然增加了腺瘤的检测，但并未增加晚期腺瘤的检测，并且对晚期ADR的影响不显著（RR, 1.35；P = .33）。一项研究提到AI方法目前并未增加癌症或大型腺瘤性息肉的检测，但有助于检测小型癌前息肉。
-
对操作时间的影响：
- AI辅助结肠镜检查通常会导致总退镜时间略微延长（0.53分钟）。
- 一项研究报告平均检查时间仅略微增加（0.47分钟）。
- 在多中心RCT中，AI辅助结肠镜组的中位退镜时间略长（8.3分钟 vs 7.8分钟）。
长期临床获益
AI辅助结肠镜检查的长期临床获益主要体现在结直肠癌（CRC）发病率和死亡率的降低，这是通过更有效地检测和切除癌前息肉来实现的。
-
降低CRC发病率和死亡率：
- 模型研究表明，与不使用AI的筛查结肠镜检查相比，AI辅助结肠镜检查能够进一步降低CRC发病率和死亡率。在一次主要分析中，与不筛查相比，无AI工具的结肠镜筛查使CRC发病率相对降低44.2%，而AI工具使发病率相对降低48.9%（增量收益为4.8%）。
- 在CRC死亡率方面，无AI筛查的相对降低率为48.7%，而AI辅助筛查为52.3%（增量收益为3.6%）。
- 在国家层面，实施AI检测工具每年可额外预防7194例CRC病例和2089例相关死亡。
- 世界内镜组织（WEO）的立场声明指出，长期来看，CADe增加的成本可能会被与癌症治疗相关的成本节省所抵消，因为CADe能够预防癌症。
-
减少间期性结直肠癌：
- 结肠镜检查中高达四分之一的结直肠肿瘤会被漏诊，这是导致间期性结直肠癌的主要原因。AI系统通过提高息肉检测率和降低漏诊率，有望减少间期性结直肠癌的发生。
- 通过移除所有可能恶变的息肉，AI辅助结肠镜检查有望降低间期性结直肠癌的发生率，使得息肉的完全清除成为衡量内镜质量的重要指标。
-
未来前景和挑战：
- 尽管有积极的初步结果，但AI辅助结肠镜检查在降低癌症发病率和死亡率方面的实际长期效果仍需进一步的大规模研究和随访数据来证实。
- 目前的证据主要集中在替代指标，如腺瘤检出率，而对患者重要的结局如癌症发病率和死亡率的直接影响尚不明确。
- 需要对AI在不同人群中的有效性和泛化能力进行验证，并结合计算机辅助诊断（CADx）工具，以实现实时组织学预测和个性化监测策略，从而最大化长期获益。
成本效果
AI辅助结肠镜检查的成本效果是一个复杂的问题，它涉及到初始技术投入、操作时间增加、不必要的息肉切除，以及通过预防CRC所带来的长期医疗成本节省。
-
潜在的成本节省：
- 多项研究表明，AI辅助结肠镜检查可能是一种成本节约策略，主要是通过预防CRC的发生及其后续治疗费用。
- 一项对美国平均风险人群的Markov模型微观模拟研究显示，AI检测工具使每次筛查的折扣成本从3400美元降低到3343美元，每人节省57美元。在全国范围内，这意味着每年节省2.9亿美元。
- 在西班牙进行的另一项成本效益分析发现，结合计算机辅助检测和诊断（CADe/CADx）的智能内镜模块与标准实践相比，是一种更具成本效益的策略。CADe/CADx在生命周期内更有效（16.37 LYG 和 14.32 QALYs），并且每次患者的总成本更低（2300.76欧元 vs 2508.75欧元）。
- 这项西班牙的研究预测，每1000名患者，CADe/CADx可避免173次息肉切除、370次组织病理学检查和7例CRC病例。
-
初始成本和增加的医疗负担：
- AI工具的实施面临经济方面的障碍和报销挑战。WEO的立场声明指出，短期内，CADe的使用可能会通过检测更多腺瘤而增加医疗成本。
- AI辅助结肠镜检查的一个“弊端”是可能导致非肿瘤性息肉的切除数量增加。一项研究指出，使用CADe系统导致每10次结肠镜检查中额外切除近2个非肿瘤性息肉。
- 对非肿瘤性息肉的过度切除以及伴随的病理评估会增加医疗费用。
- 尽管有这些成本，但长期来看，通过预防癌症实现的成本节省有望抵消这些增加的费用。
-
成本效益分析方法：
- 成本效益分析在评估AI在结肠镜检查中整合的可行性方面起着关键作用，它能揭示潜在的长期益处和初始经济负担。
- 一项基于Markov模型的成本效益分析，针对亚洲人群的结直肠癌筛查策略，发现粪便免疫化学试验（FIT）结合AI辅助结肠镜检查是最具成本效益的策略，具有最低的增量成本效益比（ICER）。在主要筛查方法中，AI辅助结肠镜检查在成本效益方面优于传统结肠镜检查（ICER -39,040美元）。
- 研究强调需要进行广泛的高质量成本效益研究，以了解AI在不同医疗系统中的实施是否能使人口和社会受益。
医保支付价值证据
医保支付的价值评估需要考虑临床获益、成本效果以及监管批准和报销途径。
-
监管批准与报销：
- AI工具的广泛采用需要解决监管批准和报销途径的问题。
- 在日本，EndoBRAIN等CADe工具已获得监管批准，并且自2024年起获得了报销，这预计将加速AI在结肠镜检查中的实施。
- 然而，目前关于AI在结肠镜检查中的临床指南仍然缺乏，这强调了需要对现有证据进行严格评估，以优化AI在结肠镜实践中的采用。
- WEO建议医疗服务提供系统和主管部门评估CADe和CADx的成本效益，以支持其在临床实践中的使用。
-
挑战与考量：
- 医保支付需要平衡AI的益处和潜在的危害，例如过度诊断和过度治疗。
- AI辅助系统虽然能提高腺瘤检出率，但可能增加对非肿瘤性息肉的切除，从而增加额外的医疗成本。这些成本需要在医保支付决策中进行权衡。
- 此外，“操作者技能退化”（deskilling）的风险也是一个需要关注的问题，即长期依赖AI可能导致内镜医师在没有AI辅助时表现下降。一项研究发现，持续接触AI可能降低标准非AI辅助结肠镜检查的ADR，这表明对内镜医师行为可能产生负面影响。这种潜在的风险可能影响医保支付对AI工具的长期评估。
- AI在真实世界中的整合、人机交互的理解、成本效益分析的进一步深入、以及建立培训和报销路径是成功采纳AI的关键。
总结
综合来看，AI辅助结肠镜筛查在提高腺瘤检出率方面具有显著优势，能够减少漏诊，并对不同经验水平的内镜医师都有效。在长期临床获益方面，模型研究表明AI有助于降低结直肠癌的发病率和死亡率，但仍需更多长期数据验证。在成本效果方面，虽然初期可能会增加不必要的息肉切除和操作时间，但通过预防癌症，长期有望实现医疗成本的节省，并被一些研究视为成本节约策略。对于医保支付，日本等国家已经开始提供报销，但全球范围内仍需更多高质量的成本效益研究、明确的临床指南以及解决过度诊断和操作者技能退化等挑战，以全面评估其支付价值。未来的发展将倾向于整合多模态AI系统、个性化监测和AI辅助治疗干预，从而进一步优化结肠镜检查实践并提升患者护理质量。
References
1Artificial Intelligence-Assisted Colonoscopy for Polyp Detection : A Systematic Review and Meta-analysis.PubMed
Saeed Soleymanjahi, Jack Huebner, Lina Elmansy, et al.
