Dynamic creation and manipulation of data visualizations

1. A method comprising:

receiving a plurality of therapies, wherein each respective therapy of the plurality of therapies is associated with a respective plurality of guidelines indicating guidance regarding whether the respective therapy is appropriate based on patient attributes;

generating a guideline tree based on the plurality of therapies, the guideline tree comprising:

a set of leaf nodes, wherein each respective leaf node represents a respective therapy,

a set of internal nodes, wherein the set of internal nodes represent attributes that inform treatment decisions, based on the respective pluralities of guidelines, and

a set of edges, wherein each respective edge represents a respective guideline;

generating a visual depiction of the guideline tree, wherein the visual depiction depicts the set of leaf nodes, the set of internal nodes, and the set of edges arranged as a tree;

receiving a first plurality of attributes associated with a first patient;

generating a first modified visual depiction of the guideline tree based on the first plurality of attributes, wherein the first modified visual depiction visually emphasizes a subset of the guideline tree;

outputting the first modified visual depiction via a graphical user interface (GUI);

receiving a second plurality of attributes associated with the first patient;

dynamically generating a second modified visual depiction of the guideline tree based on the second plurality of attributes, wherein the second modified visual depiction emphasizes a first path from a root node to a leaf node associated with a first suggested therapy based on the first plurality of attributes, and a second path from the root node to a leaf node associated with a second suggested therapy based on the second plurality of attributes;

generating a dynamic visualization of the guideline tree, wherein the dynamic visualization includes an animation comprising a morph between the first modified visual depiction and the second modified visual depiction;

generating a knowledge graph based on a plurality of Relative Efficacy Structures (RESs), wherein:

each RES is generated from a respective comparison statement between at least two identified therapies with respect to at least one outcome,

the respective comparison statement is identified by using one or more natural language processing (NLP) techniques in one or more natural language texts from a corpus of medical literature, and

the knowledge graph comprises a plurality of nodes and a plurality of edges, wherein:

each node represents a therapy associated with one or more of the RESs, and

each edge connects at least two of the nodes and represents a relative efficacy between the at least two therapies represented by the at least two nodes based on the one or more RESs;

determining an optimal therapy identified in the knowledge graph for the first patient is different from the first suggested therapy, wherein the optimal therapy corresponds to a first node of the plurality of nodes in the knowledge graph, the first suggested therapy corresponds to a second node of the plurality of nodes in the knowledge graph, and an edge of the plurality of edges in the knowledge graph indicates that the first node has a higher efficacy than the second node with respect to at least one outcome of at least one of the one or more RESs;

updating the dynamic visualization by generating a new modified visual depiction of the guideline tree to emphasize a leaf node corresponding to the optimal therapy; and

outputting the dynamic visualization via a graphical user interface (GUI).

2. The method of claim 1 , the method further comprising:

receiving a third plurality of attributes associated with a second patient; and

generating a third modified visual depiction of the guideline tree based on the third plurality of attributes, wherein the third modified visual depiction emphasizes a first path from a root node to a leaf node associated with a first suggested therapy for the first patient, and a second path from the root node to a leaf node associated with a second suggested therapy for the first patient, and a third path from the root node to a leaf node associated with a third suggested therapy for the second patient.

3. The method of claim 1 , the method further comprising:

analyzing the knowledge graph to identify the optimal therapy, of the plurality of therapies, for the first patient.

4. The method of claim 1 , the method further comprising:

identifying a first path in the guideline tree beginning at a root node and ending at the leaf node corresponding to the first suggested therapy;

identifying a second path in the guideline tree beginning at the root node and ending at the leaf node corresponding to the optimal therapy;

determining a node in the guideline tree where the first path and the second path diverge; and

providing an indication of the determined node.

5. The method of claim 1 , the method further comprising:

identifying a cohort of similar patients, based on the first plurality of attributes associated with the first patient;

analyzing one or more electronic medical records associated with at least one similar patient in the cohort of similar patients to determine a preferred therapy; and

upon determining that the preferred therapy is different from the first suggested therapy, generating a third modified visual depiction of the guideline tree to emphasize a leaf node corresponding to the preferred therapy.

