Method and system for extracting data from a plurality of electronic data stores of patient data to provide provider and patient data similarity scoring

Inventors

Naeymi-Rad, Frank • Cardwell, Matthew • Decaro, Michael • Paul, Erina • Wang, Yiqing • Thompson, James • Kanter, Andrew

Assignees

Intelligent Medical Objects Inc

Abstract

A system and method for extracting data from an electronic health record to provide provider and patient data similarity scoring includes: encoding a problem list for a plurality of patients with concepts from a common electronic health record ontology. In one aspect, the patients have electronic health records maintained by a plurality of providers. The system and method then may parse the concepts into a plurality of clusters or categories and determining, for each of the providers, a total number of patients that have at least one problem in a cluster or category or determining, for each patient, which of the plurality of clusters or categories correspond to at least one concept encoded in the patient's problem list. The system and method then may calculate for each pair of providers or patients, a distance between the providers or patients.

Core Innovation

The invention provides a method for extracting data from a plurality of electronic data repositories to provide provider and patient data similarity scoring. It encodes a plurality of electronic problem lists stored using one or more electronic health record ontologies so that the problem lists are used with a respective electronic health record software package, and the problem lists are associated with patients and providers maintained by a plurality of providers, with each patient attributable to a respective provider.

The method parses the concepts from the encoded problem lists into clusters or categories of concept groupings. It determines, for each provider and for all of the problems encoded to a respective concept within a cluster or category, a total number of patients that have at least one problem in the cluster or category, and associates a patient to the cluster or category according to whether the patient has at least one problem in the cluster or category, iterating the determining step for each remaining cluster or category.

The method calculates, for each pair of providers, a distance between the providers using the results of the determining and iterating steps. It generates, on a display of a computer, a first user interface arranging at least a subset of the providers into one or more groups reflected by the plurality of clusters or categories, with a visual representation of a closest match based on the distance between the providers and a selectable second user interface depicting a selected provider with one or more other providers calculated to be closest and visual representations of a degree of closeness.

Claims Coverage

The claim coverage includes 9 inventive features across the independent claim and dependent claims: ontology-based encoding, concept clustering, provider-level patient counting, pairwise provider distance calculation, grouped similarity user interfaces, Minkowski distance of order 2, normalization by provider patient totals, common ontology as interface terminology, and SNOMED CT or ICD / exact problem list match inputs.

Overall, the claims cover ontology-encoded problem lists tied to providers, concept clustering into clusters or categories, provider-level patient counting within each cluster or category, pairwise provider distance calculation, and grouped user interfaces that present closest matches and selectable comparison views. Dependent claims further specify the distance metric, normalization, ontology implementation, named ontology options, and exact problem list match inputs.

Stated Advantages

Documented Applications