Title: Energy entropy feature for the discrimination between the patients with amyotrophic lateral sclerosis and healthy subjects

Authors: Chandrakar Kamath

Addresses: ShanthaNilaya, 107, First stage, Ananthnagar, Manipal 576104, India

Abstract: Amyotrophic Lateral Sclerosis (ALS) is a type of neurological disease due to the degeneration of motor neurons. Hypothesising that ALS patient will experience an altered gait rhythm, with changes in the magnitude of the stride-to-stride fluctuations and perturbations in the fluctuation dynamics, the gait rhythm of ALS patients is compared with those of healthy subjects using energy entropy (EE) feature and approximate entropy (ApEn). It is found that EE (ApEn) was increased (decreased) in ALS patients compared to EE feature in healthy controls. It is found that in the diagnostic ability test ApEn performed well with AROC = 0.8222, accuracy = 83.3%, specificity = 66.7%, sensitivity = 93.3%, and precision = 85.7%, while EE feature outperformed with AROC = 0.9803, accuracy = 93.8%, specificity = 94.1%, sensitivity = 96.3%, and precision = 96.1%. Results also showed that the diagnostic performance of EE feature is much superior compared to that of ApEn in discriminating ALS from patients with Parkinson and Huntington diseases.

Keywords: amyotrophic lateral sclerosis; ALS; approximate entropy; energy entropy feature; Huntington's disease; Parkinson's disease; neurological diseases; gait rhythm; stride-to-stride fluctuations; fluctuation dynamics.

DOI: 10.1504/IJBET.2016.075423

International Journal of Biomedical Engineering and Technology, 2016 Vol.20 No.3, pp.208 - 225

Received: 20 Mar 2015
Accepted: 22 Jul 2015

Published online: 22 Mar 2016 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article