Researchers, informatics experts, doctors and insurers are faced with unique, data related challenges every day. They have to find precise sets of information in huge amount of information usually scattered across disparate data sources. None of the complex terminology is semantically indexed or connected. This results in lost time and less accurate research or care.
Ontotext Insight Platform dramatically reduces the time it takes to discover relevant content for a research task. The platform is an unique blend components that make possible linked open data integration, generation of knowledge base of terms, contextual ontology based text mining and semantic search.
AstraZeneca is a global research-based bio-pharmaceutical company with skills and resources focused on discovering, developing and marketing medicines for some of the world’s most serious illnesses, including cancer, heart disease, neurological disorders such as schizophrenia, respiratory disease and infection.
The platform offers all required component to build a system that will be capable to analyze huge amount of unstructured information and extract useful knowledge locked in the documents; to semantically integrate it together with relevant public and in-house generated data; to provide multiple mechanisms to explore it as a Linked Data.
The Ontotext Insights Platform is successfully applied for identification of relevant archived clinical study documents and unlocking the hidden knowledge locked in them, thus reducing significantly the effort and time required to answer re-occurring regulatory information requests, see Pharma Insights.
The Insights Platform was used to identify and classify medical observations, clinical procedures and historical events in narratives of patients health records and other medical documents, see Patients Insights.
The Insights Platform was successfully applied in monitoring of the major health regulatory recommendation documents; formalizing the recommendations in structured form that could be used for the definition of medical validity check rules that could be used for detection of inconsistencies in health insurance claim, see Health Insurance Insights.
Semantic annotation pipelines supporting more than 100 different semantic types to extract meaning from scientific research, clinical or patients’ medical documents and health insurance claims. Pre-integrated industry adopted biomedical ontologies like SNOMED, LOINC, RxNorm and UMLS. Experts validated Linked Data creation and maintenance process, which makes possible building an enterprise Linked Data cloud from both public and proprietary data sets.
Linked Data Creation
Automatically recognize complex biomedical terms using tested “bio-medical tagging algorithms.” Proven methodology for schema and instance mappings for development of fully functional Linked Data
Linked Data Repository
Semantic Indexing and Search
Out-of-the-box semantic search capability with easily configurable faceted search defined over all important semantic types in the repository. Development tools which help Information Extraction experts to identify important patterns in the unstructured text and to define their own complex rules for extraction of valuable information from text.