تشخیص بیماریهای عصبی پیشرونده تدریجی با استفاده از ویژگیهای زمان-فرکانسی متقارن سیگنال راه رفتن

Abstract

In the recent years, the diagnosis of neurodegenerative diseases (NDDs) has been one of the most challenging problems in the medical fields. Amyotrophic lateral sclerosis (ALS), Parkinson's disease (PD) and Huntington's disease (HD) are a group of neurological disorders affecting the quality of patient’s life. Occurrence of these diseases is due to the deterioration of motor neurons, causing human gait disturbance and asymmetry between the right and left limbs. For this purpose, in this paper various gait signals namely stride, swing, and stance intervals (from both legs) have been decomposed using a Matching pursuit (MP) algorithm. Then, two sets of differential and dynamic features have been extracted from the MP coefficients in order to quantify the amount of divergence between both limbs. Finally, the principal components of these features have been fed as an input to sparse non-negative least squares (NNLS) classifier.The proposed algorithm has been evaluated using the gait signals of 16 healthy control subjects, 13 patients with Amyotrophic lateral sclerosis (ALS), 15 patients with Parkinson’s disease (PD) and 20 patients with Huntington’s disease (HD). The results showed that the proposed method has achieved high average accuracy rates of 84.10%, 86.67%, and 91.43% for ALS, PD, and HD detection, respectively.