Intelligent search tool for answering clinical queries

Abstract

A scalable infrastructure for searching multiple disparate textual databases by mapping their contents onto a structured ontology, e.g., of medical concepts. This framework can be leveraged against any database where free-text attributes are used to describe the constituent records.

Having described our invention, what we now claim is as follows: 
     
         1 . In an information and retrieval system, the improvement comprising:
 a processor; and   computer memory storing computer program instructions executed by the processor, the computer program instructions comprising:
 program code responsive to receipt of a search query to normalize the search query using a natural language processing (NLP) mapping to generate a normalized search query, the NLP mapping normalizing text contents of a set of documents onto a structured ontology, the text contents in the set of documents representing at least one concept in the structured ontology having a set of subordinate concepts that are described by synonymous, but textually disparate phrases; and 
 program code to process the normalized search query against the structured ontology to identify documents associated with at least one concept, the processing of the normalized search query associated with the concept also returning results associated with the subordinate concepts. 
   
     
     
         2 . The improved information and retrieval system as described in  claim 1  wherein the structured ontology represents a hierarchical representation of medical knowledge. 
     
     
         3 . The improved information and retrieval system as described in  claim 1  wherein the program code to normalize maps text contents of the set of documents to a controlled vocabulary. 
     
     
         4 . The improved information and retrieval system as described in  claim 1  wherein the structured ontology is stored in a hierarchical database. 
     
     
         5 . The improved information and retrieval system as described in  claim 4  wherein a particular concept in the structured ontology is represented in the hierarchical database by a vertex, and wherein a reference to an entity associated with the particular concept is stored at the vertex. 
     
     
         6 . The improved information and retrieval system as described in  claim 5  wherein the program code to process the normalized search query implements a recursive graph-based algorithm. 
     
     
         7 . The improved information and retrieval system as described in  claim 5  wherein the program code to process the normalized search query searches first and second portions of the hierarchical database concurrently. 
     
     
         8 . The improved information and retrieval system as described in  claim 7  wherein the first and second portions of the hierarchical database are associated with first and second machines. 
     
     
         9 . The improved information and retrieval system as described in  claim 8  wherein the first and second machines are located remotely from one another.