Deep learning has for a long time been recognized as a powerful tool in the field of medicine for making predictions or detecting abnormalities in a patient’s data. However, up until recently, hosting of these neural networks has been relegated to the domain of servers and powerful workstations due to the vast amount of resources they require. This trend has been steadily shifting in the recent years, and we are now beginning to see more and more mobile applications with similar capabilities. Deep neural networks hosted completely on mobile platforms are extremely valuable for providing healthcare services to remote areas without network connectivity. Yet despite this, there is very little information regarding the migration process of an existing server-based neural network to a mobile environment. In this work, we describe the various techniques and considerations that should be taken into account when developing a deep-learning enabled mobile application with offline support. We illustrate the above by providing a concrete example through our experience in migrating to mobile an in-house developed medical application for detecting early signs of traumatic brain injuries.