Linked Data Integration for Healthcare

Ontotext Healthcare Insights helps you:

Interoperate across systems

By applying common clinical terminologies and standards.

Integrate newly generated patient knowledge

With external systems like clinical decision support systems or automatic clinical trial matching.

Develop new predictive models

Based on the newly extracted knowledge.

Find the Hidden Data in Patient Records

Ontotext Healthcare Insights breaks the productivity barrier transforming the raw patient data into structured knowledge. HealthCare Insights is designed to process large volumes of patient records, extracting and semantically indexing data about patient diagnoses, treatments, medications and events timing. In order to semantically fuse the extracted data with the available structured knowledge, the modeling of identified facts is compliant with the industry standards and accurately coded with the approved ontologies.

Unstructured patient data is enriched by applying medical ontologies (like SNOMED, LOINC, RxNorm and UMLS) and clinical terminologies for identification of entities like generic and branded drugs, recommended and prescribed dosage, adverse event reactions, contradictions, and more. The standard terminology based extraction is backed up with negation and temporality identification, application of flexible gazetteers to and definition of post-coordination patterns to capture the full meaning of the described medical observations.

All extracted medical data is normalized to resolvable instances from the Healthcare Insights Knowledge Base. The extracted information is semantically ready to be fused with the Linked Data generated from multiple reference public data set (covering disease and symptoms, anatomical structures, generic drugs and products, and many more). The “operational data” extracted from patient medical documents describe different entities, like diagnoses and conditions, treatments, medical procedures and test, different measurements – all recorded in the context of a particular patient, document and time period. The combination of real evidence patient data with the structured background knowledge allows the identifications trends and patterns that may have previously gone undetected. Such structured approach allows for integration with external data sources like DailyMed and ClinicalTrials.gov for even greater insight.

Overview

Healthcare Insights relies on industry standards and approved ontologies. The platform analyse large volumes of patient records and extract semantically structured information about diagnoses, treatments, medications and events. Healthcare Insights semantically fuse the extracted facts with the available structured knowledge and provides innovative approaches for semantic data exploration and search.

Overview

Knowledgebase

Connect to clinical linked data cloud or integrate with an existing terminology server.

Knowledgebase

Raw data

Process patient narrative information using text mining pipelines.
Ontology based text analysis is used to to detect concepts and disambiguate meaning using biomedical ontologies using the background knowledge from the clinical linked data cloud.

Raw_data

RDF Data

Extract in structured format diagnosis, medication and other important clinical data. The extracted facts are modelled in RDF, using the concept original identifiers provided by the ontologies.

RDF_Modeling

Data Fusion

The extracted and normalized information is loaded in the knowledgebase. The data is semantically fused with all the background knowledge available in the semantic repository, where could be used for definition of new inference rules and generation of new insights.

Data_Fusion

Semantic Search

Develop custom patient classifications specific for the different therapeutic areas using a high-level semantic query language.

Semantic_search

Features

Ontology Based Text Mining

Benefits

  • Eliminates ambiguity within the data.
  • Analyzes and categorizes documents based on a common classification system.
  • Provides researchers with actionable intelligence.

Semantic Enrichment

  • Tags your data to help you identify relationships between entities.
  • Enhances your knowledge base with related concepts and terms leading to more complete search results.

Knowledge Base

  • Allows for the inexpensive integration of internal data with publicly available open data.
  • Pre-populates your data with pharmaceutical products including drug indications and contraindications.
  • Stores all application concepts, their text definition, labels and full semantic context.
  • Reduces data integration between clinical decision support or automatic clinical trial matching services, for example.

Semantic Search

  • Save time locating patient data.
  • Retrieve accurate, relevant results.
  • Reduce the time required to identify trends in patient data.
  • Guarantee credibility of results by using tools and quality assessments established with scientific methods.
Back to top