Systems and methods for constructing clinical pathways within a GUI

• US 10,692,254 B2
• Filed: 03/02/2018
• Issued: 06/23/2020
• Est. Priority Date: 03/02/2018
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Yaeli, Avi, Bak, Peter, Perer, Adam, Segal, Shay
• Primary Examiner(s): Khan, Shahid K
• Application Number:
  US15/910,379
• Publication Number:
  US 20190272654A1
• Time in Patent Office: 844 Days
• Field of Search: None
• CPC Class Codes: G06F 16/9024 Graphs; Linked lists G06F16/9027 takes precedence
• CPC Class Codes: G06F 3/04847 Interaction techniques to control parameter settings, e.g. interaction with sliders or dials
• CPC Class Codes: G06F 9/451 Execution arrangements for user interfaces
• CPC Class Codes: G06T 11/206 Drawing of charts or graphs
• CPC Class Codes: G06T 2200/24 involving graphical user interfaces [GUIs]

Abstract

There is provided a method of creating a cohort clinical pathway graph based on knowledge-driven manual user input and automated data-driven mining comprising: receiving via a graphical user interface (GUI), manual selections including: knowledge-driven variable(s) denoting clinically significant values representing elements of a clinical decision making process, and an anchoring location of each knowledge-driven node denoting a respective knowledge-drive variable within a directed acyclic graph (DAG), computing individual clinical pathways for each of the sampled population of patients by automatically computing data-driven nodes denoting the data-driven discovery of event types relative to the manual selections, and aggregating the individual clinical pathways to compute a cohort clinical pathway DAG, wherein the cohort clinical pathway DAG includes nodes comprising the knowledge-driven nodes, the data-driven nodes, and links connecting the nodes, each link denoting an automatically discovered sequence between two respective nodes, and presenting the cohort clinical pathway DAG within the GUI.

Claims

1. A computer-implemented method of creating a cohort clinical pathway graph based on knowledge-driven manual user input and automated data-driven mining within a graphical user interface (GUI) presented on a display associated with a computing device, the method comprising:

receiving, via the GUI a plurality of manual selections including;

at least one knowledge-driven variable denoting clinically significant values representing elements of a clinical decision making process, and

an anchoring location of each knowledge-driven node denoting a respective knowledge-drive variable within a directed acyclic graph (DAG);

computing individual clinical pathways for each of the sampled population of patients by automatically computing data-driven nodes denoting the data-driven discovery of event types relative to the plurality of manual selections; and

aggregating the individual clinical pathways to compute a cohort clinical pathway DAG, wherein the cohort clinical pathway DAG includes a plurality of nodes comprising the knowledge-driven nodes, the data-driven nodes, and links connecting the plurality of nodes, each link denoting an automatically discovered sequence between two respective nodes;

computing a depth for each node of the plurality of nodes; and

presenting, within the GUI, an arrangement of clusters of nodes, wherein each cluster includes nodes of the plurality of nodes having a common depth.

2. The method according to claim 1, wherein a Sankey-based diagram is computed according to the cohort clinical pathway DAG, and presented within the GUI.

3. The method according to claim 1, wherein a height of a respective node and/or link of the cohort clinical pathway DAG corresponds to a relative number of sub-set of patients of the sampled population having individual pathways passing through the respective node and/or link.

4. The method according to claim 1, wherein the links are color coded according to a color coding scheme denoting the relative ratio of the sampled population meeting a certain clinical outcome value of the clinical outcome variable.

5. The method according to claim 1, further comprising, receiving, via the GUI one or both of the following manual selections:

at least one clinical state parameter denoting an initial clinical state of a sample population of patients;

at least one clinical outcome parameter denoting a final state of at least one patient of the sample population;

wherein the at least one clinical state parameter denotes an initial root node of the DAG, and the at least one outcome parameter denotes a final leaf node of the DAG,

wherein the plurality of nodes of the cohort clinical pathway DAG comprise the initial root node, and the final leaf node.

6. The method according to claim 5, wherein the at least one clinical state parameter, the at least one clinical outcome parameter, and the at least one knowledge-driven variable are denoted as one or more conditions of values of entries stored in electronic health records of patients.

7. The method according to claim 5, wherein the at least one clinical state parameter, the at least one clinical outcome parameter, and the at least one knowledge-drive variable are computed based on data stored in one or more of:

electronic health records of a population of patients, medical insurance claims of the population of patients, and billing data of the population of patients.

8. The method according to claim 1, further comprising

presenting within the GUI, the links between the plurality of nodes at each cluster,

wherein computing a depth for each node of the plurality of nodes includes traversing the cohort clinical pathway DAG depth first.

9. The method according to claim 1, wherein the computed depth includes one or both of:

a fixed depth value indicative of the last knowledge-driven node encountered in the cohort clinical pathway DAG for the respective node, and a variable depth value indicative of depth between knowledge-driven nodes.

