Scholarlink

<p>Cholangioscopy with biopsy is currently the best, guideline-compliant method for assessing indeterminate biliary strictures. However, the interpretation of cholangioscopic examinations is not very standardized and the diagnostic accuracy of cholangioscopy-assisted biopsies remains limited for various reasons. In the future, support systems using deep learning algorithms may allow optimization of cholangioscopy and patient management. Algorithms based on convolutional neural networks (CNN) in particular have been increasingly adopted in medical imaging. For biliary diagnostics using cholangioscopy, several studies investigating the use of CNN-based algorithms for the detection of malignant bile duct stenoses have been published. In this paper, we describe the general principles of CNN-based algorithms and their use in percutaneous/peroral diagnostics. We address the recommendations and limitations of cholangioscopy for the diagnosis of indeterminate biliary strictures and discuss recent CNN-based algorithms that have been scientifically developed for this indication.</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

Einsatz von künstlicher Intelligenz in der biliären Diagnostik

  • 期刊论文

  • 发表日期：

    2025年1月30日

查看全文
  • Andreas Albert,
  • Philipp Dobsch,
  • Joceline Ziegler,
  • Pia Hofstetter,
  • Philipp Heumann,
  • Martina Müller-Schilling,
  • Arne Kandulski

摘要

Cholangioscopy with biopsy is currently the best, guideline-compliant method for assessing indeterminate biliary strictures. However, the interpretation of cholangioscopic examinations is not very standardized and the diagnostic accuracy of cholangioscopy-assisted biopsies remains limited for various reasons. In the future, support systems using deep learning algorithms may allow optimization of cholangioscopy and patient management. Algorithms based on convolutional neural networks (CNN) in particular have been increasingly adopted in medical imaging. For biliary diagnostics using cholangioscopy, several studies investigating the use of CNN-based algorithms for the detection of malignant bile duct stenoses have been published. In this paper, we describe the general principles of CNN-based algorithms and their use in percutaneous/peroral diagnostics. We address the recommendations and limitations of cholangioscopy for the diagnosis of indeterminate biliary strictures and discuss recent CNN-based algorithms that have been scientifically developed for this indication.