Scholarlink

<p>Artificial Intelligence (AI) will radically transform digestive endoscopy in the next decade. The assistance of computer-vision algorithms has very high performance in pattern recognition of endoscopic procedures and has been studied for detection, diagnosis, quality control, and endoscopic training. In upper endoscopy, AI algorithms have been developed to study malignant and non-malignant diseases of the esophagus and stomach with higher performance than human endoscopists. In wireless capsule imaging, AI may reduce reading times and increase efficiency. In colonoscopy, multiple studies have shown that algorithms can increase polyp detection by twofold and assist in diagnosing adenomas. AI can provide objective quality assessment and may shorten the learning curve of trainees. However, algorithm errors, medicolegal issues, and lack of applicability in different populations may limit the physician–AI collaboration. In this manuscript, we review the applications of AI in endoscopic procedures, its impact on endoscopy education, discuss its limitations, and describe the future of computer vision in gastroenterology.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Artificial intelligence for computer assistance in endoscopic procedures and training

  • 期刊论文

  • 发表日期：

    2025年1月17日

查看全文
  • Pablo Achurra,
  • Domingo Mery,
  • Arnoldo Riquelme,
  • Chaya Shwaartz

摘要

Artificial Intelligence (AI) will radically transform digestive endoscopy in the next decade. The assistance of computer-vision algorithms has very high performance in pattern recognition of endoscopic procedures and has been studied for detection, diagnosis, quality control, and endoscopic training. In upper endoscopy, AI algorithms have been developed to study malignant and non-malignant diseases of the esophagus and stomach with higher performance than human endoscopists. In wireless capsule imaging, AI may reduce reading times and increase efficiency. In colonoscopy, multiple studies have shown that algorithms can increase polyp detection by twofold and assist in diagnosing adenomas. AI can provide objective quality assessment and may shorten the learning curve of trainees. However, algorithm errors, medicolegal issues, and lack of applicability in different populations may limit the physician–AI collaboration. In this manuscript, we review the applications of AI in endoscopic procedures, its impact on endoscopy education, discuss its limitations, and describe the future of computer vision in gastroenterology.