BACKGROUND: Randomized clinical trials (RCTs) of computer-aided detection (CADe) system-enhanced colonoscopy compared with conventional colonoscopy suggest increased adenoma detection rate (ADR) and decreased adenoma miss rate (AMR), but the effect on detection of advanced colorectal neoplasia (ACN) is unclear. PURPOSE: To conduct a systematic review to compare performance of CADe-enhanced and conventional colonoscopy. DATA SOURCES: Cochrane Library, Google Scholar, Ovid EMBASE, Ovid MEDLINE, PubMed, Scopus, and Web of Science Core Collection databases were searched through February 2024. STUDY SELECTION: Published RCTs comparing CADe-enhanced and conventional colonoscopy. DATA EXTRACTION: Average adenoma per colonoscopy (APC) and ACN per colonoscopy were primary outcomes. Adenoma detection rate, AMR, and ACN detection rate (ACN DR) were secondary outcomes. Balancing outcomes included withdrawal time and resection of nonneoplastic polyps (NNPs). Subgroup analyses were done by neural network architecture. DATA SYNTHESIS: Forty-four RCTs with 36 201 cases were included. Computer-aided detection-enhanced colonoscopies have higher average APC (12 090 of 12 279 [0.98] vs. 9690 of 12 292 [0.78], incidence rate difference [IRD] = 0.22 [95% CI, 0.16 to 0.28]) and higher ADR (7098 of 16 253 [44.7%] vs. 5825 of 15 855 [36.7%], rate ratio [RR] = 1.21 [CI, 1.15 to 1.28]). Average ACN per colonoscopy was similar (1512 of 9296 [0.16] vs. 1392 of 9121 [0.15], IRD = 0.01 [CI, -0.01 to 0.02]), but ACN DR was higher with CADe system use (1260 of 9899 [12.7%] vs. 1119 of 9746 [11.5%], RR = 1.16 [CI, 1.02 to 1.32]). Using CADe systems resulted in resection of almost 2 extra NNPs per 10 colonoscopies and longer total withdrawal time (0.53 minutes [CI, 0.30 to 0.77]). LIMITATION: Statistically significant heterogeneity in quality and sample size and inability to blind endoscopists to the intervention in included studies may affect the performance estimates. CONCLUSION: Computer-aided detection-enhanced colonoscopies have increased APC and detection rate but no difference in ACN per colonoscopy and a small increase in ACN DR. There is minimal increase in procedure time and no difference in performance across neural network architectures. PRIMARY FUNDING SOURCE: None. (PROSPERO: CRD42023422835).
2Performance of artificial intelligence in colonoscopy for adenoma and polyp detection: a systematic review and meta-analysis.PubMed
Cesare Hassan, Marco Spadaccini, Andrea Iannone, et al.
BACKGROUND AND AIMS: One-fourth of colorectal neoplasia are missed at screening colonoscopy, representing the main cause of interval colorectal cancer. Deep learning systems with real-time computer-aided polyp detection (CADe) showed high accuracy in artificial settings, and preliminary randomized controlled trials (RCTs) reported favorable outcomes in the clinical setting. The aim of this meta-analysis was to summarize available RCTs on the performance of CADe systems in colorectal neoplasia detection. METHODS: We searched MEDLINE, EMBASE, and Cochrane Central databases until March 2020 for RCTs reporting diagnostic accuracy of CADe systems in the detection of colorectal neoplasia. The primary outcome was pooled adenoma detection rate (ADR), and secondary outcomes were adenoma per colonoscopy (APC) according to size, morphology, and location; advanced APC; polyp detection rate; polyps per colonoscopy; and sessile serrated lesions per colonoscopy. We calculated risk ratios (RRs), performed subgroup and sensitivity analyses, and assessed heterogeneity and publication bias. RESULTS: Overall, 5 randomized controlled trials (4354 patients) were included in the final analysis. Pooled ADR was significantly higher in the CADe group than in the control group (791/2163 [36.6%] vs 558/2191 [25.2%]; RR, 1.44; 95% confidence interval [CI], 1.27-1.62; P < .01; I = 42%). APC was also higher in the CADe group compared with control (1249/2163 [.58] vs 779/2191 [.36]; RR, 1.70; 95% CI, 1.53-1.89; P < .01; I = 33%). APC was higher for ≤5-mm (RR, 1.69; 95% CI, 1.48-1.84), 6- to 9-mm (RR, 1.44; 95% CI, 1.19-1.75), and ≥10-mm adenomas (RR, 1.46; 95% CI, 1.04-2.06) and for proximal (RR, 1.59; 95% CI, 1.34-1.88), distal (RR, 1.68; 95% CI, 1.50-1.88), flat (RR, 1.78; 95% CI, 1.47-2.15), and polypoid morphology (RR, 1.54; 95% CI, 1.40-1.68). Regarding histology, CADe resulted in a higher sessile serrated lesion per colonoscopy (RR, 1.52; 95% CI, 1.14-2.02), whereas a nonsignificant trend for advanced ADR was found (RR, 1.35; 95% CI, .74-2.47; P = .33; I = 69%). Level of evidence for RCTs was graded as moderate. CONCLUSIONS: According to available evidence, the incorporation of artificial intelligence as aid for detection of colorectal neoplasia results in a significant increase in the detection of colorectal neoplasia, and such effect is independent from main adenoma characteristics.
