Microstructural MRI provides non-invasive measures of the microstructure of white-matter axons, including diameter and g-ratio. These can theoretically be used to infer axonal conduction velocity (CV). However, several other morphological and physiological parameters also contribute to CV, which are not accessible through MRI. Using sensitivity analysis on a comprehensive model of axon electrophysiology, we test the feasibility of modelling CV and associated uncertainty using only MRI-accessible parameters. Results show 88.5% of the variance in CV is accounted for by axon diameter and g-ratio and uncertainty due to non-imageable parameters can be easily modelled by a simple linear function. When these measurements can be made reliably it is feasible to obtain estimates of axonal CVs from micro-structural parameters obtained from MRI.