Numerical investigation of transcranial direct current stimulation on cortical modulation

Transcranial direct current stimulation (tDCS) is a non-invasive and sub-convulsive functional stimulation technique with applications in both clinical therapy and neuro-science research. The technique provides researchers and clinicians with a unique tool capable of modulating the neural excitability in both the central and peripheral nervous system. On a clinical level, the procedure has been used quite extensively for its potential therapeutic applications in a number of neurological disorders. Despite the advantages of being safe, low cost and easy to administer, our limited under-standing on interaction mechanisms between the stimulation parameters and biologi-cal materials has impeded the development and optimisation of tDCS based therapies. The focus of this thesis is to develop a realistic finite element based human head model to address the problems involved in the forward modelling of transcranial direct current stimulation. The study explores the effects of model complexities and anisotropic material properties on field estimations. The sensitivity of electric field and current density on accurate modelling of cortical and non-cortical structures, and the influence of heterogeneously defined anisotropic electric conductivity on field parameters were analysed in an incremental manner. Using the averaged and the subject specific Magnetic Resonance Imaging (MRI) and Diffusion Tensor Imaging (DTI) data, the head models with detailed anatomical features and realistic tissue conductive properties, were developed and employed to specifically address the role of stimulation parameters, such as: morphological variations, structural details, tissue behaviour, inter-subject variations, electrode montages and neural fibre pathways for defining the site and strength of modulation/stimulation. This thesis demonstrates the importance of human head modelling in elucidating the complex electric field and current density profiles instigated by the non-invasive electric stimulation. The results of this study strongly support the initial hypothesis that model complexity and accurate conductivity estimation play a crucial role in determining the accurate predictions of field variables. The study also highlighted the inadequacy of scalar field maps to decipher the complex brain current flow patterns and axonal/neural polarization. With the proposed refinements, model based strategies can be employed to optimally select the required stimulation strength and electrode montage specific to individual dose requirements. Therefore, the work con-ducted in this study will bridge the gap between the current clinical practices and the subject specific treatments by providing accurate physiologically representative simulation.