3Cost-effectiveness of artificial intelligence for screening colonoscopy: a modelling study.PubMed
Miguel Areia, Yuichi Mori, Loredana Correale, et al.
BACKGROUND: Artificial intelligence (AI) tools increase detection of precancerous polyps during colonoscopy and might contribute to long-term colorectal cancer prevention. The aim of the study was to investigate the incremental effect of the implementation of AI detection tools in screening colonoscopy on colorectal cancer incidence and mortality, and the cost-effectiveness of such tools. METHODS: We conducted Markov model microsimulation of using colonoscopy with and without AI for colorectal cancer screening for individuals at average risk (no personal or family history of colorectal cancer, adenomas, inflammatory bowel disease, or hereditary colorectal cancer syndrome). We ran the microsimulation in a hypothetical cohort of 100 000 individuals in the USA aged 50-100 years. The primary analysis investigated screening colonoscopy with versus without AI every 10 years starting at age 50 years and finishing at age 80 years, with follow-up until age 100 years, assuming 60% screening population uptake. In secondary analyses, we modelled once-in-life screening colonoscopy at age 65 years in adults aged 50-79 years at average risk for colorectal cancer. Post-polypectomy surveillance followed the simplified current guideline. Costs of AI tools and cost for downstream treatment of screening detected disease were estimated with 3% annual discount rates. The main outcome measures included the incremental effect of AI-assisted colonoscopy versus standard (no-AI) colonoscopy on colorectal cancer incidence and mortality, and cost-effectiveness of screening projected for the average risk screening US population. FINDINGS: In the primary analyses, compared with no screening, the relative reduction of colorectal cancer incidence with screening colonoscopy without AI tools was 44·2% and with screening colonoscopy with AI tools was 48·9% (4·8% incremental gain). Compared with no screening, the relative reduction in colorectal cancer mortality with screening colonoscopy with no AI was 48·7% and with screening colonoscopy with AI was 52·3% (3·6% incremental gain). AI detection tools decreased the discounted costs per screened individual from $3400 to $3343 (a saving of $57 per individual). Results were similar in the secondary analyses modelling once-in-life colonoscopy. At the US population level, the implementation of AI detection during screening colonoscopy resulted in yearly additional prevention of 7194 colorectal cancer cases and 2089 related deaths, and a yearly saving of US$290 million. INTERPRETATION: Our findings suggest that implementation of AI detection tools in screening colonoscopy is a cost-saving strategy to further prevent colorectal cancer incidence and mortality. FUNDING: European Commission and Japan Society of Promotion of Science.
4Cost-Effectiveness for Artificial Intelligence in Colonoscopy.PubMed
Natalie Halvorsen, Yuichi Mori
Artificial intelligence (AI) is set to transform the field of colonoscopy through the implementation of computer-assisted detection and diagnosis. While over 20 randomized controlled trials have demonstrated the efficacy of AI in increasing adenoma detection rate, the broader implementation of these technologies faces hurdles due to economic considerations and reimbursement challenges. Cost-effectiveness analysis plays a crucial role in determining the viability of integrating AI into clinical practice, highlighting both the potential long-term benefits and the initial economic burdens. Comprehensive evaluation and large-scale studies are essential to realize AI's potential in optimizing colonoscopy procedures and reducing health care costs.
5Artificial intelligence for colorectal neoplasia detection during colonoscopy: a systematic review and meta-analysis of randomized clinical trials.PubMed
Shenghan Lou, Fenqi Du, Wenjie Song, et al.
BACKGROUND: The use of artificial intelligence (AI) in detecting colorectal neoplasia during colonoscopy holds the potential to enhance adenoma detection rates (ADRs) and reduce adenoma miss rates (AMRs). However, varied outcomes have been observed across studies. Thus, this study aimed to evaluate the potential advantages and disadvantages of employing AI-aided systems during colonoscopy. METHODS: Using Medical Subject Headings (MeSH) terms and keywords, a comprehensive electronic literature search was performed of the Embase, Medline, and the Cochrane Library databases from the inception of each database until October 04, 2023, in order to identify randomized controlled trials (RCTs) comparing AI-assisted with standard colonoscopy for detecting colorectal neoplasia. Primary outcomes included AMR, ADR, and adenomas detected per colonoscopy (APC). Secondary outcomes comprised the poly missed detection rate (PMR), poly detection rate (PDR), and poly detected per colonoscopy (PPC). We utilized random-effects meta-analyses with Hartung-Knapp adjustment to consolidate results. The prediction interval (PI) and statistics were utilized to quantify between-study heterogeneity. Moreover, meta-regression and subgroup analyses were performed to investigate the potential sources of heterogeneity. This systematic review and meta-analysis is registered with PROSPERO (CRD42023428658). FINDINGS: This study encompassed 33 trials involving 27,404 patients. Those undergoing AI-aided colonoscopy experienced a significant decrease in PMR (RR, 0.475; 95% CI, 0.294-0.768; = 87.49%) and AMR (RR, 0.495; 95% CI, 0.390-0.627; = 48.76%). Additionally, a significant increase in PDR (RR, 1.238; 95% CI, 1.158-1.323; = 81.67%) and ADR (RR, 1.242; 95% CI, 1.159-1.332; = 78.87%), along with a significant increase in the rates of PPC (IRR, 1.388; 95% CI, 1.270-1.517; = 91.99%) and APC (IRR, 1.390; 95% CI, 1.277-1.513; = 86.24%), was observed. This resulted in 0.271 more PPCs (95% CI, 0.144-0.259; = 65.61%) and 0.202 more APCs (95% CI, 0.144-0.259; = 68.15%). INTERPRETATION: AI-aided colonoscopy significantly enhanced the detection of colorectal neoplasia detection, likely by reducing the miss rate. However, future studies should focus on evaluating the cost-effectiveness and long-term benefits of AI-aided colonoscopy in reducing cancer incidence. FUNDING: This work was supported by the Heilongjiang Provincial Natural Science Foundation of China (LH2023H096), the Postdoctoral research project in Heilongjiang Province (LBH-Z22210), the National Natural Science Foundation of China's General Program (82072640) and the Outstanding Youth Project of Heilongjiang Natural Science Foundation (YQ2021H023).
