Evaluating false‐positive detection in a computer‐aided detection system for colonoscopy

Background and Aim Computer‐aided detection (CADe) systems can efficiently detect polyps during colonoscopy. However, false‐positive (FP) activation is a major limitation of CADe. We aimed to compare the rate and causes of FP using CADe before and after an update designed to reduce FP. Methods We an...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Journal of Gastroenterology and Hepatology Ročník 39; číslo 5; s. 927 - 934
Hlavní autoři:	Okumura, Taishi, Imai, Kenichiro, Misawa, Masashi, Kudo, Shin‐ei, Hotta, Kinichi, Ito, Sayo, Kishida, Yoshihiro, Takada, Kazunori, Kawata, Noboru, Maeda, Yuki, Yoshida, Masao, Yamamoto, Yoichi, Minamide, Tatsunori, Ishiwatari, Hirotoshi, Sato, Junya, Matsubayashi, Hiroyuki, Ono, Hiroyuki
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	Australia Wiley 01.05.2024 Wiley Subscription Services, Inc
Témata:	Adenoma Artificial intelligence Colon Colonoscopy Colorectal cancer Computer‐aided detection Polyps Adenoma Artificial intelligence Computer-aided detection Colonoscopy Colorectal cancer
ISSN:	0815-9319, 1440-1746, 1440-1746
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Popis
Shrnutí:	Background and Aim Computer‐aided detection (CADe) systems can efficiently detect polyps during colonoscopy. However, false‐positive (FP) activation is a major limitation of CADe. We aimed to compare the rate and causes of FP using CADe before and after an update designed to reduce FP. Methods We analyzed CADe‐assisted colonoscopy videos recorded between July 2022 and October 2022. The number and causes of FPs and excessive time spent by the endoscopist on FP (ET) were compared pre‐ and post‐update using 1:1 propensity score matching. Results During the study period, 191 colonoscopy videos (94 and 97 in the pre‐ and post‐update groups, respectively) were recorded. Propensity score matching resulted in 146 videos (73 in each group). The mean number of FPs and median ET per colonoscopy were significantly lower in the post‐update group than those in the pre‐update group (4.2 ± 3.7 vs 18.1 ± 11.1; P < 0.001 and 0 vs 16 s; P < 0.001, respectively). Mucosal tags, bubbles, and folds had the strongest association with decreased FP post‐update (pre‐update vs post‐update: 4.3 ± 3.6 vs 0.4 ± 0.8, 0.32 ± 0.70 vs 0.04 ± 0.20, and 8.6 ± 6.7 vs 1.6 ± 1.7, respectively). There was no significant decrease in the true positive rate (post‐update vs pre‐update: 95.0% vs 99.2%; P = 0.09) or the adenoma detection rate (post‐update vs pre‐update: 52.1% vs 49.3%; P = 0.87). Conclusions The updated CADe can reduce FP without impairing polyp detection. A reduction in FP may help relieve the burden on endoscopists.
Bibliografie:	The ethics committee of our hospital approved the study protocol (J‐2022‐159‐2022‐1‐3). Informed consent The opt‐out method was used to obtain consent from study participants. Ethical approval Declaration of conflict of interest M.M. received speaking honoraria from Olympus Corporation and has ownership interests in the products of Cybernet Systems. The other authors declare no conflict of interest for this study. ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 content type line 23
ISSN:	0815-9319 1440-1746 1440-1746
DOI:	10.1111/jgh.16491