The goal of this work is to recover semantic and structural information from medical documents in electronic format. Despite the progressive diffusion of Electronic Health Record systems, a lot of medical information, also for legacy reasons, is available to patients and physicians in image-only or textual format. The difficulties of obtaining such information when needed result in high costs for health providers. In this work we develop the concept of a system designed to convert legacy medical documents into a standard and interoperable format compliant with the Clinical Document Architecture model by the means of semantic annotation.
展开▼