6Endoscopist deskilling risk after exposure to artificial intelligence in colonoscopy: a multicentre, observational study.PubMed
Krzysztof Budzyń, Marcin Romańczyk, Diana Kitala, et al.
BACKGROUND: It is not known if continuous exposure to artificial intelligence (AI) changes endoscopists' behaviour when conducting colonoscopy. We assessed how endoscopists who regularly used AI performed colonoscopy when AI was not in use. METHODS: We conducted a retrospective, observational study at four endoscopy centres in Poland taking part in the ACCEPT (Artificial Intelligence in Colonoscopy for Cancer Prevention) trial. These centres introduced AI tools for polyp detection at the end of 2021, after which colonoscopies had been randomly assigned to be conducted with or without AI assistance according to the date of examination. We evaluated the quality of colonoscopy by comparing two different phases: 3 months before and 3 months after AI implementation. We included all diagnostic colonoscopies, excluding those involving intensive anticoagulant use, pregnancy, or a history of colorectal resection or inflammatory bowel disease. The primary outcome was change in adenoma detection rate (ADR) of standard, non-AI assisted colonoscopy before and after AI exposure. Multivariable logistic regression was done to identify independent factors affecting ADR. FINDINGS: Between Sept 8, 2021, and March 9, 2022, 1443 patients underwent non-AI assisted colonoscopy before (n=795) and after (n=648) the introduction of AI (median age 61 years [IQR 45-70], 847 [58·7%] female, 596 [41·3%] male). The ADR of standard colonoscopy decreased significantly from 28·4% (226 of 795) before to 22·4% (145 of 648) after exposure to AI, corresponding with an absolute difference of -6·0% (95% CI -10·5 to -1·6; p=0·0089). In multivariable logistic regression analysis, exposure to AI (odds ratio 0·69 [95% CI 0·53-0·89]), male versus female patient sex (1·78 [1·38-2·30]), and patient age ≥60 years versus <60 years (3·60 [2·74-4·72]) were the independent factors significantly associated with ADR. INTERPRETATION: Continuous exposure to AI might reduce the ADR of standard non-AI assisted colonoscopy, suggesting a negative effect on endoscopist behaviour. FUNDING: European Commission and Japan Society for the Promotion of Science.
7Impact of artificial intelligence on colorectal polyp detection.PubMed
Giulio Antonelli, Matteo Badalamenti, Cesare Hassan, et al.
Since colonoscopy and polypectomy were introduced, Colorectal Cancer (CRC) incidence and mortality decreased significantly. Although we have entered the era of quality measurement and improvement, literature shows that a considerable amount of colorectal neoplasia is still missed by colonoscopists up to 25%, leading to an high rate of interval colorectal cancer that account for nearly 10% of all diagnosed CRC. Two main reasons have been recognised: recognition failure and mucosal exposure. For this purpose, Artificial Intelligence (AI) systems have been recently developed that identify a "hot" area during the endoscopic examination. In retrospective studies, where the systems are tested with a batch of unknown images, deep learning systems have shown very good performances, with high levels of accuracy. Of course, this setting may not reflect actual clinical practice where different pitfalls can occur, like suboptimal bowel preparation or poor examination technique. For this reason, a number of randomised clinical trials have recently been published where AI was tested in real time during endoscopic examinations. We present here an overview on recent literature addressing the performance of Computer Assisted Detection (CADe) of colorectal polyps in colonoscopy.
8Translation of Artificial Intelligence in Colonoscopy.PubMed
Jabed Ahmed, Ahmed El-Sayed, Rawen Kader
BACKGROUND: Artificial intelligence (AI) has progressed rapidly in gastroenterology, especially in colonoscopy, which is well positioned to benefit from AI due to the high global procedure volume and variability in quality across operators. In this review, we summarise the latest updates in the field, its current benefits, and further work required to accelerate its translation in day-to-day clinical practice. SUMMARY: Computer-aided detection systems are the most established AI system in colonoscopy, with robust evidence from randomised controlled trials showing significant improvements in adenoma detection rates. However, translation into real-world clinical practice has been less impactful, hindered by implementation challenges and lack of reimbursement pathways. Computer-aided diagnosis systems aim to support histological decision-making for diminutive polyps but have shown inconsistent benefits in clinical trials, reflecting complex human-computer interactions. Computer-aided quality systems, while in earlier stages, hold promise for standardising quality metrics. Novel applications in IBD demonstrate the potential of AI to standardise disease activity scoring and predict relapse, while therapeutic applications remain in proof-of-concept phases. KEY MESSAGES: Successful adoption of AI will depend on seamless workflow integration, better understanding of human-AI interaction, cost-effectiveness, establishing reimbursement and training pathways, clinician endorsement, and frameworks addressing fairness, accountability, and bias. The more distant future directions are likely to involve fully integrated multi-modal AI systems, personalised surveillance, and AI-assisted therapeutic interventions.
