How it works: Treato technology
Years of intense research and development efforts have been invested in creating Treato's technology that identifies the billions of relevant patient discussions and distills their
intelligence in real-time.
Understanding patient language
Healthcare is one of the most complex and sensitive fields discussed online. Patients have their own way of expressing how they feel, using their own words, rather than medical terminology.
Treato’s proprietary algorithms turn the unstructured patient discussions found on the Web into actionable insights. With multiple patents pending, Treato’s intelligent Natural Language
Processing goes far beyond traditional sentiment analysis. The system can differentiate between actual patient experiences and hearsay, rumors, or observations, enabling Treato to truly embody
its mission of discerning the voice of the patient.
Comprehensive medical ontology
Treato has brought together multiple medical
dictionaries, organized them under a new hierarchy, and added its proprietary patient language vocabulary. This enormous knowledge base consists of a terminology of over one million medical terms and consumer-friendly
synonyms, as well as many hundreds of thousands of medical concepts, the most comprehensive medical ontology.
"Big data" infrastructure
Scalability is a cornerstone of Treato technology, since health-related user-generated content is spread across hundreds of thousands of medical and healthcare sites which are growing daily. To
crawl, parse and analyze such extensive amounts of data, Treato is based on a state-of-the-art, open-source Hadoop distributed computational platform. Built as a result of extensive research and
proofs of concept, the fully integrated system includes HBase and the Solr search platform over Lucene, enabling the infrastructure to provide real-time coverage of the continuously increasing
volumes of health-related data on the Web.