Automated prediction and answering of medical professional questions directed to patient based on EMR

1. A method, in a data processing system comprising a processor and a memory, the memory comprising instructions that are executed by the processor to specifically configure the processor to implement a question prediction and answering engine for predicting questions a medical professional is attempting to answer, the method comprising:

training, by a personalized interaction learning engine, a machine learning model, comprising:

monitoring interaction of a user with an electronic medical record (EMR) to identify one or more natural language questions entered by the user into an input device and submitted to a cognitive question answering system;

identifying a first set of questions within a predetermined set of questions that match the one or more natural language questions entered by the user;

correlating the one or more natural language questions entered by the user to the first set of questions; and

training the machine learning model based on the correlation between the one or more natural language questions entered by the user and the first set of questions;

monitoring, by an interaction monitoring component executing within the question prediction and answering engine, interaction of a medical professional with a patient electronic medical record (EMR) to identify a medical professional interaction pattern;

applying, by a question selection component executing within the question prediction and answering engine, the machine learning model to the medical professional interaction pattern to select a second set of questions, from the predetermined set of questions, the medical professional is attempting to obtain an answer to from the patient EMR, wherein the machine learning model is configured to receive the medical professional interaction pattern and determine the second set of questions that correlate to the medical professional interaction pattern;

submitting, by a question submission component executing within the question prediction and answering engine, the second set of questions to the cognitive question answering system;

receiving, by an answer processing component executing within the question prediction and answering engine, answers to the second set of questions from the cognitive question answering system based on the patient EMR;

processing, by the answer processing component, the answers to the second set of questions to generate a set of answers to the second set of questions from at least a portion of the patient EMR; and

outputting, by the question prediction and answering engine, a report correlating the second set of questions and the set of answers to the medical professional at least by generating a graphical user interface (GUI) to present the answers to the second set of questions to the medical professional as a tailored, cognitive information retrieval, wherein the GUI presents areas needing additional data retrieval based on questions in the second set of questions for which an answer could not be found in the patient EMR.

2. The method of claim 1 , wherein monitoring the interaction of the medical professional with the EMR comprises detecting an interaction from the set consisting of portions of the EMR the medical professional views, an order in which the medical professional views EMR portions, the medical professional hovering a cursor over a particular location in the EMR, or the medical professional zooming on the at least a portion of the patient EMR.

3. The method of claim 1 , further comprising:

determining a context of the interaction of the medical professional with the patient EMR; and

prioritizing the second set of questions based on the determined context.

4. The method of claim 3 , wherein determining the context of the interaction comprises detecting an interaction from the set consisting of portions of the EMR the medical professional views, an order in which the medical professional views EMR portions, the medical professional hovering a cursor over a particular location in the EMR, or the medical professional zooming on the at least a portion of the patient EMR.

5. The method of claim 3 , wherein prioritizing the second set of questions comprises prioritizing the second set of questions based on what medications the patient is taking, medication side effects, patient vital signs, or blood test results.

6. The method of claim 1 , wherein processing the answers to the second set of questions comprises modifying a form of the answers to be included in a graphical user interface (GUI).

7. The method of claim 1 , wherein the GUI presents the answers to the second set of questions in association with the second set of questions.

8. The method of claim 1 , wherein the GUI provides links to portions of the patient EMR in association with the answers to the second set of questions.

9. A computer program product comprising a computer readable storage medium having a computer readable program stored therein, wherein the computer readable program, when executed on at least one processor of a data processing system, causes the data processing system to implement a question prediction and answering engine for predicting questions a medical professional is attempting to answer, wherein the computer readable program causes the data processing system to:

train, by a personalized interaction learning engine, a machine learning model, comprising:

monitoring interaction of a user with an electronic medical record (EMR) to identify one or more natural language questions entered by the user into an input device and submitted to a cognitive question answering system;

identifying a first set of questions within a predetermined set of questions that match the one or more natural language questions entered by the user;

correlating the one or more natural language questions entered by the user to the first set of questions; and

training the machine learning model based on the correlation between the one or more natural language questions entered by the user and the first set of questions;

monitor, by an interaction monitoring component executing within the question prediction and answering engine, interaction of a medical professional with a patient electronic medical record (EMR) to identify a medical professional interaction pattern;

apply, by a question selection component executing within the question prediction and answering engine, the machine learning model to the medical professional interaction pattern to select a second set of questions, from the predetermined set of questions, the medical professional is attempting to obtain an answer to from the patient EMR, wherein the machine learning model is configured to receive the medical professional interaction pattern and determine the set of questions that correlate to the medical professional interaction pattern;

