The yield of artificial intelligence (GI genius) in Lynch syndrome -A randomized tandem-colonoscopy trial.
Аннотация
BACKGROUND AND AIMS: Artificial intelligence (AI)- assisted colonoscopy has been shown to increase the adenoma-detection rate in the general population but there is a paucity of data on its benefit in Lynch syndrome. We aimed to investigate the incremental detection rate of polyps using AI- assisted colonoscopy compared with high-definition white-light endoscopy (HD-WLE).
METHODS
We conducted a single-center randomized tandem colonoscopy trial of patients with Lynch syndrome. Patients underwent a first colonoscopy with HD-WLE, and were randomized at 1:1 ratio to undergo a second colonoscopy with either HD-WLE or an AI- assisted device. All visualized polyps were removed and size, histology and numbers of polyps detected on each exam were recorded.
RESULTS
Of 107 patients screened, 100 were enrolled (median age 45.5 years, 39% males). On a per polyp analysis, 12 adenomas were detected at the first colonoscopy and 6 at the second one in the AI group, corresponding to incremental detection rate (IDR) of 33.3%, whereas in the HD-WLE group, 16 adenomas were detected at first colonoscopy and 2 at second, corresponding to IDR of 11.1% [OR 4.0, 95% CI (0.68-23.40); p value- 0.124]. On per patient analysis, the increment in adenoma detection rate and polyp detection rate at second colonoscopy were also not significantly increased with AI compared with HD-WLE (35.3%vs. 13.3%, p - 0.165; and 26.6% vs. 14.8%, p - 0.278; respectively).
CONCLUSIONS
No significant difference was displayed by artificial intelligence in detecting adenomas in Lynch-syndrome patients in the setting of high-definition colonoscopy with prolonged withdrawal time.
Подготовка и перевод публикации
Обработка статьи · этапы, ошибки и управление
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Тип публикации указан по данным источника. Уровень доказательности автоматически не присваивается.