10. The method according to claim 1, wherein intermediate nodes of the cohort clinical pathway DAG denote the data-driven discovered event types, and links between nodes denote data-driven discovered sequences between event types denoted by the linked nodes.

11. The method according to claim 1, wherein links and/or nodes of the cohort clinical pathway DAG are associated with statistics computed based on the individual pathways from which the links and/or nodes were aggregated from.

12. The method according to claim 11, wherein the statistics include at least one of:

number of individual patients of the sampled population associated with the respective link and/or node, and a measure of central tendency computed for a value associated with the respective link and/or node associated with the sub-set of patients.

13. The method according to claim 1, wherein the aggregating comprises:

computing unique sequences of nodes based on the individual clinical pathways;

computing the number of patients for each discovered unique sequence; and

aggregating patient level metrics for each unique sequence.

14. The method according to claim 1, further comprising receiving, via the GUI, a manual designation indicative of automated data-driven discovery of events at least one of:

before, after, and in-between knowledge-driven nodes of the at least one knowledge-driven variables, wherein the automatically computing the data-driven discovery of events is performed according to the manual designation.

15. The method according to claim 1, further comprising, receiving via the GUI, a manual collapse instruction, and aggregating the data-driven event type nodes of the cohort clinical pathway DAG to create a collapsed cohort clinical pathway DAG depicting direct flows between the knowledge-driven nodes.

16. The method according to claim 1, wherein nodes of individual clinical pathway are associated with timestamps indicative of time of occurrence of the event represented by the respective node, wherein the individual clinical pathway are aggregated to compute the cohort clinical pathway DAG according to the timestamps.

17. The method according to claim 1, wherein the data-driven discovery of events is implemented based on temporal sequence mining of electronic health records of the sample population of patients.

18. The method according to claim 1, wherein data-driven nodes represent data selected from the group comprising:

a medical drug-based treatment, a medical intervention procedure, and a certain clinical state.

19. A system for creating a cohort clinical pathway graph based on knowledge-driven manual user input and automated data-driven mining within a graphical user interface (GUI) presented on a display associated with a computing device, comprising:

a non-transitory memory having stored thereon a code for execution by at least one hardware processor of the computing device, the code comprising;

code for receiving, via the GUI a plurality of manual selections including;

at least one knowledge-driven variable denoting clinically significant values representing elements of a clinical decision making process, and

an anchoring location of each knowledge-driven node denoting a respective knowledge-drive variable within a directed acyclic graph (DAG);

code for computing individual clinical pathways for each of the sampled population of patients by automatically computing data-driven nodes denoting the data-driven discovery of event types relative to the plurality of manual selections; and

code for aggregating the individual clinical pathways to compute a cohort clinical pathway DAG, wherein the cohort clinical pathway DAG includes a plurality of nodes comprising the knowledge-driven nodes, the data-driven nodes, and links connecting the plurality of nodes, each link denoting an automatically discovered sequence between two respective nodes;

code for receiving, via the GUI, a manual collapse instruction, and, in response to receiving the manual collapse instruction, aggregating the data-driven event type nodes of the cohort clinical pathway DAG to create a collapsed cohort clinical pathway DAG depicting direct flows between the knowledge-driven nodes; and

code for presenting the collapsed cohort clinical pathway DAG within the GUI.

20. A computer program product for creating a cohort clinical pathway graph based on knowledge-driven manual user input and automated data-driven mining within a graphical user interface (GUI) presented on a display associated with a computing device, comprising:

a non-transitory memory having stored thereon a code for execution by at least one hardware processor of the computing device, the code comprising;

instructions for receiving, via the GUI a plurality of manual selections including;

at least one knowledge-driven variable denoting clinically significant values representing elements of a clinical decision making process, and

an anchoring location of each knowledge-driven node denoting a respective knowledge-drive variable within a directed acyclic graph (DAG);

instructions for computing individual clinical pathways for each of the sampled population of patients by automatically computing data-driven nodes denoting the data-driven discovery of event types relative to the plurality of manual selections; and

instructions for aggregating the individual clinical pathways to compute a cohort clinical pathway DAG, wherein the cohort clinical pathway DAG includes a plurality of nodes comprising the knowledge-driven nodes, the data-driven nodes, and links connecting the plurality of nodes, each link denoting an automatically discovered sequence between two respective nodes;

instructions for receiving, via the GUI, a manual collapse instruction, and, in response to receiving the manual collapse instruction, aggregating the data-driven event type nodes of the cohort clinical pathway DAG to create a collapsed cohort clinical pathway DAG depicting direct flows between the knowledge-driven nodes; and

instructions for presenting the collapsed cohort clinical pathway DAG within the GUI.