评估机器学习模型在诊断帕金森病时使用声学参数的临床实用性
Zhiheng Xu1, Liwen Li2, Xining Liu1
1Department of Neurology and National Research Center for Aging and Medicine & National Center for Neurological Disorders, State Key Laboratory of Medical Neurobiology, Huashan Hospital, Fudan University, Shanghai, China.
概括
对语音录音的声学分析显示,在中国人群中诊断帕金森病 (PD) 的潜力. 这种方法在PD患者中发现了显著的声管缺陷,提供了一种新的诊断方法.
科学领域:
- 神经学 神经学
- 语音科学 语言科学
- 生物医学工程 生物医学工程
背景情况:
- 目前的帕金森病 (PD) 诊断方法有局限性.
- 脱发性关节障碍,是PD的核心症状,为新的诊断方法提供了机会.
- 语音录音的声学分析被探索为PD的潜在诊断工具.
研究的目的:
- 评估使用声学分析来诊断PD的可行性.
- 为了评估中国人口的诊断准确性.
主要方法:
- 140名中国参与者 (96名PD患者,44名健康对照) 收集了持续元音 (/a:/, /i:/, /u:/) 的语音录音.
- 声学分析确定了具有诊断价值的参数.
- 基于语音特征的机器学习模型被开发用于基于语音特征的自动化PD诊断.
主要成果:
- 在PD患者和对照人群之间发现了8个声学参数的显著差异.
- 一种光梯度增强机模型实现了高诊断性能:AUC为0.96,准确率为92.86%,灵敏度为89.47%,特异性为100.00%.
结论:
- 在中国人群中的初步发现表明,在PD患者中,有明显的语音器缺陷.
- 声学分析可用于PD诊断,并可能指导未来的临床研究.
相关概念视频
01:24Parkinson's Disease: Treatment
391
Neurodegenerative disorders, such as Parkinson's Disease (PD), involve the gradual and irreversible destruction of neurons in particular brain areas. These disorders exhibit standard features like proteinopathies, selective vulnerability of some neurons, and an interaction of intrinsic properties, genetics, and environmental influences in neural injury.
Parkinson's Disease is primarily a result of the loss of dopaminergic neurons in the substantia nigra pars compacta. The cornerstone of...
Parkinson's Disease is primarily a result of the loss of dopaminergic neurons in the substantia nigra pars compacta. The cornerstone of...
391
01:15Parkinson's Disease: Overview
722
Neurodegenerative disorders are progressive diseases that cause irreversible damage and loss to neurons in specific brain areas. Examples of these disorders include Parkinson's disease, Alzheimer's disease, Multiple Sclerosis (MS), and Amyotrophic Lateral Sclerosis (ALS). These disorders share characteristics such as proteinopathies, selective neuronal vulnerability, and a complex interplay between genetic and environmental factors. The primary therapeutic goal for these conditions is...
722