9Computer-assisted detection of colorectal polyps: a narrative review of clinical utility, ongoing limitations, and opportunities for advancement.PubMed
Madeline L D'Aquila, Samantha M Linhares, Kurt S Schultz, et al.
BACKGROUND AND OBJECTIVE: Colorectal cancer (CRC) is the second leading cause of cancer-related deaths worldwide and remains a public health challenge despite widespread screening. Colonoscopy is the gold standard for screening by enabling detection and removal of precancerous lesions, yet it is not without its limitations. Interval CRCs still occur, largely due to variability in adenoma detection rate (ADR), the primary quality indicator of colonoscopy. Artificial intelligence (AI)-powered computer-assisted polyp detection (CADe) systems have emerged as promising tools to enhance colonoscopy performance. This review synthesizes current evidence on CADe in colonoscopy, highlighting clinical efficacy, limitations, and future directions. METHODS: This review is based on a comprehensive PubMed search of articles published from database inception through July 31, 2025, related to CADe and AI in colonoscopy. Eligible studies included randomized controlled trials (RCTs), systematic and narrative reviews, meta-analyses, observational studies, case reports, guidelines, consensus conferences, and comparative studies. KEY CONTENT AND FINDINGS: Multiple RCTs and meta-analyses consistently demonstrate that the use of CADe in colonoscopy can increase ADR with minimal impact on colonoscope withdrawal time (WT). Benefits extend to both experienced and less experienced endoscopists across varied settings. However, concerns about false positive (FP) rates, automation bias, operator deskilling, system integration, generalizability, and long-term outcomes persist. Health-economic models suggest CADe may be cost-effective, though real-world cost-effectiveness and long-term outcome data remain limited. Emerging directions include integration with computer-assisted diagnosis tools (CADx), real-time histology prediction, and personalized surveillance strategies. CONCLUSIONS: CADe can improve ADR and is a promising step toward consistent, high-quality, equitable CRC prevention. However, uncertainties remain regarding generalizability, cost-effectiveness, and long-term outcomes. Continued work with validation, post-market surveillance, and integration with CADx are critical to fully realize the potential of CADe.
10Artificial intelligence and colonoscopy experience: lessons from two randomised trials.PubMed
Alessandro Repici, Marco Spadaccini, Giulio Antonelli, et al.
BACKGROUND AND AIMS: Artificial intelligence has been shown to increase adenoma detection rate (ADR) as the main surrogate outcome parameter of colonoscopy quality. To which extent this effect may be related to physician experience is not known. We performed a randomised trial with colonoscopists in their qualification period (AID-2) and compared these data with a previously published randomised trial in expert endoscopists (AID-1). METHODS: In this prospective, randomised controlled non-inferiority trial (AID-2), 10 non-expert endoscopists (<2000 colonoscopies) performed screening/surveillance/diagnostic colonoscopies in consecutive 40-80 year-old subjects using high-definition colonoscopy with or without a real-time deep-learning computer-aided detection (CADe) (GI Genius, Medtronic). The primary outcome was ADR in both groups with histology of resected lesions as reference. In a post-hoc analysis, data from this randomised controlled trial (RCT) were compared with data from the previous AID-1 RCT involving six experienced endoscopists in an otherwise similar setting. RESULTS: In 660 patients (62.3±10 years; men/women: 330/330) with equal distribution of study parameters, overall ADR was higher in the CADe than in the control group (53.3% vs 44.5%; relative risk (RR): 1.22; 95% CI: 1.04 to 1.40; p<0.01 for non-inferiority and p=0.02 for superiority). Similar increases were seen in adenoma numbers per colonoscopy and in small and distal lesions. No differences were observed with regards to detection of non-neoplastic lesions. When pooling these data with those from the AID-1 study, use of CADe (RR 1.29; 95% CI: 1.16 to 1.42) and colonoscopy indication, but not the level of examiner experience (RR 1.02; 95% CI: 0.89 to 1.16) were associated with ADR differences in a multivariate analysis. CONCLUSIONS: In less experienced examiners, CADe assistance during colonoscopy increased ADR and a number of related polyp parameters as compared with the control group. Experience appears to play a minor role as determining factor for ADR. TRIAL REGISTRATION NUMBER: NCT:04260321.
11Artificial intelligence for identification and characterization of colonic polyps.PubMed
Nasim Parsa, Michael F Byrne
Colonoscopy remains the gold standard exam for colorectal cancer screening due to its ability to detect and resect pre-cancerous lesions in the colon. However, its performance is greatly operator dependent. Studies have shown that up to one-quarter of colorectal polyps can be missed on a single colonoscopy, leading to high rates of interval colorectal cancer. In addition, the American Society for Gastrointestinal Endoscopy has proposed the "resect-and-discard" and "diagnose-and-leave" strategies for diminutive colorectal polyps to reduce the costs of unnecessary polyp resection and pathology evaluation. However, the performance of optical biopsy has been suboptimal in community practice. With recent improvements in machine-learning techniques, artificial intelligence-assisted computer-aided detection and diagnosis have been increasingly utilized by endoscopists. The application of computer-aided design on real-time colonoscopy has been shown to increase the adenoma detection rate while decreasing the withdrawal time and improve endoscopists' optical biopsy accuracy, while reducing the time to make the diagnosis. These are promising steps toward standardization and improvement of colonoscopy quality, and implementation of "resect-and-discard" and "diagnose-and-leave" strategies. Yet, issues such as real-world applications and regulatory approval need to be addressed before artificial intelligence models can be successfully implemented in clinical practice. In this review, we summarize the recent literature on the application of artificial intelligence for detection and characterization of colorectal polyps and review the limitation of existing artificial intelligence technologies and future directions for this field.
12Cost-effectiveness analysis of artificial intelligence-aided colonoscopy for adenoma detection and characterization in Spain.PubMed
Marco Bustamante-Balén, Beatriz Merino Rodríguez, Luis Barranco, et al.
