Diagnosis of Parkinson’s disease using Gait Dynamics and Images

• Published in: Procedia computer science Vol. 165; pp. 428 - 434
• Main Authors: Nancy Noella, R S, Gupta, Divyansh, Priyadarshini, J
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 2019
• Subjects: Gait Dynamics; Parallel Factor Analysis (PARAFAC); Parkinson’s disease (PD); Substantia Nigra; Tensor decomposition; Tensor decomposition; Parallel Factor Analysis (PARAFAC); Parkinson’s disease (PD); Substantia Nigra; Gait Dynamics
• ISSN: 1877-0509; 1877-0509
• DOI: 10.1016/j.procs.2020.01.002

Parkinson’s disease is a neuro-degenerative disorder in which dopamine producing neurons in the brain structure called substantia nigra has damaged entire over time. PD leads to number of modal problems and mental disabilities. This paper presents a gait dynamic technology for the early stage prediction of Parkinson’s disease and a discussion on the Image processing technologies related to PD diagnosis. The analysis of gait in Parkinson’s disease helps to understand the behaviour of the neural system and so the early detection of Parkinson’s disease is possible. This can help neurologists to improve their treatment and to give guidance in reintegrate programs. This paper introduces an efficient multi-sensor data analysis of gait force in PD with respect to healthy subjects using PARAFAC model. Tensor decomposition is proposed in the work for the analysis of multi-sensors data. The data are collected from PhysioNet gait database consist of multichannel recording from force sensors of 93 patients with PD and 73 healthy subjects.