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

Consensus recommendations for clinical functional MRI applied to language mapping.

Aperture neuro·2026
Same author

The Effect of Age on Post-Stroke Language Outcomes.

Journal of aging research·2026
Same author

'An adrenaline-fueled emergency': a qualitative thematic analysis of reviews of deaths in England where DNACPR (Do Not Attempt Cardiopulmonary Resuscitation) recommendations are not implemented for people with a learning disability.

BMJ open·2026
Same author

SONIVA database: Speech recognition validation in aphasia.

Scientific data·2026
Same author

Testing and tracking in the UK: A dynamic causal modelling study.

Wellcome open research·2026
Same author

Fragile memories for fleeting percepts.

Consciousness and cognition·2026

相关实验视频

Updated: Jun 21, 2025

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.0K

预测中风后的康复:使用可解释的AI进行深度学习,多式联络数据和特征选择.

Adam White1, Margarita Saranti2, Artur d'Avila Garcez1

  • 1Department of Computer Science, City, University of London, UK.

NeuroImage. Clinical
|July 13, 2024
PubMed
概括

机器学习使用结合的神经成像和临床数据准确地预测中风后失语. 使用卷积神经网络 (CNN) 与感兴趣区域 (ROI) 的新方法实现了最高的准确性,改善了患者的结果预测.

更多相关视频

Determining the Functional Status of the Corticospinal Tract Within One Week of Stroke
09:10

Determining the Functional Status of the Corticospinal Tract Within One Week of Stroke

Published on: February 22, 2020

8.5K
Electroencephalography Network Indices as Biomarkers of Upper Limb Impairment in Chronic Stroke
06:37

Electroencephalography Network Indices as Biomarkers of Upper Limb Impairment in Chronic Stroke

Published on: July 14, 2023

862

相关实验视频

Last Updated: Jun 21, 2025

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches
09:47

Author Spotlight: Advancing Alzheimer's Research – Exploring Early Detection and Multi-Omics Approaches

Published on: December 15, 2023

1.0K
Determining the Functional Status of the Corticospinal Tract Within One Week of Stroke
09:10

Determining the Functional Status of the Corticospinal Tract Within One Week of Stroke

Published on: February 22, 2020

8.5K
Electroencephalography Network Indices as Biomarkers of Upper Limb Impairment in Chronic Stroke
06:37

Electroencephalography Network Indices as Biomarkers of Upper Limb Impairment in Chronic Stroke

Published on: July 14, 2023

862

科学领域:

  • 神经科学是一个神经科学.
  • 医疗成像医学成像
  • 机器学习 机器学习

背景情况:

  • 机器学习 (ML) 对预测中风后症状和康复反应具有前景.
  • 高维神经成像数据,小数据集和整合多种数据类型构成重大挑战.
  • 准确的预测对于个性化中风恢复策略至关重要.

研究的目的:

  • 评估结合神经成像和表格数据用于中风后失语预测的策略.
  • 引入和评估一种新的卷积神经网络 (CNN) 方法,将感兴趣区域 (ROI) 与表格数据集成在一起.
  • 在中风幸存者中预测口语图片描述能力 (非法斯 vs 非法斯).

主要方法:

  • 对比MRI扫描和特征选择策略的2D图像总结.
  • 开发了一个CNN,训练有素的组合MRI衍生ROI和象征性表格数据表示.
  • 评估了MRI上的2D和3DCNN架构以及758名中风幸存者的表格数据.
  • 使用五组分组进行培训,验证和持有测试集 (锁盒).

主要成果:

  • 基线逻辑回归实现了0.678的准确度 (仅仅是损伤大小).
  • 精度增加到0.757 (增加症状严重程度) 和0.813 (增加恢复时间).
  • 最好的性能 (0.854准确度,0.899AUC,0.901F1) 是通过使用8个ROI和表格数据的2D残余神经网络 (ResNet) 实现的.
  • 这种模型在286名中度至重度初始失语症 (AUC = 0.865) 参与者的具有挑战性的子集上也表现最好.

结论:

  • 通过CNN将神经成像 (ROI) 和表格数据结合起来,显著提高了中风后失语分类的准确性.
  • 这种方法甚至在ML中典型的有限数据集大小的情况下也显示出有效性.
  • 未来的工作可以通过直接从医院扫描仪中整合图像来进一步提高准确性.