BACKGROUND AND STUDY AIMS: The aim of this study was to assess the cost-effectiveness of an intelligent endoscopy module for computer-assisted detection and characterization (CADe/CADx) compared with standard practice, from a Spanish National Health System perspective. METHODS: A Markov model was designed to estimate total costs, life years gained (LYG), and quality-adjusted life years (QALYs) over a lifetime horizon with annual cycles. A hypothetical cohort of 1,000 patients eligible for colonoscopy (mean age 61.32 years) was distributed between Markov states according to polyp size, location, and histology based on national screening program data. CADe/CADx efficacy was determined based on adenoma miss rates and natural disease evolution was simulated according to annual transition probabilities. Detected polyp management involved polypectomy and histopathology in standard practice, whereas with CADe/CADx leave-in-situ strategy was applied for ≤ 5 mm rectosigmoid non-adenomas and resect-and-discard strategy for the rest of ≤ 5mm polyps. Unit costs (€,2024) included the diagnostic procedure and polyp and colorectal cancer (CRC) management. A 3% annual discount rate was applied to costs and outcomes. Model inputs were validated by an expert panel. RESULTS: CADe/CADx was more effective (16.37 LYG and 14.32 QALYs) than standard practice (16.33 LYG and 14.27 QALYs) over a lifetime horizon. Total cost per patient was €2,300.76 with CADe/CADx and €2,508.75 with colonoscopy alone. In a hypothetical cohort of 1,000 patients, CADe/CADx avoided 173 polypectomies, 370 histopathologies, and 7 CRC cases. Sensitivity analyses confirmed model robustness. CONCLUSIONS: The results of this analysis suggest that CADe/CADx would result in a dominant strategy versus standard practice in patients undergoing colonoscopy in Spain.
13Implementation of Artificial Intelligence in Colonoscopy Practice in Japan.PubMed
Masashi Misawa, Shin-Ei Kudo, Yuichi Mori
This review outlines the implementation of artificial intelligence (AI) into colonoscopy procedures which includes its history, processes, and challenges. We highlight the importance of the collaborative effort between medical and computer science researchers in the development of AI tools in colonoscopy, particularly focusing on the roles of computer-aided detection (CADe) and computer-aided characterization (CADx) in a real time analysis of colonoscopy videos. Some of the proposed technologies are considered to improve the important clinical outcomes of patients such as adenoma detection rate in colonoscopy. Regulatory approval is considered mandatory before introducing AI tools into the market owing to the potential risks associated with the introduction of AI tools in healthcare. We share the experience of obtaining regulatory approval for EndoBRAIN in Japan, emphasizing the challenges in establishing examination criteria and performance levels at the period. Reimbursement is also identified as necessary for the widespread adoption of medical innovation. With the introduction of reimbursement for a CADe tool in Japan in 2024, we expect to accelerate implementation of AI in colonoscopy in general. Despite regulatory approval and reimbursement, concerns are raised with regard to the assessment of the balance between benefits and harms of AI in colonoscopy. Questions about its impact on cancer prevention, healthcare burden, patient acceptance, and effectiveness across different populations remain unsolved. The lack of clinical guidelines for AI in colonoscopy emphasizes the need for a rigorous assessment of available evidence in optimizing the adoption of AI in colonoscopy practice. While it is always exciting to strive for medical innovation, ensuring rigorous evaluation to optimize patient care is mandatory to improve the quality of health and society.
14Real-Time Computer-Aided Detection of Colorectal Neoplasia During Colonoscopy : A Systematic Review and Meta-analysis.PubMed
Cesare Hassan, Marco Spadaccini, Yuichi Mori, et al.
BACKGROUND: Artificial intelligence computer-aided detection (CADe) of colorectal neoplasia during colonoscopy may increase adenoma detection rates (ADRs) and reduce adenoma miss rates, but it may increase overdiagnosis and overtreatment of nonneoplastic polyps. PURPOSE: To quantify the benefits and harms of CADe in randomized trials. DESIGN: Systematic review and meta-analysis. (PROSPERO: CRD42022293181). DATA SOURCES: Medline, Embase, and Scopus databases through February 2023. STUDY SELECTION: Randomized trials comparing CADe-assisted with standard colonoscopy for polyp and cancer detection. DATA EXTRACTION: Adenoma detection rate (proportion of patients with ≥1 adenoma), number of adenomas detected per colonoscopy, advanced adenoma (≥10 mm with high-grade dysplasia and villous histology), number of serrated lesions per colonoscopy, and adenoma miss rate were extracted as benefit outcomes. Number of polypectomies for nonneoplastic lesions and withdrawal time were extracted as harm outcomes. For each outcome, studies were pooled using a random-effects model. Certainty of evidence was assessed using the GRADE (Grading of Recommendations Assessment, Development and Evaluation) framework. DATA SYNTHESIS: Twenty-one randomized trials on 18 232 patients were included. The ADR was higher in the CADe group than in the standard colonoscopy group (44.0% vs. 35.9%; relative risk, 1.24 [95% CI, 1.16 to 1.33]; low-certainty evidence), corresponding to a 55% (risk ratio, 0.45 [CI, 0.35 to 0.58]) relative reduction in miss rate (moderate-certainty evidence). More nonneoplastic polyps were removed in the CADe than the standard group (0.52 vs. 0.34 per colonoscopy; mean difference [MD], 0.18 polypectomy [CI, 0.11 to 0.26 polypectomy]; low-certainty evidence). Mean inspection time increased only marginally with CADe (MD, 0.47 minute [CI, 0.23 to 0.72 minute]; moderate-certainty evidence). LIMITATIONS: This review focused on surrogates of patient-important outcomes. Most patients, however, may consider cancer incidence and cancer-related mortality important outcomes. The effect of CADe on such patient-important outcomes remains unclear. CONCLUSION: The use of CADe for polyp detection during colonoscopy results in increased detection of adenomas but not advanced adenomas and in higher rates of unnecessary removal of nonneoplastic polyps. PRIMARY FUNDING SOURCE: European Commission Horizon 2020 Marie Skłodowska-Curie Individual Fellowship.
