智能技术学报

文章详情

稿件标题: Diagnosis System for Parkinson's Disease using Speech Characteristics of Patients and Deep Belief Network
稿件作者: Ali H. Al-Fatlawi, Sai Ho Ling*, Mohammed H. Jabardi
关键字词: Parkinson’s disease; deep belief network; Restricted Boltzmann Machine
文章摘要: Parkinson's disease (PD) is a degenerative disorder of the nervous system. The early diagnosis of PD is important and necessary. Thus, in this paper, an efficient technique to diagnose the Parkinson's disease (PD) is proposed. Deep belief network (DBN) has been adopted to diagnose the speech characteristic of patients with PD. Through distinguishing and analysing the speech characteristic, the DBN takes the main role to diagnose PD. In this paper, DBN is used to classify the Parkinson's disease, which composes of two stacked layers of Restricted Boltzmann Machines (RBMs) and one output layer. To illustrate the efficiency of the proposed system, a case study with 31 subjects (including 23 with PD and 8 healthy) is given. The experimental results are comparedwith different approaches and related works. The overall testing accuracy of the proposed system is 94%, which is better than all of the compared methods. In short, the DBN is an efficient method to diagnose Parkinson's disease by using the speech characteristics of patients.
收录刊物: 2017年2卷4期
稿件基金:
浏览次数: 234
下载次数: 66
点击下载