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

An Experience-driven Interpretable Multi-task Model for Segmentation and Classification of Small Cell Lung Cancer and Non-small Cell Lung Cancer from CT Images.

IEEE journal of biomedical and health informatics·2026
Same author

PCSK9 inhibitor failure in a statin-intolerant FH patient with a novel LDLR variant: a case report.

Frontiers in cardiovascular medicine·2025
Same author

Recent Advancements in Hyperspectral Image Reconstruction from a Compressive Measurement.

Sensors (Basel, Switzerland)·2025
Same author

EISegNet: Enhancing Instrument Segmentation Network via Dual-View Disparity Estimation.

IEEE journal of biomedical and health informatics·2025
Same author

Multi-Task Collaborative Assisted Training Method for Grouping Fuzzy Categories Classification of Cervical Cancer Cells.

IEEE journal of biomedical and health informatics·2025
Same author

NpCIPK6-NpSnRK1 module facilitates intersubgeneric hybridization barriers in water lily (<i>Nymphaea</i>) by reducing abscisic acid content.

Horticulture research·2025

相关实验视频

Updated: Jun 28, 2025

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

6.8K

使用RA-DL注意模块引导深度特征提取与放射学特征的小型PET数据集的多实例瘤亚型分类方法.

Zhaoshuo Diao1, Huiyan Jiang2

  • 1Software College, Northeastern University, Shenyang 110819, China.

Computers in biology and medicine
|April 16, 2024
PubMed
概括

这项研究引入了一种新的放射学-深度学习 (RA-DL) 注意力方法,用于在小型正子发射断层扫描 (PET) 数据集中准确地对瘤亚型进行分类. 这种方法提高了肝癌,肺癌和淋巴瘤的诊断能力,即使患者数据有限.

关键词:
生物医学分类生物医学分类深度学习的特点是深度学习的特点.多实例学习是指多实例的学习.定子发射断层扫描 (PET) 是一种定子发射断层扫描.无线电学 (Radiomics) 是一种辐射学.

更多相关视频

Quantifying the Brain Metastatic Tumor Micro-Environment using an Organ-On-A Chip 3D Model, Machine Learning, and Confocal Tomography
09:53

Quantifying the Brain Metastatic Tumor Micro-Environment using an Organ-On-A Chip 3D Model, Machine Learning, and Confocal Tomography

Published on: August 16, 2020

7.2K
PET and MRI Guided Irradiation of a Glioblastoma Rat Model Using a Micro-irradiator
10:48

PET and MRI Guided Irradiation of a Glioblastoma Rat Model Using a Micro-irradiator

Published on: December 28, 2017

9.5K

相关实验视频

Last Updated: Jun 28, 2025

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
07:15

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

6.8K
Quantifying the Brain Metastatic Tumor Micro-Environment using an Organ-On-A Chip 3D Model, Machine Learning, and Confocal Tomography
09:53

Quantifying the Brain Metastatic Tumor Micro-Environment using an Organ-On-A Chip 3D Model, Machine Learning, and Confocal Tomography

Published on: August 16, 2020

7.2K
PET and MRI Guided Irradiation of a Glioblastoma Rat Model Using a Micro-irradiator
10:48

PET and MRI Guided Irradiation of a Glioblastoma Rat Model Using a Micro-irradiator

Published on: December 28, 2017

9.5K

科学领域:

  • 医学成像和诊断 医学成像和诊断
  • 医疗保健中的人工智能
  • 瘤学和癌症研究研究.

背景情况:

  • 定子发射断层扫描 (PET) 对于癌症诊断和分期至关重要,但由于有限和不平衡的数据集,准确的瘤亚型分类具有挑战性.
  • 有效的治疗计划依赖于精确的瘤亚型识别,其中包括扩散的大B细胞淋巴瘤,霍奇金淋巴瘤,腺癌,小细胞癌,状细胞癌,胆管癌和肝细胞癌.

研究的目的:

  • 开发一种新的Radiomics-DeepLearning (RA-DL) 关注方法,用于在小,不平衡的PET数据集中精确地分类瘤亚型.
  • 克服小样本大小在实现精确癌症亚型识别方面的局限性.

主要方法:

  • 利用支持矢量机器 (SVM) 作为瘤亚型的分类器,结合放射学和从PET图像中提取的深度特征.
  • 采用了用于放射学特征压缩的自动编码器和RA-DL-Attention机制来提取互补的深度特征,最大限度地减少冗余.
  • 集成的二维感兴趣区域 (ROI) 分段和图像重建作为辅助任务来解决数据限制,使用多实例学习汇总损伤特征.

主要成果:

  • 在肺癌亚型分类中实现了高性能,曲线下的面积 (AUC) 值为0.82,0.84和0.83.
  • 在淋巴瘤二元分类中表现出显著的结果,AUC值为0.95和0.75.
  • 在肝癌二元分类中表现出有前途的表现,AUC值为0.84和0.86.

结论:

  • 拟议的RA-DL注意力方法显著优于在小型PET数据集上进行瘤亚型分类的替代方法.
  • 互补的放射学和深度特征的整合提高了分类准确性,为具有挑战性的临床场景提供了强大的解决方案.