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

Understanding the Content and Purpose of S3 Clinical Practice Guidelines in Dentistry.

International endodontic journal·2026
Same author

AI-assisted Software Development in Digital Dentistry: A Technical Innovation Report with Three Open-Source Applications.

Journal of dentistry·2026
Same author

Predicting Perceived Profile Attractiveness From Cephalometric Measurements Using Machine Learning.

International dental journal·2026
Same author

Diagnosis and Dentists' Treatment Preferences for Vestibular Enamel Defects-A Cross-Sectional Survey.

Journal of esthetic and restorative dentistry : official publication of the American Academy of Esthetic Dentistry ... [et al.]·2026
Same author

Correction: Quantum-inspired fused explainable deep learning framework for early enamel caries classification in intraoral photographs.

Odontology·2026
Same author

What If the External Crown Surface of Teeth Could Predict the Pulp Chamber? A DeepSDF-Based Approach.

International endodontic journal·2026
Same journal

Adhesion of resin composites to 3D-printed dental resins: A study on the effect of surface conditioning methods and repair materials.

Journal of dentistry·2026
Same journal

DGADS: A Graph-based Agentic Decision Support System for Precision Dental Question Answering.

Journal of dentistry·2026
Same journal

Preventive Effects of a Strontium‑Containing Nano Bioactive Glass Hydrogel Against Enamel Caries: An In Vitro Study.

Journal of dentistry·2026
Same journal

Does robotic surgery offer the highest accuracy for delayed implant placement in single-tooth spaces? A network meta-analysis of randomized clinical trials.

Journal of dentistry·2026
Same journal

Multiomics Analyses in Young Grade C Molar Incisor Pattern Periodontitis.

Journal of dentistry·2026
Same journal

Breath-Based Detection of Oral Diseases Using Sensors and Machine Learning.

Journal of dentistry·2026
查看所有相关文章

相关实验视频

Updated: May 23, 2025

Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
05:49

Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization

Published on: February 23, 2024

754

通过合成数据改进基于机器学习的咬翅细分.

Ekaterina Tolstaya1, Antonin Tichy2, Sebastian Paris3

  • 1Department of Conservative Dentistry and Periodontology, LMU University Hospital, LMU Munich, Goethestraße 70, 80 336, Munich, Germany.

Journal of dentistry
|March 11, 2025
PubMed
概括
此摘要是机器生成的。

合成数据有助于牙植入物细分在咬伤翼X射线图,当微调在原始数据. 仅在合成数据上的训练可能会降低性能,但微调可以提高代表性不足的类型的模型准确性.

关键词:
人工智能的人工智能是人工智能.数据集不平衡的原因牙科 牙科是指牙科的专业.扩散模型是一个扩散模型.生成性的对抗性网络.合成医疗数据 合成医疗数据

更多相关视频

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

相关实验视频

Last Updated: May 23, 2025

Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization
05:49

Author Spotlight: Advancing CBCT and Digital Dental Image Integration with AI-Assisted Digitization

Published on: February 23, 2024

754
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

科学领域:

  • 医疗图像分析 医学图像分析
  • 在牙科领域的机器学习.
  • 放射学中的人工智能

背景情况:

  • 阶级不平衡在医学图像分析中构成了重大挑战,特别是在对代表性不足的结构进行细分时,例如咬翼放射图中的牙科植入物.
  • 在训练和测试数据集中,植入物的像素级表示非常低 (0.03%和0.07%),需要先进的技术来提高模型性能.

研究的目的:

  • 通过扩散模型和生成对抗网络 (pix2pix) 生成的合成数据用于解决牙科植入物细分中的阶级不平衡的有效性.
  • 为了比较在各种数据策略上训练的U-Net细分模型的性能,包括原始,合成,过量采样和微调数据集.

主要方法:

  • 通过扩散和pix2pix模型生成与牙科植入物丰富的合成数据集.
  • 在四个不同的数据集上训练U-Net细分模型:原始,合成,合成,然后对原始数据进行微调,然后对原始数据进行纯粹的过量采样.
  • 使用精度,子得分,回忆,F1得分和ROC AUC等指标评估模型性能.

主要成果:

  • 仅在原始数据集上进行训练的U-Net模型未能对植入物进行细分.
  • 纯粹的过量采样产生了最高的精度,而仅在合成数据上进行训练则导致所有指标的表现较差.
  • 在合成数据上预先训练的模型与原始数据集的微调实现了最高的子得分,回忆,F1得分和ROC AUC,优于其他方法.

结论:

  • 合成数据本身就会降低细分模型的性能,特别是在少数群体中.
  • 在合成数据和原始数据上预先训练的微调模型显著提高了代表性不足的班级的表现,为人工智能驱动的牙科成像提供了有前途的方法.