PREDICTING GLUCOSE TRENDS FOR POPULATION MANAGEMENT

• US 20160180040A1
• Filed: 12/23/2014
• Published: 06/23/2016
• Est. Priority Date: 12/23/2014
• Status: Active Grant

• Current Assignee: Cerner Innovation, Inc. (Oracle Corporation)
• Original Assignee: Cerner Innovation, Inc. (Oracle Corporation)
• Inventors: QUICK, MEGAN KATHLEEN, CROUGH, DANIEL CRAIG, RYAN, HUGH H.
• Granted Patent: US 10,120,979 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0635 by changing the path, e.g. traffic rerouting, path reconfiguration
• CPC Class Codes: G06F 3/0659 Command handling arrangements, e.g. command buffers, queues, command scheduling
• CPC Class Codes: G06F 3/0685 Hybrid storage combining heterogeneous device types, e.g. hierarchical storage, hybrid arrays
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

Computerized systems and methods facilitate preventing dangerous blood glucose levels using a predictive model to predict whether a particular patient is trending to have dangerous blood glucose levels. The predictive model may be built using logistic or linear regression models incorporating glucose data associated with a plurality of patients received from a plurality of sources. The glucose data may include context data and demographic data associated with the glucose data and the plurality of patients. The predictive model may be employed to predict a likelihood of a particular patient to have dangerous blood glucose levels. Based on the likelihood, the prediction and one or more interventions are communicated to a care team or the patient. The one or more interventions may be incorporated into a clinical device workflow associated with a clinician on the care team or the patient.

Claims

1. One or more computer storage media storing computer-useable instructions that, when used by one or more computing devices, cause the one or more computing devices to perform operations comprising:

receiving glucose data at a prediction server, the glucose data received from a plurality of sources including an electronic medical record associated with a patient, one or more care facilities, one or more laboratories, or one or more integrated home devices;

determining, based on the glucose data, a real-time prediction indicating whether the patient is trending to have dangerous blood glucose levels; and

communicating, based on the determining, the real-time prediction and one or more interventions to a care team and the patient.

2. The one or more computer storage media of claim 1, further comprising receiving additional data from the care team and the patient.

3. The one or more computer storage media of claim 2, further comprising utilizing the additional data to update the real-time prediction and the interventions.

4. The one or more computer storage media of claim 1, wherein the real-time prediction is based on a prediction model using logistic or linear regression models.

5. The one or more computer storage media of claim 2, wherein the prediction model predicts whether the patient is likely to have abnormally high or abnormally low glucose levels.

6. The one or more computer storage media of claim 3, wherein the prediction model that predicts whether the patient is likely to have abnormally high glucose levels leverages the glucose data including:

a most recent glucose lab value;

a month, a time, and a year the most recent glucose lab value was drawn;

whether the most recent glucose lab value was drawn on a weekend; and

an age, marital status, and race of the patient.

7. The one or more computer storage media of claim 3, wherein the prediction model that predicts whether the patient is likely to have abnormally low glucose levels leverages the glucose data including:

a most recent glucose lab value;

a month, a time, and a year the most recent glucose lab value was drawn;

whether the most recent glucose lab value was drawn on a weekend; and

an age and race of the patient;

geographic region where the most recent lab value was drawn; and

whether the most recent lab value was drawn at a teaching facility.

8. The one or more computer storage media of claim 1, further comprising incorporating the interventions into a workflow provided on a clinical device associated with a clinician on the care team.

9. The one or more computer storage media of claim 1, wherein the interventions include specific recommended actions.

10. The one or more computer storage media of claim 2, wherein the additional data includes data received from the patient in an electronic questionnaire communicated to the patient as at least a portion of the intervention.

11. The one or more computer storage media of claim 1, further comprising automatically adjusting medication levels associated with the one or more integrated home devices.

12. A computer-implemented method in a clinical computing environment comprising:

receiving, via a first computing process, glucose data for a plurality of patients from a plurality of sources, the glucose data including context data and demographic data associated with the glucose data and the plurality of patients;

generating, via a second computing process, a predictive model based on the glucose data using one or more logistic or linear regression models;

employing, via a third computing process, the predictive model to predict a likelihood of a particular patient to have dangerous blood glucose levels;

communicating, via a fourth computing process, a prediction and one or more interventions to a care team and the patient based on the likelihood; and

incorporating, via a fifth computing process, the one or more interventions into a clinical device workflow associated with a clinician on the care team;

wherein each of the computing processes is performed by one or more computing devices.

13. The method of claim 12, wherein the glucose data associated with the predictive model to predict a likelihood of a particular patient to have abnormally high glucose levels includes:

a most recent glucose lab value;

a month, a time, and a year the most recent glucose lab value was drawn;

whether the most recent glucose lab value was drawn on a weekend; and

an age, marital status, and a race.

14. The method of claim 12, wherein the glucose data associated with the predictive model to predict a likelihood of a particular patient to have abnormally low glucose levels includes:

a most recent glucose lab value;

a month, a time, and a year the most recent glucose lab value was drawn;

whether the most recent glucose lab value was drawn on a weekend; and

an age and race of the patient;

geographic region where the most recent lab value was drawn; and

whether the most recent lab value was drawn at a teaching facility.

15. The method of claim 12, further comprising receiving, via a sixth computing process, additional data from the care team and the patient.

16. The method of claim 15, further comprising updating, via a seventh computing process, the real-time prediction and clinical device workflow based on the additional data.

17. The method of claim 12, further comprising automatically adjusting, via an eight computing process, medication levels associated with one or more integrated home devices.

18. The method of claim 12, further comprising communicating, via a ninth computing process, an electronic questionnaire to the patient in association with the one or more interventions.

20. The system of claim 16, wherein the instructions further cause the one or more processors to:

determine a context based predictive model of the one or more predictive models to employ for the particular patient.

19. A system comprising:

a glucose database storing glucose data received for a plurality of patients from a plurality of sources, the glucose data including context data and demographic data associated with the glucose data and the plurality of patients;

one or more processors; and

one or more computer storage media storing instructions that, when used by the one or more processors, cause the one or more processors to;

generate one or more predictive models based on the glucose data using one or more logistic or linear regression models;

employ the one or more predictive models to predict a likelihood of a particular patient to have dangerous blood glucose levels; and

communicate a prediction and one or more interventions to a care team or the particular patient based on the likelihood.