Back to results
US 2015/0317446 A1Published application
Health forecaster
AssigneeCerner Innovation, Inc.
InventorsMichael A. Ash, Todd Jeffrey Reynolds, Harlen Hays
Life of this patent
Estimating the term…
Abstract
Methods, computer systems, and computer-readable storage media are provided for utilizing health-related variables to generate a current health state of a user along with a most-likely future health state of the user and an ideal future health state of the user. Graphical representations of these health states are generated and presented on a graphical user interface. The user is able to interact with the graphical representations to see the impact of the user's choices on the user's future health.
Claims20
One or more computer memory having embodied thereon computer executable instructions that, when executed by a server, perform a method of using health-related variables associated with a user having a health condition to generate current and future health states of the user, the method comprising:
- receiving structured and unstructured health-related data associated with the user having the health condition, the structured and unstructured health-related data comprising data over a past period of time up to a current period of time, the structured and unstructured health-related data comprising data received from at least one of personal health risk assessments, biometric screenings, and health-related insurance claims associated with the user;
- codifying the structured and unstructured health-related data into a set of codified concepts of standard nomenclature;
- utilizing a principal component analysis (PCA), reducing the set of discrete concepts of standard nomenclature to a finite set of factors that represent the current and future health states of the user, wherein each factor in the finite set of factors is a rotated linear weighted combination of the set of codified concepts;
- utilizing the PCA:
- (A) generating at least a first health score for the each factor in the finite set of factors, wherein the at least first health score comprises a percentile comparison of the user's current health state to health states of people who share one or more demographic traits with the user,
- (B) generating at least a second health score for the each factor in the finite set of factors, wherein the at least second health score represents the future health state of the user, further wherein the at least second health score comprises a percentile comparison of the user's future health state to health states of the people who share one or more demographic traits with the user, and wherein the at least second health score is determined by applying regression analysis to the finite set of factors;
- and
- generating a set of goals related to the user for achieving an ideal future health state.
5 more elements
A computing system for performing a method for using health-related variables associated with a user having a health condition to generate current and future health states of the user, the system comprising:
- one or more computer processors;
- at least one computer server;
- one or more computer memory storage devices having embodied thereon computer executable instructions that, when executed by the at least one server, perform the method for using health-related variables associated with a user having a health condition to generate current and future health states of the user;
- a receiving component that receives structured and unstructured health-related data associated with the user having the health condition, the structured and unstructured health-related data comprising data over a past period of time up to a current period of time, the structured and unstructured health-related data comprising data received from at least one of personal health risk assessments, biometric screenings, and health-related insurance claims associated with the user, and codifies the structured and unstructured health-related data into a set of codified concepts of standard nomenclature;
- a computation component that uses the codified concepts in a principal component analysis (PCA) to generate:
- (A) at least a first health score that represents the current health state of the user, wherein the at least the first health score comprises a percentile comparison of the user's current health state to health states of people who share one or more demographic traits with the user,
- (B) at least a second health score that represents the future health state of the user, wherein the at least the second health score comprises a percentile comparison of the user's future health state to health states of the people who share one or more demographic traits with the user, and wherein the at least the second health score is determined by applying regression analysis to the set of codified concepts of standard nomenclature;
- and
- a rendering component that generates a first graphical representation of the current health state of the user and generates a second graphical representation of the future health state of the user, and renders the first and second graphical representations on a graphical user interface (GUI).
6 more elements
A computer implemented method for using health-related variables associated with a user having a health condition to generate current and future health states of the user, the method comprising:
- receiving structured and unstructured health-related data associated with the user having the health condition, the structured and unstructured health-related data comprising data over a past period of time up to a current period of time, the structured and unstructured health-related data comprising data received from at least one of personal health risk assessments, biometric screenings, and health-related insurance claims associated with the user;
- codifying the structured and unstructured health-related data into a set of codified concepts of standard nomenclature;
- utilizing a principal component analysis (PCA), reducing the set of discrete concepts of standard nomenclature to a finite set of factors that represent the current and future health states of the user, wherein each factor in the finite set of factors is a rotated linear weighted combination of the set of codified concepts;
- utilizing the PCA:
- (A) generating at least a first health score for the each factor in the finite set of factors, wherein the at least first health score comprises a percentile comparison of the user's current health state to health states of people who share one or more demographic traits with the user,
- (B) generating at least a second health score for the each factor in the finite set of factors, wherein the at least second health score represents the future health state of the user, further wherein the at least second health score comprises a percentile comparison of the user's future health state to health states of the people who share one or more demographic traits with the user, and wherein the at least second health score is determined by applying regression analysis to the finite set of factors;
- and
- generating a set of goals related to the user for achieving an ideal future health state.
5 more elements
Dependencies are inferred from the claim text.