submit, by a question submission component executing within the question prediction and answering engine, the second set of questions to the cognitive question answering system;

receive, by an answer processing component executing within the question prediction and answering engine, answers to the second set of questions from the cognitive question answering system based on the patient EMR;

process, by the answer processing component, the answers to the second set of questions to generate a set of answers to the second set of questions from at least a portion of the patient EMR; and

output, by the question prediction and answering engine, a report correlating the second set of questions and the set of answers to the medical professional at least by generating a graphical user interface (GUI) to present the answers to the second set of questions to the medical professional as a tailored, cognitive information retrieval, wherein the GUI presents areas needing additional data retrieval based on questions in the second set of questions for which an answer could not be found in the patient EMR.

10. The computer program product of claim 9 , wherein the computer readable program further causes the data processing system to:

determine a context of the interaction of the medical professional with the patient EMR; and

prioritize the second set of questions based on the determined context.

11. The computer program product of claim 9 , wherein processing the answers to the second set of questions comprises modifying a form of the answers to be included in a graphical user interface (GUI).

12. An apparatus comprising:

a processor; and

a memory coupled to the processor, wherein the memory comprises instructions which, when executed by the processor, cause the processor to implement a question prediction and answering engine for predicting questions a medical professional is attempting to answer, wherein the instructions cause the processor to:

train, by a personalized interaction learning engine, a machine learning model, comprising:

monitoring interaction of a user with an electronic medical record (EMR) to identify one or more natural language questions entered by the user into an input device and submitted to a cognitive question answering system;

identifying a first set of questions within a predetermined set of questions that match the one or more natural language questions entered by the user;

correlating the one or more natural language questions entered by the user to the first set of questions; and

training the machine learning model based on the correlation between the one or more natural language questions entered by the user and the first set of questions;

monitor, by an interaction monitoring component executing within the question prediction and answering engine, interaction of a medical professional with a patient electronic medical record (EMR) to identify a medical professional interaction pattern;

apply, by a question selection component executing within the question prediction and answering engine, the machine learning model to the medical professional interaction pattern to select a second set of questions, from the predetermined set of questions, the medical professional is attempting to obtain an answer to from the patient EMR, wherein the machine learning model is configured to receive the medical professional interaction pattern and determine the set of questions that correlate to the medical professional interaction pattern;

submit, by a question submission component executing within the question prediction and answering engine, the second set of questions to the cognitive question answering system;

receive, by an answer processing component executing within the question prediction and answering engine, answers to the second set of questions from the cognitive question answering system based on the patient EMR;

process, by the answer processing component, the answers to the second set of questions to generate a set of answers to the second set of questions from at least a portion of the patient EMR; and

output, by the question prediction and answering engine, a report correlating the second set of questions and the set of answers to the medical professional at least by generating a graphical user interface (GUI) to present the answers to the second set of questions to the medical professional as a tailored, cognitive information retrieval, wherein the GUI presents areas needing additional data retrieval based on questions in the second set of questions for which an answer could not be found in the patient EMR.

13. The method of claim 1 , wherein identifying the first set of questions comprises:

examining features from the user interaction pattern;

treating the features as if they are evidence passages or portions for candidate answers to a question; and

determining which questions from a predetermined set of questions the evidence passages would answer.

14. The method of claim 1 , wherein identifying the first set of questions comprises:

submitting each question from a predetermined set of questions to the cognitive question answering system to determine a set of evidence passages or portions that support an answer to each question; and

determining which sets of evidence passages or portions match the user interaction pattern.

15. The computer program product of claim 9 , wherein identifying the first set of questions comprises:

examining features from the user interaction pattern;

treating the features as if they are evidence passages or portions for candidate answers to a question; and

determining which questions from a predetermined set of questions the evidence passages would answer.

16. The computer program product of claim 9 , wherein identifying the first set of questions comprises:

submitting each question from a predetermined set of questions to the cognitive question answering system to determine a set of evidence passages or portions that support an answer to each question; and

determining which sets of evidence passages or portions match the user interaction pattern.

17. The apparatus of claim 12 , wherein identifying the first set of questions comprises:

examining features from the user interaction pattern;

treating the features as if they are evidence passages or portions for candidate answers to a question; and

determining which questions from a predetermined set of questions the evidence passages would answer.

18. The apparatus of claim 12 , wherein identifying the first set of questions comprises:

submitting each question from a predetermined set of questions to the cognitive question answering system to determine a set of evidence passages or portions that support an answer to each question; and

determining which sets of evidence passages or portions match the user interaction pattern.