Modeling of Transcranial Magnetic Stimulation Versus Pallidal Deep Brain Stimulation for Parkinson's Disease

Non-invasive repetitive transcranial magnetic stimulation (TMS) is a neuromodulation technique in which time-varying magnetic fields induce electric fields on the patient's brain, thus stimulating neurons in the targeted cortical region. We have created a new model of the basal ganglia motor pathway in the human brain, using the Python package called NEST. We simulated healthy nuclei with random noise and connection weights optimized to mimic established background neuronal activity noise in the connecting nuclei. We then simulated Parkinsonism by removing dopaminergic (DAergic) input from the substantia nigra pars compacta to the striatum. The results showed the production of characteristic Parkinsonian neuronal features, ranging from pathological highly irregular to burst signals matching those found during in vivo animal and human microelectrode recording studies. With the model thus validated, we modeled deep brain stimulation (DBS) current delivery to the globus pallidus internus (GPi) in a Parkinsonian brain and simulated neuron firing as an effect of DBS. Finally, repetitive TMS at various frequencies was simulated at the level of the motor cortex, which produced predicted downstream firing rate effects and intra-neuronal synchrony at the level of the GPi. Although typical clinical TMS at 10 pulses/s was suggested to offer some beneficial effects, firing rates and patterns and desynchronous firing were most effectively normalized to match those seen in a healthy GPi upon stimulation at 50 pulses/s. Thus, our modeling suggests that cortical, non-invasive TMS may be a viable alternative to DBS surgery for patients with Parkinson's disease (PD).