Automated Knowledge Structuring of Medical Charts Data

We offer an Environmental Surveillance Framework that will have the ability to quickly run data filters over a consolidated set of medical charts to “raise red flags” and draw attention to anomalies used to help identify the onset or origin of changes in health functionally acting as diagnostic decision aids. The ability to trace results back to the original individual medical cases, for evaluators to create and edit filters so data can be evaluated in real or near real- time. The Environmental Surveillance Framework comprises three major components to extract, understand and access the data. The first component extracts and encodes narrative free text from databases such as AHLTA using natural language processing (NLP) with an ontology. The second component harmonizes extracted data elements with other databases and data sources such as environmental site locations (e.g. oil wells, land fills, etc.) and news feeds using an ontology to mediate different database constructs and datasets. The third component provides user interfaces and query tools. The architecture approach is to be non-invasive, so for example changes to the AHLTA system are not required, all we need is access to the knowledgebase.