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

Incidence and Radiographic Progression of Rheumatoid Arthritis in Anti-Citrullinated Protein Antibody-Positive Individuals: A Six-Year Population- Based Cohort Study.

Modern rheumatology·2026
Same author

New insights into spinopelvic-hip motion age-related changes in postural adaptability.

Journal of orthopaedic science : official journal of the Japanese Orthopaedic Association·2026
Same author

Impact of Unfractionated Heparin Use on Maternal Bone Mineral Density During Pregnancy: A Retrospective Study Using AI-Assisted Radiographic Analysis.

Reproductive sciences (Thousand Oaks, Calif.)·2026
Same author

Dynamic Weight-Bearing Lateral Radiographs to Detect Occult Nonunion After Ankle Arthrodesis: A Diagnostic Technical Tip and Case Series of 2 Patients.

Foot & ankle orthopaedics·2026
Same author

Overlap syndrome between sarcoidosis and Sjögren's disease presenting as hypopituitarism: case report and review of literature.

Endocrine journal·2026
Same author

Functional Resilience in Chronic Low Back Pain: Dissociating Magnetic Resonance Imaging Abnormalities From Real-World Disability in the Wakayama Spine Study.

Neurospine·2026

相关实验视频

Updated: Jul 1, 2025

An Anatomical Study of Nerves at Risk During Minimally Invasive Hallux Valgus Surgery
15:04

An Anatomical Study of Nerves at Risk During Minimally Invasive Hallux Valgus Surgery

Published on: February 17, 2018

11.7K

使用深度神经网络与基于轴的注释自动估计角.

Ryutaro Takeda1, Hiroyasu Mizuhara1, Akihiro Uchio1

  • 1Department of Orthopaedic Surgery, Faculty of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-Ku, Tokyo, 113-8655, Japan.

Skeletal radiology
|March 13, 2024
PubMed
概括

一个深度神经网络 (DNN) 模型准确地测量了足部放射图上的角 (HVA) 和间角 (IMA). 这种人工智能工具在这些关键的骨科测量中显示了与专家外科医生可比的准确性.

关键词:
深度学习是一种深度学习.脚 脚 脚 脚 脚 脚 脚 脚哈卢克斯瓦尔古斯 (Hallux Valgus) 是一种疾病.放射学 放射学 放射学 放射学

更多相关视频

Author Spotlight: Deciphering Electrical Networks Behind Complex Brain Activities and Disorders
05:49

Author Spotlight: Deciphering Electrical Networks Behind Complex Brain Activities and Disorders

Published on: November 1, 2024

779
Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
09:41

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

Published on: April 21, 2023

1.6K

相关实验视频

Last Updated: Jul 1, 2025

An Anatomical Study of Nerves at Risk During Minimally Invasive Hallux Valgus Surgery
15:04

An Anatomical Study of Nerves at Risk During Minimally Invasive Hallux Valgus Surgery

Published on: February 17, 2018

11.7K
Author Spotlight: Deciphering Electrical Networks Behind Complex Brain Activities and Disorders
05:49

Author Spotlight: Deciphering Electrical Networks Behind Complex Brain Activities and Disorders

Published on: November 1, 2024

779
Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
09:41

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

Published on: April 21, 2023

1.6K

科学领域:

  • 整形外科成像分析分析
  • 医学中的人工智能
  • 放射性测量技术的使用.

背景情况:

  • 角 (HVA) 和间角 (IMA) 是诊断脚形的关键指标.
  • 在X光照上手动测量HVA和IMA可能是主观的,耗时的.
  • 开发用于精确放射分析的自动化方法对于高效的临床实践至关重要.

研究的目的:

  • 开发和验证深度神经网络 (DNN) 模型,用于自动测量HVA和IMA.
  • 通过将其测量结果与经验丰富的足外科医生的测量结果进行比较,评估DNN模型的准确性.
  • 评估模型的性能与已建立的评级者之间的可靠性标准相比.

主要方法:

  • 一个DNN模型被训练来识别足部X射线图上的骨轴,用于HVA和IMA计算.
  • 该模型是使用来自联合队列的1798张放射图开发的.
  • 对92张X射线图进行了追溯验证,将DNN测量与三个外科医生手工测量的中位数进行了比较.

主要成果:

  • 在DNN模型中,HVA的平均绝对误差 (MAE) 为1.3°,远远低于外科医生的评分间差异 (2.0°).
  • 对于IMA,模型的MAE为0.8°,与外科医生的评分间差异 (1.0°) 相比,没有显著差异.
  • 该模型在测量HVA和IMA时表现出高精度和可靠性.

结论:

  • 开发的DNN模型准确地测量了HVA和IMA在足部放射图上.
  • 该模型的准确性与专业脚外科医生的准确性相当.
  • 这种自动化方法提供了一个可靠的替代方法,用于对脚对齐的放射评估.