15Can Real-time Computer-Aided Detection Systems Diminish the Risk of Postcolonoscopy Colorectal Cancer?PubMed
Mariusz Madalinski, Roger Prudham
The adenoma detection rate is the constant subject of research and the main marker of quality in bowel cancer screening. However, by improving the quality of endoscopy via artificial intelligence methods, all polyps, including those with the potential for malignancy, can be removed, thereby reducing interval colorectal cancer rates. As such, the removal of all polyps may become the best marker of endoscopy quality. Thus, we present a viewpoint on integrating the computer-aided detection (CADe) of polyps with high-accuracy, real-time colonoscopy to challenge quality improvements in the performance of colonoscopy. Colonoscopy for bowel cancer screening involving the integration of a deep learning methodology (ie, integrating artificial intelligence with CADe systems) has been assessed in an effort to increase the adenoma detection rate. In this viewpoint, a few studies are described, and their results show that CADe systems are able to increase screening sensitivity. The detection of adenomatous polyps, which are associated with a potential risk of progression to colorectal cancer, and their removal are expected to reduce cancer incidence and mortality rates. However, so far, artificial intelligence methods do not increase the detection of cancer or large adenomatous polyps but contribute to the detection of small precancerous polyps.
16Cost-effectiveness Analysis of Colorectal Cancer Screening Strategies Using Active Learning and Monte Carlo Simulation.PubMed
Amirhossein Fouladi, Amin Asadi, Eric A Sherer, et al.
INTRODUCTION: Detection of colorectal cancer (CRC) in the early stages through available screening tests increases the patient's survival chances. Multimodal screening policies can benefit patients by providing more diverse screening options and balancing the risks and benefits of screening tests. We investigate the cost-effectiveness of a wide variety of multimodal CRC screening policies. METHODS: We developed a Monte Carlo simulation framework to model CRC dynamics. We proposed an innovative calibration process using machine learning models to estimate age- and size-specific adenomatous polyps' progression and regression rates. The proposed approach significantly expedites the model parameter space search. RESULTS: Two multimodal proposed policies (i.e., 1] colonoscopy at 50 y and fecal occult blood test annually between 60 and 75 y and 2] colonoscopy at 50 and 60 y and fecal immunochemical test annually between 70 and 75 y) are identified as efficient frontier policies. Both policies are cost-effective at a willingness to pay of $50,000. Sensitivity analyses were performed to assess the sensitivity of results to a change in screening test costs as well as adherence behavior. The sensitivity analysis results suggest that the proposed policies are mostly robust to the considered changes in screening test costs, as there is a significant overlap between the efficient frontier policies of the baseline and the sensitivity analysis cases. However, the efficient frontier policies were more sensitive to changes in adherence behavior. CONCLUSION: Generally, combining stool-based tests with visual tests will benefit patients with higher life expectancy and a lower expected cost compared with unimodal screening policies. Colonoscopy at younger ages (when the colonoscopy complication risk is lower) and stool-based tests at older ages are shown to be more effective. HIGHLIGHTS: We propose a detailed Markov model to capture the colorectal cancer (CRC) dynamics. The proposed Markov model presents the detailed dynamics of adenomas progression to CRC.We use more than 44,000 colonoscopy reports and available data in the literature to calibrate the proposed Markov model using an innovative approach that leverages machine learning models to expedite the calibration process.We investigate the cost-effectiveness of a wide variety of multimodal CRC screening policies and compare their performances with the current in-practice policies.
17Benefits and challenges in implementation of artificial intelligence in colonoscopy: World Endoscopy Organization position statement.PubMed
Yuichi Mori, James E East, Cesare Hassan, et al.
The number of artificial intelligence (AI) tools for colonoscopy on the market is increasing with supporting clinical evidence. Nevertheless, their implementation is not going smoothly for a variety of reasons, including lack of data on clinical benefits and cost-effectiveness, lack of trustworthy guidelines, uncertain indications, and cost for implementation. To address this issue and better guide practitioners, the World Endoscopy Organization (WEO) has provided its perspective about the status of AI in colonoscopy as the position statement. WEO Position Statement: Statement 1.1: Computer-aided detection (CADe) for colorectal polyps is likely to improve colonoscopy effectiveness by reducing adenoma miss rates and thus increase adenoma detection; Statement 1.2: In the short term, use of CADe is likely to increase health-care costs by detecting more adenomas; Statement 1.3: In the long term, the increased cost by CADe could be balanced by savings in costs related to cancer treatment (surgery, chemotherapy, palliative care) due to CADe-related cancer prevention; Statement 1.4: Health-care delivery systems and authorities should evaluate the cost-effectiveness of CADe to support its use in clinical practice; Statement 2.1: Computer-aided diagnosis (CADx) for diminutive polyps (≤5 mm), when it has sufficient accuracy, is expected to reduce health-care costs by reducing polypectomies, pathological examinations, or both; Statement 2.2: Health-care delivery systems and authorities should evaluate the cost-effectiveness of CADx to support its use in clinical practice; Statement 3: We recommend that a broad range of high-quality cost-effectiveness research should be undertaken to understand whether AI implementation benefits populations and societies in different health-care systems.
18British Society of Gastroenterology guidelines on colorectal surveillance in inflammatory bowel disease.PubMed
James Edward East, Morris Gordon, Gaurav Bhaskar Nigam, et al.
Patients with inflammatory bowel disease (IBD) remain at increased risk for colorectal cancer and death from colorectal cancer compared with the general population despite improvements in inflammation control with advanced therapies, colonoscopic surveillance and reductions in environmental risk factors. This guideline update from 2010 for colorectal surveillance of patients over 16 years with colonic inflammatory bowel disease was developed by stakeholders representing UK physicians, endoscopists, surgeons, specialist nurses and patients with GRADE (Grading of Recommendations Assessment, Development and Evaluation) methodological support.An a priori protocol was published describing the approach to three levels of statement: GRADE recommendations, good practice statements or expert opinion statements. A systematic review of 7599 publications, with appraisal and GRADE analysis of trials and network meta-analysis, where appropriate, was performed. Risk thresholding guided GRADE judgements.We made 73 statements for the delivery of an IBD colorectal surveillance service, including outcome standards for service and endoscopist audit, and the importance of shared decision-making with patients.Core areas include: risk of colorectal cancer, IBD-related post-colonoscopy colorectal cancer; service organisation and supporting patient concordance; starting and stopping surveillance, who should or should not receive surveillance; risk stratification, including web-based multivariate risk calculation of surveillance intervals; colonoscopic modalities, bowel preparation, biomarkers and artificial intelligence aided detection; chemoprevention; the role of non-conventional dysplasia, serrated lesions and non-targeted biopsies; management of dysplasia, both endoscopic and surgical, and the structure and role of the multidisciplinary team in IBD dysplasia management; training in IBD colonoscopic surveillance, sustainability (green endoscopy), cost-effectiveness and patient experience. Sixteen research priorities are suggested.