6. The method of claim 1 , the method further comprising determining a set of edges in the guideline tree based on the first plurality of attributes, wherein the first modified visual depiction of the guideline tree further emphasizes the set of edges, and wherein the set of edges corresponds to a path in the guideline tree beginning at a root node and ending at the leaf node corresponding to the first suggested therapy.

7. A computer-readable storage medium having computer-readable program code embodied therewith, the computer-readable program code executable by one or more computer processors to perform an operation comprising:

receiving a plurality of therapies, wherein each respective therapy of the plurality of therapies is associated with a respective plurality of guidelines indicating guidance regarding whether the respective therapy is appropriate based on patient attributes;

generating a guideline tree based on the plurality of therapies, the guideline tree comprising:

a set of leaf nodes, wherein each respective leaf node represents a respective therapy,

a set of internal nodes, wherein the set of internal nodes represent attributes that inform treatment decisions, based on the respective pluralities of guidelines, and

a set of edges, wherein each respective edge represents a respective guideline;

generating a visual depiction of the guideline tree, wherein the visual depiction depicts the set of leaf nodes, the set of internal nodes, and the set of edges arranged as a tree;

receiving a first plurality of attributes associated with a first patient;

generating a first modified visual depiction of the guideline tree based on the first plurality of attributes, wherein the first modified visual depiction visually emphasizes a subset of the guideline tree;

outputting the first modified visual depiction via a graphical user interface (GUI);

receiving a second plurality of attributes associated with the first patient;

dynamically generating a second modified visual depiction of the guideline tree based on the second plurality of attributes, wherein the second modified visual depiction emphasizes a first path from a root node to a leaf node associated with a first suggested therapy based on the first plurality of attributes, and a second path from the root node to a leaf node associated with a second suggested therapy based on the second plurality of attributes;

generating a dynamic visualization of the guideline tree, wherein the dynamic visualization includes an animation comprising a morph between the first modified visual depiction and the second modified visual depiction;

generating a knowledge graph based on a plurality of Relative Efficacy Structures (RESs), wherein:

each RES is generated from a respective comparison statement between at least two identified therapies with respect to at least one outcome,

the respective comparison statement is identified by using one or more natural language processing (NLP) techniques in one or more natural language texts from a corpus of medical literature, and

the knowledge graph comprises a plurality of nodes and a plurality of edges, wherein:

each node represents a therapy associated with one or more of the RESs, and

each edge connects at least two of the nodes and represents a relative efficacy between the at least two therapies represented by the at least two nodes based on the one or more RESs;

determining an optimal therapy identified in the knowledge graph for the first patient is different from the first suggested therapy, wherein the optimal therapy corresponds to a first node of the plurality of nodes in the knowledge graph, the first suggested therapy corresponds to a second node of the plurality of nodes in the knowledge graph, and an edge of the plurality of edges in the knowledge graph indicates that the first node has a higher efficacy than the second node with respect to at least one outcome of at least one of the one or more RESs;

updating the dynamic visualization by generating a new modified visual depiction of the guideline tree to emphasize a leaf node corresponding to the optimal therapy; and

outputting the dynamic visualization via a graphical user interface (GUI).

8. The computer-readable storage medium of claim 7 , the operation further comprising:

identifying a first path in the guideline tree beginning at a root node and ending at the leaf node corresponding to the first suggested therapy;

identifying a second path in the guideline tree beginning at the root node and ending at the leaf node corresponding to the optimal therapy;

determining a node in the guideline tree where the first path and the second path diverge; and

providing an indication of the determined node.

9. The computer-readable storage medium of claim 7 , the operation further comprising:

identifying a cohort of similar patients, based on the first plurality of attributes associated with the first patient;

analyzing one or more electronic medical records associated with at least one similar patient in the cohort of similar patients to determine a preferred therapy; and

upon determining that the preferred therapy is different from the first suggested therapy, generating a third modified visual depiction of the guideline tree to emphasize a leaf node corresponding to the preferred therapy.

10. The computer-readable storage medium of claim 7 , the operation further comprising determining a set of edges in the guideline tree based on the first plurality of attributes, wherein the first modified visual depiction of the guideline tree further emphasizes the set of edges, and wherein the set of edges corresponds to a path in the guideline tree beginning at a root node and ending at the leaf node corresponding to the first suggested therapy.

