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

Parkinson Disease l: Introduction01:24

Parkinson Disease l: Introduction

Parkinson’s disease is a chronic, progressive neurodegenerative disorder that primarily affects movement. It is characterized by motor symptoms such as resting tremors, muscle rigidity, bradykinesia (slowness of movement), and postural instability. Patients may notice hand tremors at rest, stiffness during movement, or a shuffling gait. In addition to motor features, non-motor symptoms include sleep disturbances, mood and behavioral changes, constipation, and cognitive impairment, all of which...

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

Synthetic data augmentation for improving performance in deep learning models for anatomical landmark localization on the distal upper limb.

Frontiers in bioengineering and biotechnology·2026
Same author

MXene-based hybrid nanocomposites for enhanced enzyme-electrode electron transfer for high-performance glucose biosensing.

Bioelectrochemistry (Amsterdam, Netherlands)·2026
Same author

Liquid-Gated Field-Effect Transistor-Based Biosensor for Uric Acid Detection.

Biosensors·2026
Same author

Scattering Coefficient Estimation Using Thin-Film Phantoms with a Spectral-Domain Dental OCT System.

Sensors (Basel, Switzerland)·2026
Same author

High-performance uric acid detection using a hierarchical NiO nanostructure-based biosensor.

Journal of materials chemistry. B·2026
Same author

Environmental DNA reveals the distinct genetic diversity and evolutionary pathways of the Chinese Minnow Rhynchocypris oxycephalus in Korean freshwater systems.

Scientific reports·2025

相关实验视频

Updated: Jul 18, 2026

Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody
09:09

Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody

Published on: September 27, 2024

528

基于语音的帕金森病检测使用预训练的自主监督自动语音识别 (ASR) 模型和监督的对比学习.

Hadi Sedigh Malekroodi1, Nuwan Madusanka2, Byeong-Il Lee1,2,3

  • 1Industry 4.0 Convergence Bionics Engineering, Pukyong National University, Busan 48513, Republic of Korea.

Bioengineering (Basel, Switzerland)
|July 29, 2025
PubMed
概括

这项研究表明,使用自动语音识别 (ASR) 模型的高级语音分析可以有效检测帕金森病 (PD). 微调的ASR模型与对比学习显著优于早期PD诊断的传统方法.

关键词:
胡伯特·赫伯特 (Hubert) 是一个著名的作家.帕金森病 (PD) 是一种疾病.在Wav2Vec 2.0中使用.深度学习是一种深度学习.监督的对比学习学习.转移学习转移学习

更多相关视频

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception
05:48

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Published on: August 9, 2024

1.6K
Asthma Detection Research Based on Voice Signal Processing and Machine Learning
04:04

Asthma Detection Research Based on Voice Signal Processing and Machine Learning

Published on: July 22, 2025

413

相关实验视频

Last Updated: Jul 18, 2026

Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody
09:09

Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody

Published on: September 27, 2024

528
Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception
05:48

Author Spotlight: Investigating the Impact of Emotional Prosodies on Voice Recognition and Perception

Published on: August 9, 2024

1.6K
Asthma Detection Research Based on Voice Signal Processing and Machine Learning
04:04

Asthma Detection Research Based on Voice Signal Processing and Machine Learning

Published on: July 22, 2025

413

科学领域:

  • 神经学 神经学
  • 计算语言学 计算语言学
  • 机器学习 机器学习

背景情况:

  • 语音障碍是帕金森病 (PD) 的早期指标.
  • 传统的PD诊断方法可能具有侵入性或缺乏早期检测的敏感性.
  • 自动语音识别 (ASR) 模型为非侵入性PD评估提供了潜力.

研究的目的:

  • 评估微调的预训练ASR模型 (Wav2Vec 2.0,HuBERT) 对帕金森病检测的有效性.
  • 通过将监督对比 (SupCon) 学习与ASR模型集成来增强PD检测.
  • 将ASR衍生特征与已确定的声学特征 (MFCC,eGeMAPS) 进行比较,并使用Grad-CAM评估模型的可解释性.

主要方法:

  • 在用于PD检测的NeuroVoz数据集上微调Wav2Vec 2.0和HuBERT模型.
  • 实施监督对比 (SupCon) 学习以改善特征歧视.
  • 将ASR特征与Mel-frequency cepstral系数 (MFCCs) 和eGeMAPS进行比较.
  • 使用Grad-CAM可视化语音区域对于PD预测至关重要.

主要成果:

  • 预先训练的ASR模型在PD检测中显著超过了基线声学特征.
  • SupCon学习方法始终比传统的交叉 (CE) 模型产生更好的结果.
  • 与CE模型相比,Wav2Vec 2.0和HuBERT与SupCon实现了高F1得分 (90.0%和88.99%) 和更高的AUC值.

结论:

  • 精心调整的ASR模型,特别是Wav2Vec 2.0和HuBERT与SupCon,显示了非侵入性PD检测的高精度.
  • 基于ASR的语音分析为早期诊断和监测帕金森病提供了一个可扩展和有前途的工具.
  • 集成先进的机器学习技术增强了语音分析在神经疾病评估中的潜力.