UNSTRUCTURED
Although NLP techniques support automated information extraction in number of industries, the adoption of NLP methods to extract patient level information from Electronic Health Records has been slow. This could be attributed to a disconnect between state-of-the-art systems developed by researchers and their ability to support healthcare decision making that leads to improved outcomes. We enumerate a set of practical considerations for developing NLP system that are scientifically innovative and have potential to improve health outcomes. The key considerations that we propose include determining (1) the readiness of the data and compute resources for NLP, (2) the organizational incentives to use and maintain the NLP systems and (3) the feasibility of implementation and evaluation. They are intended to help to enable a system that is well-positioned to scale to other health systems in the US, and globally.