19Artificial Intelligence-Assisted Colonoscopy for Colorectal Cancer Screening: A Multicenter Randomized Controlled Trial.PubMed
Hong Xu, Raymond S Y Tang, Thomas Y T Lam, et al.
BACKGROUND AND AIMS: Artificial intelligence (AI)-assisted colonoscopy improves polyp detection and characterization in colonoscopy. However, data from large-scale multicenter randomized controlled trials (RCT) in an asymptomatic population are lacking. METHODS: This multicenter RCT aimed to compare AI-assisted colonoscopy with conventional colonoscopy for adenoma detection in an asymptomatic population. Asymptomatic subjects 45-75 years of age undergoing colorectal cancer screening by direct colonoscopy or fecal immunochemical test were recruited in 6 referral centers in Hong Kong, Jilin, Inner Mongolia, Xiamen, and Beijing. In the AI-assisted colonoscopy, an AI polyp detection system (Eagle-Eye) with real-time notification on the same monitor of the endoscopy system was used. The primary outcome was overall adenoma detection rate (ADR). Secondary outcomes were mean number of adenomas per colonoscopy, ADR according to endoscopist's experience, and colonoscopy withdrawal time. This study received Institutional Review Board approval (CRE-2019.393). RESULTS: From November 2019 to August 2021, 3059 subjects were randomized to AI-assisted colonoscopy (n = 1519) and conventional colonoscopy (n = 1540). Baseline characteristics and bowel preparation quality between the 2 groups were similar. The overall ADR (39.9% vs 32.4%; P < .001), advanced ADR (6.6% vs 4.9%; P = .041), ADR of expert (42.3% vs 32.8%; P < .001) and nonexpert endoscopists (37.5% vs 32.1%; P = .023), and adenomas per colonoscopy (0.59 ± 0.97 vs 0.45 ± 0.81; P < .001) were all significantly higher in the AI-assisted colonoscopy. The median withdrawal time (8.3 minutes vs 7.8 minutes; P = .004) was slightly longer in the AI-assisted colonoscopy group. CONCLUSIONS: In this multicenter RCT in asymptomatic patients, AI-assisted colonoscopy improved overall ADR, advanced ADR, and ADR of both expert and nonexpert attending endoscopists. (ClinicalTrials.gov, Number: NCT04422548).
20The Cost-Effectiveness of AI-Assisted Colonoscopy as a Primary or Secondary Screening Test in a Population-Based Colorectal Cancer Screening Program: Markov Modeling-Based Cost Effectiveness Analysis.PubMed
Martin Cs Wong, Junjie Huang, Thomas Yt Lam, et al.
BACKGROUND: Colorectal cancer (CRC) is the third most common cancer worldwide and poses a heavy burden on health care systems. Early screening for CRC through colonoscopy can effectively reduce both the incidence and mortality associated with CRC. However, the sensitivity of conventional colonoscopy is limited by the level of experience of physicians. Recently, artificial intelligence (AI)-assisted colonoscopy has been shown to have higher sensitivity in detecting CRC and mitigating the limitations concerning physician experience, but few studies have evaluated the cost-effectiveness of AI-assisted colonoscopy in CRC screening. OBJECTIVE: This study aimed to evaluate the cost-effectiveness of various CRC screening strategies, including no screening, fecal immunochemical test (FIT) positive result followed by a conventional colonoscopy, FIT positive result followed by AI-assisted colonoscopy, direct colonoscopy, and direct AI-assisted colonoscopy. METHODS: This study modeled a hypothetical population based on current clinical practice in Asia, where CRC screening typically begins at the age of 50 years. The cost-effectiveness of various population-based CRC screening strategies, including AI-assisted colonoscopy, was evaluated by comparing incremental cost-effectiveness ratios (ICERs) and outcome measures such as cancer-related life years lost, number of CRC cases prevented, life years saved, and total cost per life year saved. Data from the international literature and the government gazette were accessed to calculate relevant cost and performance estimates. The data were entered into a decision analysis algorithm based on a Markov model. RESULTS: Compared to no screening strategy, the ICERs of FIT+colonoscopy (FIT followed by conventional colonoscopy if the FIT result is positive), FIT+AI-assisted colonoscopy (FIT followed by AI-assisted colonoscopy if the FIT result is positive), colonoscopy alone, and AI-assisted colonoscopy were US $138,539, US $122,539, US $203,929, and US $180,444, respectively. When compared with FIT+colonoscopy, the FIT+AI-assisted colonoscopy strategy resulted in fewer cancer-related life years lost (5355 y vs 5327 y), a higher number and proportion of CRC cases prevented (120 vs 132 and 3.7% vs 4.1%), more life years saved (280 y vs 308 y), and lower total cost per life year saved (US $944,008 vs US $854,367). FIT+AI-assisted colonoscopy, which had the lowest ICER (US $122,539) dominated all other strategies, particularly compared to FIT+colonoscopy, with an ICER of -US $36,462. Among primary screening methods, AI-assisted colonoscopy dominated conventional colonoscopy (ICER -US $39,040). CONCLUSIONS: For an Asian population, FIT followed by AI-assisted colonoscopy represented the most cost-effective CRC screening strategy. It had the lowest ICER and the lowest additional cost among all 4 evaluated strategies.