As a natural form of nanoscale communication, neuro-spike communication inspires the deployment of nanomachines inside the human body for healthcare. To this end, the identification of failure mechanisms in normal and diseased connections of nervous nano-networks is crucial. Thus, in this paper, we investigate the information transmission through a single myelinated axon segment. We introduce a realistic multi-compartmental model for a single myelinated segment by incorporating the axon's paranodal regions to the model. Next, we characterize the myelinated segment communication channel in terms of attenuation over the range of frequencies. Based on this, we derive the rate per channel use and upper bound on the information capacity. The performance evaluations reveal that our approach provides dramatic correction regarding frequency response. We believe that this result could have a significant effect on the characterization of demyelinated axons from the information and communication technology (ICT) perspective.