Jove
Visualize
联系我们

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

A Conceptual Framework for Integrated and Collaborative Orthodontic AI.

Studies in health technology and informatics·2026
Same author

DilatedToothSegNet: Tooth Segmentation Network on 3D Dental Meshes Through Increasing Receptive Vision.

Journal of imaging informatics in medicine·2024
Same author

Survival of retained permanent canines after autotransplantation: A retrospective cohort study.

American journal of orthodontics and dentofacial orthopedics : official publication of the American Association of Orthodontists, its constituent societies, and the American Board of Orthodontics·2024
Same author

A Critical Assessment of Generative Models for Synthetic Data Augmentation on Limited Pneumonia X-ray Data.

Bioengineering (Basel, Switzerland)·2023
Same author

Augmentation strategies for an imbalanced learning problem on a novel COVID-19 severity dataset.

Scientific reports·2023
Same author

ResNetFed: Federated Deep Learning Architecture for Privacy-Preserving Pneumonia Detection from COVID-19 Chest Radiographs.

Journal of healthcare informatics research·2023
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: May 17, 2025

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

2.6K

评估面具自主监督学习框架,用于3D牙科模型细分任务.

Lucas Krenmayr1,2, Reinhold von Schwerin3, Daniel Schaudt3

  • 1Cooperative Doctoral Program for Data Science and Analytics, University of Ulm, 89081, Ulm, Germany. lucas.krenmayr@uni-ulm.de.

Scientific reports
|May 14, 2025
PubMed
概括

蒙面自主监督学习增强了牙科模型的深度学习,改善了自动化治疗规划. 当标记数据稀缺时,这种方法最有利,提高了牙细分等任务的准确性.

更多相关视频

Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures
09:10

Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures

Published on: August 5, 2021

1.7K
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

2.3K

相关实验视频

Last Updated: May 17, 2025

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

2.6K
Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures
09:10

Digital Hybrid Model Preparation for Virtual Planning of Reconstructive Dentoalveolar Surgical Procedures

Published on: August 5, 2021

1.7K
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

2.3K

科学领域:

  • 3D 计算机视觉 3D 计算机视觉
  • 医学成像分析分析 医学成像分析
  • 人工智能在牙科中的应用

背景情况:

  • 牙科的自动化计算机辅助治疗规划依赖于使用3D牙科模型的深度学习.
  • 高精度模型的开发受到标记医疗数据的稀缺性所阻碍.
  • 蒙面自主监督学习 (SSL) 为数据稀缺性挑战提供了潜在的解决方案.

研究的目的:

  • 调查四个面具自主监督学习框架 (Point-BERT,Point-MAE,Point-GPT,Point-M2AE) 对3D牙科模型的有效性.
  • 评估预训练对下游任务的影响,例如牙和支架细分.
  • 确定在牙科应用中应用掩盖SSL的最佳条件.

主要方法:

  • 在4000多个未标记的3D牙科模型上进行了四个掩盖的SSL框架的预训练.
  • 在Teeth3DS数据集上微调预训练模型以进行牙细分.
  • 对自定义数据集进行微调,用于支架细分.
  • 在下游任务中对提高绩效的实验性评估.

主要成果:

  • 前期培训显著提高了下游任务的性能,特别是在有限或不平衡的标记数据的情况下.
  • 掩盖SSL预训练的好处在数据稀缺条件下最为明显.
  • 随着可用的标记数据数量的增加,性能收益会减少.

结论:

  • 蒙面自主监督学习是改善牙科应用的深度学习模型的可行策略,特别是在数据有限的临床环境中.
  • 使用掩盖的SSL进行预训练可以提高自动化治疗计划系统的准确性和临床可用性.
  • 了解标记数据可用性和性能增长之间的关系对于有效实施至关重要.