11. The computer-readable storage medium of claim 7 , the operation further comprising:

analyzing the knowledge graph to identify the optimal therapy, of the plurality of therapies, for the first patient.

12. A system comprising:

one or more computer processors; and

a memory containing a program which when executed by the one or more computer processors performs an operation, the operation comprising:

receiving a plurality of therapies, wherein each respective therapy of the plurality of therapies is associated with a respective plurality of guidelines indicating guidance regarding whether the respective therapy is appropriate based on patient attributes;

generating a guideline tree based on the plurality of therapies, the guideline tree comprising:

a set of leaf nodes, wherein each respective leaf node represents a respective therapy,

a set of internal nodes, wherein the set of internal nodes represent attributes that inform treatment decisions, based on the respective pluralities of guidelines, and

a set of edges, wherein each respective edge represents a respective guideline;

generating a visual depiction of the guideline tree, wherein the visual depiction depicts the set of leaf nodes, the set of internal nodes, and the set of edges arranged as a tree;

receiving a first plurality of attributes associated with a first patient;

generating a first modified visual depiction of the guideline tree based on the first plurality of attributes, wherein the first modified visual depiction visually emphasizes a subset of the guideline tree;

outputting the first modified visual depiction via a graphical user interface (GUI);

receiving a second plurality of attributes associated with the first patient;

dynamically generating a second modified visual depiction of the guideline tree based on the second plurality of attributes, wherein the second modified visual depiction emphasizes a first path from a root node to a leaf node associated with a first suggested therapy based on the first plurality of attributes, and a second path from the root node to a leaf node associated with a second suggested therapy based on the second plurality of attributes;

generating a dynamic visualization of the guideline tree, wherein the dynamic visualization includes an animation comprising a morph between the first modified visual depiction and the second modified visual depiction;

generating a knowledge graph based on a plurality of Relative Efficacy Structures (RESs), wherein:

each RES is generated from a respective comparison statement between at least two identified therapies with respect to at least one outcome,

the respective comparison statement is identified by using one or more natural language processing (NLP) techniques in one or more natural language texts from a corpus of medical literature, and

the knowledge graph comprises a plurality of nodes and a plurality of edges, wherein:

each node represents a therapy associated with one or more of the RESs, and

each edge connects at least two of the nodes and represents a relative efficacy between the at least two therapies represented by the at least two nodes based on the one or more RESs;

determining an optimal therapy identified in the knowledge graph for the first patient is different from the first suggested therapy, wherein the optimal therapy corresponds to a first node of the plurality of nodes in the knowledge graph, the first suggested therapy corresponds to a second node of the plurality of nodes in the knowledge graph, and an edge of the plurality of edges in the knowledge graph indicates that the first node has a higher efficacy than the second node with respect to at least one outcome of at least one of the one or more RESs;

updating the dynamic visualization by generating a new modified visual depiction of the guideline tree to emphasize a leaf node corresponding to the optimal therapy; and

outputting the dynamic visualization via a graphical user interface (GUI).

13. The system of claim 12 , the operation further comprising:

analyzing the knowledge graph to identify the optimal therapy, of the plurality of therapies, for the first patient.

14. The system of claim 12 , the operation further comprising:

identifying a first path in the guideline tree beginning at a root node and ending at the leaf node corresponding to the first suggested therapy;

identifying a second path in the guideline tree beginning at the root node and ending at the leaf node corresponding to the optimal therapy;

determining a node in the guideline tree where the first path and the second path diverge; and

providing an indication of the determined node.

15. The system of claim 12 , the operation further comprising:

identifying a cohort of similar patients, based on the first plurality of attributes associated with the first patient;

analyzing one or more electronic medical records associated with at least one similar patient in the cohort of similar patients to determine a preferred therapy; and

upon determining that the preferred therapy is different from the suggested therapy, generating a third modified visual depiction of the guideline tree to emphasize a leaf node corresponding to the preferred therapy.

16. The system of claim 12 , the operation further comprising determining a set of edges in the guideline tree based on the first plurality of attributes, wherein the first modified visual depiction of the guideline tree further emphasizes the set of edges, and wherein the set of edges corresponds to a path in the guideline tree beginning at a root node and ending at the leaf node corresponding to the first suggested therapy.