Effect of a Computer-Aided Device for Detecting Gastric Neoplasms: A Multicenter, Randomized Controlled Trial.

We aimed to explore whether AI can enhance gastric neoplasm detection.

我们的目标是探讨人工智能是否能提高胃肿瘤的检测率。

Methods

Participants from 24 hospitals in China from December 21, 2021, to November 11, 2023, were randomized to AI-assisted or nonassisted esophagogastroduodenoscopy.

2021年12月21日至2023年11月11日，来自中国24家医院的参与者被随机分配到人工智能辅助或非辅助的食管胃十二指肠镜检查组。

Primary outcome was detection rate of gastric neoplasms after pathologic review.

主要结果是经过病理复审后的胃肿瘤检出率。

Secondary outcomes included detection rate of gastric neoplasms before review, relative early gastric cancer detection ratio, detection rate of intestinal metaplasia and/or gastric atrophy before or after review, biopsy rate, number of blind spots, and procedure/inspection time.

次要结果包括审查前胃肿瘤的检出率、相对早期胃癌检出比例、审查前后肠上皮化生和/或胃萎缩的检出率、活检率、盲点数量以及程序/检查时间。

We did intention-to-treat(ITT), per-protocol, and exploratory subgroup analysis.

我们进行了意向治疗(ITT)、按方案(PP)和探索性亚组分析。

Results

In the ITT cohort, 29,514 patients were enrolled.

在意向治疗(ITT)队列中，共有29,514名患者被纳入。

AI did not improve detection rate of gastric neoplasm after pathological review (RR, 1.13; 0.92-1.38; 1.42 vs 1.25%; P = .25).

在病理复审后，人工智能并未提高胃肿瘤的检出率（相对风险，1.13；95%置信区间0.92-1.38；1.42%对比1.25%；P = .25）。

However, based on original pathology, an improvement with AI was observed (RR, 1.14; 1.0-1.28; 4.06 vs 3.57%; P = .03).

然而，基于原始病理学，观察到人工智能（AI）的改善（相对风险，1.14；95%置信区间1.0-1.28；4.06%对比3.57%；P = .03）。

AI reduced blind spots number from 2.52 to 1.07 (P < .001) and prolonged procedure and inspection time.

人工智能（AI）将盲点数量从2.52减少到1.07（P < .001），并延长了手术和检查时间。

No significant differences were observed for relative early gastric cancer detection ratio or detection rate of intestinal metaplasia and/or gastric atrophy before/after pathologic review in ITT cohort.

在ITT队列中，无论是在病理复审前还是病理复审后，相对早期胃癌检出率或肠上皮化生和/或胃萎缩的检出率均未观察到显著差异。

Subgroup analysis suggested potential benefit among less experienced endoscopists and during fatigue periods.

亚组分析表明，在经验较少的内镜医师和疲劳期间可能存在潜在的益处。

In the experimental group, AI diagnosed 100%, 91.9%, and 57.1% of pathologically confirmed gastric adenocarcinoma, high-grade, and low-grade intraepithelial neoplasia, respectively.

在实验组中，人工智能诊断出的病理证实的胃腺癌、高级别和低级别上皮内瘤变的检出率分别为100%、91.9%和57.1%。

Conclusions

AI did not improve the detection rate of gastric neoplasms.

人工智能并未提高胃肿瘤的检出率。

Further real-world studies are needed to fully address the adaptability of AI.

需要进一步的现实世界研究来全面解决人工智能的适应性问题。

(Chinese Clinical Trial Registry, ChiCTR2100054449.).

（中国临床试验注册中心，ChiCTR2100054449）。