Jove
Visualize
联系我们
JoVE
x logofacebook logolinkedin logoyoutube logo
关于 JoVE
概览领导团队博客JoVE 帮助中心
作者
出版流程编辑委员会范围与政策同行评审常见问题投稿
图书馆员
用户评价订阅访问资源图书馆顾问委员会常见问题
研究
JoVE JournalMethods CollectionsJoVE Encyclopedia of Experiments存档
教育
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab Manual教师资源中心教师网站
使用条款与条件
隐私政策
政策

相关实验视频

Updated: Jan 10, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.3K

评估深度学习模型用于多类上端内镜疾病细分:一项全面的比较研究.

In Neng Chan1, Pak Kin Wong2, Tao Yan3

  • 1Department of Electromechanical Engineering, University of Macau, Macau 999078, China.

World journal of gastroenterology
|November 20, 2025
PubMed
概括

相关概念视频

您也可能阅读

相关文章

通过共同作者、期刊和引用图与本文相关的文章。

排序
Same author

Chemokine CCL21 promotes the pathological progression of endometriosis by regulating inflammatory cytokine expression and activating the NF-κB signaling pathway.

International immunopharmacology·2026
Same author

Combined nocturnal sleep pattern and napping duration in relation to metabolic dysfunction-associated steatotic liver disease risk and prediction in type 2 diabetes mellitus.

Diabetology & metabolic syndrome·2026
Same author

Poriae cutis triterpenes ameliorate metabolic-associated steatotic liver disease by inhibiting PXR/NLRP3 pathways.

Chinese medical journal·2026
Same author

Comment on "BMJ Publications on Interventional Techniques Do Not Meet Appropriateness Criteria of Conducting a Rapid Review: A Comprehensive Review".

Pain physician·2026
Same author

Dopamine neurons in the ventral tegmental area regulate emotional disturbances in a mouse model of acute respiratory distress syndrome.

Science bulletin·2026
Same author

Long-Range Exciton Diffusion and Efficient Hole Transfer in Molecular Non-Fullerene Acceptors with Strong Resonance Structure.

The journal of physical chemistry letters·2026
Same journal

Lessons from extended induction and practical evidence for improving tofacitinib therapy in ulcerative colitis.

World journal of gastroenterology·2026
Same journal

Small animal <i>ex vivo</i> machine perfusion of the liver: A comprehensive literature review.

World journal of gastroenterology·2026
Same journal

Comparable remission and health care use in real-world inflammatory bowel disease patients initiating originator biologics <i>vs</i> biosimilars.

World journal of gastroenterology·2026
Same journal

Diagnosis and management of metabolic dysfunction-associated steatohepatitis in patients with chronic hepatitis B infection.

World journal of gastroenterology·2026
Same journal

Simultaneous treatment of concomitant achalasia coexisting with epiphrenic diverticulum: The practice of submucosal tunneling technique.

World journal of gastroenterology·2026
Same journal

Lianhe Xiaozhi ointment ameliorates metabolic dysfunction-associated steatotic liver disease <i>via</i> peroxisome proliferator-activated receptor alpha pathway activation.

World journal of gastroenterology·2026
查看所有相关文章
此摘要是机器生成的。

深度学习模型显示出对上部胃肠道疾病的细分有希望,像Swin-UMamba和SegFormer这样的等级架构显示出高准确性. 进一步的临床验证对于内镜的实际应用至关重要.

科学领域:

  • 医疗成像医学成像
  • 人工智能的人工智能
  • 计算机视觉 计算机视觉

背景情况:

  • 上部胃肠道 (UGI) 疾病在内镜中构成诊断挑战,原因是视觉相似性和观察者可变性.
  • 使用深度学习 (DL) 的自动细分可以帮助内镜医生,但多类UGI疾病细分尚未得到充分探索.
  • 用于UGI疾病细分的DL模型的临床验证是有限的,阻碍了实际应用.

研究的目的:

  • 评估17个最先进的深度学习 (DL) 模型,用于多类上胃肠道 (UGI) 疾病细分.
  • 评估UGI内镜不同DL架构的临床翻译和现实世界的适用性.
  • 确定可以减少诊断错误和支持临床决策的DL模型.

主要方法:

  • 评估了17个DL模型 (基于CNN,变压器,mamba) 在自收集的数据集 (3313张图像,9个类) 和EDD2020数据集 (386张图像,5个类) 上.
  • 评估了每个模型的细分性能 (IoU) 和性能效率权衡.
  • 进行统计分析和交叉数据集评估,以衡量性能差异和概括能力.

主要成果:

  • 斯温-UMamba实现了最高的细分性能 (IoU: 89.06%自收集, 77.53% EDD2020),其次是SegFormer和ConvNeXt + UPerNet.
关键词:
深度学习是一种深度学习.疾病细分 疾病细分胃肠道疾病 胃肠道疾病医学成像医学成像上部内镜检查上部内镜检查

更多相关视频

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

3.2K

相关实验视频

Last Updated: Jan 10, 2026

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography
04:48

Application of Deep Learning-Based Medical Image Segmentation via Orbital Computed Tomography

Published on: November 30, 2022

3.3K
Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application
05:56

Objectification of Tongue Diagnosis in Traditional Medicine, Data Analysis, and Study Application

Published on: April 14, 2023

3.2K
  • SegFormer提供了最佳的精度-效率平衡 (92.02%),适合实时临床使用.
  • 基于变压器的模型在交叉数据集评估中更好地泛化 (64.78%-71.52%的保留率),尽管整体性能下降.

    • 结论:

    • 像Swin-UMamba和SegFormer这样的等级DL架构显示了UGI疾病细分的巨大潜力.
    • 这些模型可以帮助减少错过的诊断,并提高内镜工作流程的效率.
    • 在内镜实践中广泛部署现实世界之前,强大的临床验证至关重要.