Reducing adenoma miss rate of colonoscopy assisted by artificial intelligence: a multicenter randomized controlled trial

Kamba, Shunsuke; Tamai, Naoto; Saitoh, Iduru; Matsui, Hiroaki; Horiuchi, Hiroki; Kobayashi, Mitsuo; Sakamoto, Taku; Ego, Mai; Fukuda, Akira; Tonouchi, Aya; Shimahara, Yuki; Nishikawa, Masako; Nishino, Hitoo; Saito, Yutaka; Sumiyama, Kazuki

doi:10.1007/s00535-021-01808-w

Cited by 61 publications

(89 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We believe that such validation methods would be more preferable in a developmental phase to decide the appropriate direction of research and also to reveal the strengths and weaknesses of CADe. In fact, a prospective multi-center randomized trial evaluating AMR as the primary outcome in a tandem study design, in which the CADe-assisted group achieved better AMR than the non-assisted group, was warranted from the desirable results of this retrospective trial [23].…”

Section: Discussionmentioning

confidence: 99%

“…Meanwhile, the advantage of assistance from a DL-based CADe system over non-assisted human perception has already been demonstrated in several randomized controlled trials (RCTs), including our study evaluating ADR or the adenoma miss rate (AMR) as the primary outcome. The result was confirmed in a series of meta-analyses, some of which analyzed the outcomes of RCTs [22][23][24][25]. However, RCT demands a significant amount of effort and time; retesting each updated DL-based algorithm using RCT may be suboptimal as a reference to set the direction of algorithmic developments.…”

Section: Introductionmentioning

confidence: 87%

See 1 more Smart Citation

Detection Accuracy and Latency of Colorectal Lesions with Computer-Aided Detection System Based on Low-Bias Evaluation

et al. 2021

Self Cite

View full text Add to dashboard Cite

We developed a computer-aided detection (CADe) system to detect and localize colorectal lesions by modifying You-Only-Look-Once version 3 (YOLO v3) and evaluated its performance in two different settings. The test dataset was obtained from 20 randomly selected patients who underwent endoscopic resection for 69 colorectal lesions at the Jikei University Hospital between June 2017 and February 2018. First, we evaluated the diagnostic performances using still images randomly and automatically extracted from video recordings of the entire endoscopic procedure at intervals of 5 s, without eliminating poor quality images. Second, the latency of lesion detection by the CADe system from the initial appearance of lesions was investigated by reviewing the videos. A total of 6531 images, including 662 images with a lesion, were studied in the image-based analysis. The AUC, sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were 0.983, 94.6%, 95.2%, 68.8%, 99.4%, and 95.1%, respectively. The median time for detecting colorectal lesions measured in the lesion-based analysis was 0.67 s. In conclusion, we proved that the originally developed CADe system based on YOLO v3 could accurately and instantaneously detect colorectal lesions using the test dataset obtained from videos, mitigating operator selection biases.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 87%

Detection Accuracy and Latency of Colorectal Lesions with Computer-Aided Detection System Based on Low-Bias Evaluation

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…In the last couple of years, automated polyp detection has been studied in several prospective trials looking primarily at AMR and ADR. Three relatively new tandem randomized controlled trials (RCTs) published in 2020 and 2021 demonstrated that AMR was significantly decreased in the CADe group compared to the standard colonoscopy group [13][14][15] (e.g., 13,8% vs. 36,7% in a Japanese multicenter trial and 20.12% vs. 31.25% in a US multicenter trial). Additionally, an increasing number of prospective trials studying the effect of AI on ADR have been published, mainly from China [16][17][18][19][20][21][22][23] but also from Italy conducted in a multi-center fashion.…”

Section: Cade and Computer-aided Quality Improvementmentioning

confidence: 99%

Artificial intelligence in colonoscopy: A review on the current status

Larsen

Mori

2022

DEN Open

View full text Add to dashboard Cite

Artificial intelligence has become an increasingly hot topic in the last several years, and it has also gained its way into the medical field. In recent years, the application of artificial intelligence in the gastroenterology field has been of increasing interest, particularly in the colonoscopy setting. Novel technologies such as deep neural networks have enabled real‐time computer‐aided polyp detection and diagnosis during colonoscopy. This might lead to increased performance of endoscopists as well as potentially reducing the costs of unnecessary polypectomies of hyperplastic polyps. Newly published prospective trials studying computer‐aided detection showed that the assistance of artificial intelligence significantly increased the detection of polyps and non‐advanced adenomas approximately by 10%, while three tandem randomized control trials proved that the adenoma miss rate was significantly reduced (e.g., 13.8% vs. 36.7% in one Japanese multicenter trial). Promising results have also been shown in prospective single‐arm trials on computer‐aided polyp diagnosis, but the evidence is insufficient to reach a conclusion.

show abstract

“…Recently published tandem randomized controlled trial in Japan showed that two LST-NG-type advanced adenomas (13.3% of all the advanced adenomas) were missed with standard colonoscopy but immediately picked up in the following AI-assisted colonoscopy. 7 It would be interesting to see how more accumulated data in randomized controlled trials and meta-analyses will change our mind towards the use of AI in terms of detection of subtle, elusive, but advanced lesions. Actually, another meta-analysis indicated that the use of AI increased the mean number of advanced adenomas per colonoscopy.…”

mentioning

confidence: 99%