System and method for clinical trial candidate matching
Inventors
Labkoff, Steven E. • Perkowitz, Marc • Masarie, JR., Fred E. • McGinness, Doris J. • Wang, Amy Y. • Koyyalamudi, Krishna Chaitanya
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-10878010-B2
Publication Date
2020-12-29
Expiration Date
Abstract
A computer-implemented system and method for identifying potential clinical trial participants from one or more databases of patient electronic health information includes analyzing the clinical trial requirements and mapping those requirements to an interface terminology, where concepts of the interface terminology include a key concept and a group of one or more additional concepts that are related in the context of the clinical trial. The method further includes mapping the patient electronic health information to the interface terminology, building a query of one or more interface terminology elements; analyzing the patient health information for matches to the one or more interface terminology elements and, if necessary, iterating the process by adding additional interface terminology elements to the query.
Core Innovation
The invention relates to a computer-implemented system and method for identifying potential clinical trial participants by matching clinical trial inclusion criterion and exclusion criterion information to patient electronic health record information. A data file including data pertaining to a clinical trial is received by a computer, where the data includes at least one inclusion criterion and at least one exclusion criterion. The method determines a key concept of an interface terminology ontology pertaining semantically to the data, where the interface terminology ontology comprises a plurality of concepts with corresponding codes.
The method identifies one or more entries of a reference or administrative code set relating to the key concept in the context of the clinical trial and maps one or more interface terminology concepts to the reference or administrative code set entries. The invention emphasizes structuring clinical-trial criteria in a way that is computable and then coding and mapping the interface terminology ontology concepts to the reference or administrative code sets, using semantic grouping in the context of the clinical trial. The matching is represented as direct match to the key concept or a match to one or more interface terminology concepts mapped to the one or more reference or administrative code set entries.
The invention analyzes anonymized patient electronic health record information for matches by building and executing Boolean queries, where the analyzing step comprises building a query including the key concept and one or more interface terminology concepts and comparing the database against the query. In implementations, the query comprises a SQL statement including a call to an XML function that executes one or more Boolean operators acting on the inclusion criterion and the exclusion criterion, and the analysis can operate across databases from different institutions. The method aggregates and returns matched patient information to a user via a graphical user interface, and it can include iterative query expansion by determining additional interface terminology ontology concepts relating to the key concept.
Claims Coverage
The partial content provides three independent claims (clm-00001, clm-00006, clm-00012). Across these claims, the inventive coverage centers on mapping clinical trial inclusion/exclusion criteria to interface terminology ontology concepts, mapping those concepts to reference or administrative code sets or ontology elements, and using SQL with an XML function executing Boolean operators to analyze anonymized patient EHR databases and present matching patients via a graphical user interface.
Key concept mapping and inclusion/exclusion matching across institutions
Receiving a data file including data pertaining to a clinical trial with at least one of an inclusion criterion and an exclusion criterion; determining a key concept of an interface terminology ontology pertaining semantically to the data; identifying one or more entries of a reference or administrative code set relating to the key concept; mapping one or more interface terminology concepts to the one or more reference or administrative code set entries; analyzing anonymized patient electronic health record information from separate institutions for matches based on whether the key concept pertains to an inclusion criterion or exclusion criterion; aggregating and returning matched patients to a user; direct match to the key concept or a match to mapped interface terminology concepts.
Interface terminology ontology concept mapping with GUI presentation using Boolean SQL/XML query
Receiving a data file including data pertaining to a clinical trial with at least one inclusion criterion and at least one exclusion criterion; mapping one or more concepts of an interface terminology ontology to each of the inclusion criterion and the exclusion criterion; determining a key concept of the interface terminology ontology pertaining to the data; identifying one or more concepts of the interface terminology ontology relating to the key concept in the context of the clinical trial; analyzing a database of patient electronic health record information for a match to the key concept or related interface terminology concepts; presenting information relating to a matched patient via a graphical user interface; building a query including the key concept and the related concepts; the query comprises a SQL statement including a call to an XML function that executes one or more Boolean operators acting on the inclusion criterion and the exclusion criterion.
Ontology element mapping with SQL/XML Boolean query over inclusion and exclusion and GUI presentation
Receiving a data file including data pertaining to a clinical trial including at least one inclusion criterion and at least one exclusion criterion; mapping one or more concepts of an interface terminology ontology to each of the inclusion criterion and the exclusion criterion; mapping elements of a database of patient electronic health record information to one or more interface terminology ontology elements; building a SQL statement including a call to an XML function that executes one or more Boolean operators acting on the inclusion criterion and the exclusion criterion; determining which elements of the interface terminology ontology map to the inclusion criterion and the exclusion criterion; analyzing the database of patient electronic health record information for a match to the determined elements; presenting to a user information relating to a patient to which the match pertains via a graphical user interface.
Across the independent claims, the central inventive elements are determining and mapping a key concept of an interface terminology ontology to clinical trial inclusion/exclusion criteria, mapping interface terminology ontology concepts to reference or administrative code set entries or to interface terminology ontology elements, analyzing anonymized patient electronic health record databases by building a SQL statement with an XML function executing Boolean operators acting on inclusion and exclusion criteria, and presenting matched patient information via a graphical user interface. One independent claim further extends the matching to separate databases from different electronic health record providers and aggregates the results.
Stated Advantages
Improve match accuracy versus direct interface-concept mapping.
Supports matching for potential clinical trial participants using anonymized patient electronic health record information from different institutions/electronic health record providers.
Enables computable querying by structuring and coding/mapping interface terminology and reference/administrative code set concepts into a queryable format.
Supports iterative query expansion by determining additional interface terminology ontology concepts relating to the key concept.
Documented Applications
Matching clinical trial inclusion/exclusion criteria to patient EHR information (including anonymized patient EHR information) for identifying potential clinical trial participants.
Outbreak or geographic hotspot analysis using matched patient information, including use of a heat map.
A large-scale EHR dataset use case where key-first iterative querying improves results.
Interested in licensing this patent?