Kwon Yeong Seok, Park Tae Yong, Kim So Eui, Park Yehyun, Lee Jae Gon, Lee Sang Pyo, Kim Kyeong Ok, Jang Hyun Joo, Yang Young Joo, Cho Bum-Joo
Department of Internal Medicine, Hallym University College of Medicine, Chuncheon-si 24253, South Korea.
Medical Artificial Intelligence Center, Hallym University Medical Center, Anyang 14068, South Korea.
World J Gastroenterol. 2025 Jul 21;31(27):106819. doi: 10.3748/wjg.v31.i27.106819.
Small-bowel capsule endoscopy (SBCE) is widely used to evaluate obscure gastrointestinal bleeding; however, its interpretation is time-consuming and reader-dependent. Although artificial intelligence (AI) has emerged to address these limitations, few models simultaneously perform small-bowel (SB) localization and abnormality detection.
To develop an AI model that automatically distinguishes the SB from the stomach and colon and diagnoses SB abnormalities.
We developed an AI model using 87005 CE images (11925, 33781, and 41299 from the stomach, SB, and colon, respectively) for SB localization and 28405 SBCE images (1337 erosions/ulcers, 126 angiodysplasia, 494 bleeding, and 26448 normal) for abnormality detection. The diagnostic performances of AI-assisted reading and conventional reading were compared using 32 SBCE videos in patients with suspicious SB bleeding.
Regarding organ localization, the AI model achieved an area under the receiver operating characteristic curve (AUC) and accuracy exceeding 0.99 and 97%, respectively. For SB abnormality detection, the performance was as follows: Erosion/ulcer: 99.4% accuracy (AUC, 0.98); angiodysplasia: 99.8% accuracy (AUC, 0.99); bleeding: 99.9% accuracy (AUC, 0.99); normal: 99.3% accuracy (AUC, 0.98). In external validation, AI-assisted reading (8.7 minutes) was significantly faster than conventional reading (53.9 minutes; < 0.001). The SB localization accuracies (88.6% 72.7%, = 0.07) and SB abnormality detection rates (77.3% 77.3%, = 1.00) of the conventional reading and AI-assisted reading were comparable.
Our AI model decreased SBCE reading time and achieved performance comparable to that of experienced endoscopists, suggesting that AI integration into SBCE reading enables efficient and reliable SB abnormality detection.
小肠胶囊内镜检查（SBCE）被广泛用于评估不明原因的胃肠道出血；然而，其解读耗时且依赖阅片者。尽管人工智能（AI）已出现以解决这些局限性，但很少有模型能同时进行小肠（SB）定位和异常检测。
开发一种人工智能模型，能自动区分小肠与胃和结肠，并诊断小肠异常。
我们使用87005张胶囊内镜图像（分别来自胃、小肠和结肠的11925张、33781张和41299张）开发了一个用于小肠定位的人工智能模型，并使用28405张小肠胶囊内镜图像（1337处糜烂/溃疡、126处血管发育异常、494处出血和26448处正常）进行异常检测。使用32例疑似小肠出血患者的小肠胶囊内镜视频比较了人工智能辅助阅片和传统阅片的诊断性能。
在器官定位方面，人工智能模型的受试者操作特征曲线下面积（AUC）和准确率分别超过0.99和97%。对于小肠异常检测，性能如下：糜烂/溃疡：准确率99.4%（AUC，0.98）；血管发育异常：准确率99.8%（AUC，0.99）；出血：准确率99.9%（AUC，0.99）；正常：准确率99.3%（AUC，0.98）。在外部验证中，人工智能辅助阅片（8.7分钟）明显快于传统阅片（53.9分钟；<0.001）。传统阅片和人工智能辅助阅片的小肠定位准确率（88.6%对72.7%，P = 0.07）和小肠异常检测率（77.3%对77.3%，P = 1.00）相当。
我们的人工智能模型缩短了小肠胶囊内镜的阅片时间，且性能与经验丰富的内镜医师相当，这表明将人工智能整合到小肠胶囊内镜阅片中可实现高效且可靠的小肠异常检测。
