Research Publications


Publication Profile

Search for Complex Queries

Study after 2010, done in the US, Hypertension treatment in people over 40 years old, Hispanics, who have type II diabetes.

CHiPS™ for Health Care: Complex search queries made simple

Customer Scenario

Find healthcare publications relevant to a particular case with complex biomedical & socio-demographic criteria. Populate treatment efficency templates from clinical trials across a varying array of people of specific demographics and medical problems, and the treatments that they underwent, as well as the outcome of those treatments.


In typical document retrieval, the more specific your search need, the more general your search results are. Traditional search systems find the documents containing simple keywords. More specific queries contain more keywords, and thus more documents are retrieved. The actual meaning of the query gets lost.

Lymba's Solution

Collaborative High Precision Search (CHiPS) - Lymba’s intelligent search system - allows to formulate long complex queries in plain English and takes into account synonyms and inference to provide relevant results. CHiPS uses K-Extractor for deep semantic analysis of biomedical publications, extracting important medical and socio-demographic concepts as well as relations between them. The search query undergoes the same semantic analysis to recognize expressed criteria. The criteria are semantically matched to information in the publications to yield high precision results.


Unstructured data: The last 10 years of MEDLINE®/PubMed®'s biomedical journal citation data, which comprises over 25 million articles from clinical trials, life science journals, and online books.

Query: Lymba provides an interactive user interface as well as an Application Program Interface (API) to allow the customer to perform detailed natural language queries, or to use an entire document as a query.


Sorted list of search results based on semantic similarity. Each result is paired with a result profile which allows you to, at a glance, determine the most relevant documents and why they are relevant before opening.

Key Features


K-Extractor recognizes concepts from existing medical lexical resources including UMLS Metathesaurus, SNOMED-CT, Medline (health signs, symptoms, diseases, medication, etc) as well as socio-demographic characteristics and relations (race, gender, age, nationality, insurance, financial status, education level) and semantic relations between these concepts.

Big Data:

Extracted information is efficiently stored to allow fast access. For this use case the system supports real-time querying across several multimillion document collections all on a single PC.

Document Comparison View:

CHiPS doesn't just surface sorted results, it can also explain the semantic matches it finds.

Check out CHiPS for more details on the product and its features.