Bioinformatic Approaches Used in Modelling Human Tremor

Manto      Mario; Grimaldi      Giuliana; Lorivel      Thomas; Farina      Dario; Popovic      Lana; Conforto      Silvia; D'Alessio      Tommaso; Belda-Lois      Juan-Manuel; Pons      Jose-Luis; Rocon      Eduardo

doi:10.2174/157489309788184747

Abstract

Bioinformatics is a field of information technology concerning the storage, retrieval, analysis, visualization, prediction and analysis of sets of data with biological or clinical significance. Tremor is a common movement disorder, for which pharmacological and neurophysiological models have been developed these last 3 decades, and which is at the frontier of biology, health sciences and computer technologies. Recently, new biomechanical modelling approaches of tremor have been proposed, based upon ambulatory systems and body area networks (BAN). Use of digital signal processing (DSP) techniques taking into account the non-linearity and non stationarity features of tremor time-series is reviewed in the present article. In particular, algorithms for instantaneous assessments of oscillations and direct online cancellations have been suggested. We discuss the advantages and drawbacks of the tremor detection algorithms, as well as prediction tools. In addition, promising models based upon neural networks, conductance studies and brain neurotransmitters are under development. These models will allow the accurate simulation of the behaviour of limbs. Their impact is outlined. The field of tremor research represents an excellent application of bioinformatics in medicine and rehabilitation.

Keywords: Tremor, bioinformatics, brain, animal models, time-series, Fast Fourier Transform, tremor detection algorithms, computational models