Deep brain stimulation (DBS) electrodes are intended to stimulate specific areas of the brain to treat movement disorders including essential tremor, Parkinson’s disease and dystonia. An important goal in the design of next generation DBS electrodes is to minimize the power needed to stimulate specific regions of the brain. A reduction in power consumption will prolong battery life and reduce the size of implanted pulse generator. Electrode geometry is one approach to increase the efficiency of neural stimulation and reduce the power required to produce the level of activation required for clinical efficacy.

We first characterized the impedance of the presently used clinical DBS electrodes in vitro and in vivo. Characterization of the electrode-tissue interface impedance is required to quantify the composition of charge transfer to the brain tissue. The composition of charge transfer was dependent on both the current density and the sinusoidal frequency. The assumption of the DBS electrode being ideally polarizable was not valid under clinical stimulating conditions. This implies that irreversible processes that can cause electrode or tissue damage might occur when high charge injection is required for DBS.

Current density distribution is an important factor in determining patterns of neural excitation, tissue damage and electrode corrosion. We developed a recursive simulation scheme to calculate the current density distribution that incorporates the nonlinear electrode-tissue interface into finite-element based models of electrodes. The current density distributions on the electrode surface were strongly dependent on the sinusoidal frequency. The primary current density distribution without including the electrode-tissue interface can be used to estimate neural excitation, tissue damage and electrode corrosion with rectangular stimulus pulses as most of the signal power is at frequencies where the secondary current density distribution matches closely the primary current density distribution.

We designed and analyzed novel electrode geometries to decrease stimulation thresholds, thus reducing power consumption of implanted stimulators. Our hypothesis was that high-perimeter electrode geometries that increase the variation of current density on the electrode surface will generate larger activating functions for surrounding neurons and thereby increase stimulation efficiency. We investigated three classes of electrodes: segmented cylindrical electrodes, serpentine-perimeter planar electrodes, and serpentine-perimeter cylindrical electrodes. An approach that combined finite element models of potentials and cable models of axonal excitation was used to quantify the stimulation efficiency of electrodes with various geometries. Increasing the electrode perimeter increased the electrode efficiency by decreasing stimulation threshold. Both segmentation and serpentine edges provided means to increase the efficiency of stimulation. Novel cylindrical electrodes that combined segmentation with serpentine edges decreased power consumption by ∼20% for axons parallel to the electrode and by ∼35% for axons perpendicular to the electrode. These electrode designs could potentially prolong the average battery life of deep brain stimulator by more than one year.