一种利用内镜超声区分胃肠道间质瘤和平滑肌瘤的人工智能系统。
An artificial intelligence system for distinguishing between gastrointestinal stromal tumors and leiomyomas using endoscopic ultrasonography.
机构信息
Department of Pediatric Surgery, The Affiliated Hospital of Qingdao University, Qingdao, China.
Shandong Key Laboratory of Digital Medicine and Computer Assisted Surgery, The Affiliated Hospital of Qingdao University, Qingdao, China.
出版信息
Endoscopy. 2022 Mar;54(3):251-261. doi: 10.1055/a-1476-8931. Epub 2022 Jun 9.
BACKGROUND
Gastrointestinal stromal tumors (GISTs) and gastrointestinal leiomyomas (GILs) are the most common subepithelial lesions (SELs). All GISTs have malignant potential; however, GILs are considered benign. Current imaging cannot effectively distinguish GISTs from GILs. We aimed to develop an artificial intelligence (AI) system to differentiate these tumors using endoscopic ultrasonography (EUS).
METHODS
The AI system was based on EUS images of patients with histologically confirmed GISTs or GILs. Participants from four centers were collected to develop and retrospectively evaluate the AI-based system. The system was used when endosonographers considered SELs to be GISTs or GILs. It was then used in a multicenter prospective diagnostic test to clinically explore whether joint diagnoses by endosonographers and the AI system can distinguish between GISTs and GILs to improve the total diagnostic accuracy for SELs.
RESULTS
The AI system was developed using 10 439 EUS images from 752 participants with GISTs or GILs. In the prospective test, 132 participants were histologically diagnosed (36 GISTs, 44 GILs, and 52 other types of SELs) among 508 consecutive subjects. Through joint diagnoses, the total accuracy of endosonographers in diagnosing the 132 histologically confirmed participants increased from 69.7 % (95 % confidence interval [CI] 61.4 %-76.9 %) to 78.8 % (95 %CI 71.0 %-84.9 %; = 0.01). The accuracy of endosonographers in diagnosing the 80 participants with GISTs or GILs increased from 73.8 % (95 %CI 63.1 %-82.2 %) to 88.8 % (95 %CI 79.8 %-94.2 %; = 0.01).
CONCLUSIONS
We developed an AI-based EUS diagnostic system that can effectively distinguish GISTs from GILs and improve the diagnostic accuracy of SELs.
背景
胃肠道间质瘤（GISTs）和胃肠道平滑肌瘤（GILs）是最常见的黏膜下病变（SELs）。所有 GIST 均具有恶性潜能；然而，GIL 被认为是良性的。目前的影像学检查无法有效区分 GIST 与 GIL。我们旨在开发一种人工智能（AI）系统，通过内镜超声（EUS）来区分这些肿瘤。
方法
该 AI 系统基于经组织学证实的 GIST 或 GIL 患者的 EUS 图像。从四个中心收集参与者以开发并回顾性评估基于 AI 的系统。当内镜超声医师认为 SEL 为 GIST 或 GIL 时，系统将被使用。然后，它将在多中心前瞻性诊断测试中用于临床探索，即内镜超声医师和 AI 系统的联合诊断是否可以区分 GIST 和 GIL，以提高 SEL 的总体诊断准确性。
结果
该 AI 系统使用来自 752 名 GIST 或 GIL 患者的 10439 张 EUS 图像进行开发。在前瞻性测试中，在 508 例连续患者中，有 132 例患者经组织学诊断为（36 例 GIST、44 例 GIL 和 52 例其他类型的 SEL）。通过联合诊断，内镜超声医师诊断 132 例经组织学证实的参与者的总准确率从 69.7%（95%置信区间 [CI] 61.4%-76.9%）提高到 78.8%（95%CI 71.0%-84.9%；=0.01）。内镜超声医师诊断 80 例 GIST 或 GIL 患者的准确率从 73.8%（95%CI 63.1%-82.2%）提高到 88.8%（95%CI 79.8%-94.2%；=0.01）。
结论
我们开发了一种基于 AI 的 EUS 诊断系统，可有效区分 GIST 和 GIL，并提高 SEL 的诊断准确性。
Zotero 插件