Determining correlation between medical symptoms and environmental factors

Patent metadata

Field	Value
Publication number	US-10698982-B2
Application number	US-201615052195-A
Country	US
Kind code	B2
Filing date	Feb 24, 2016
Priority date	Feb 24, 2016
Publication date	Jun 30, 2020
Grant date	Jun 30, 2020

Abstract

A method, a processing device, and a computer program product are provided. Unstructured text may be analyzed to identify medical condition information of multiple occurrences of a medical condition for at least one subject. Times and geographic locations corresponding to the multiple occurrences of the medical condition may be obtained. Environmental information that corresponds to the times and the geographic locations of the multiple medical condition occurrences, may be retrieved. Correlations between the medical condition information and the retrieved environmental information for the at least one subject may be determined. Environmental factors affecting the medical condition, based on the determined correlations, are identified.

First claim

We claim as our invention:

1. A computer-implemented method for identifying factors affecting a medical condition, the method comprising:
analyzing unstructured text to identify medical condition information of a plurality of occurrences of the medical condition for at least one subject, the analyzing further comprising:
creating, by a first stage of an annotator executing on a computing device, annotations covering sections of the unstructured text, and
performing by subsequent stages of the annotator:
reading annotations created by earlier stages of the annotator, and
performing at least one from a group of adding to and modifying the annotations created by the earlier stages of the annotator, wherein
the annotator includes one or more predefined dictionaries and rules that identify medical symptoms, degree of the medical symptoms, dates and a location;

obtaining times and geographic locations corresponding to the plurality of occurrences of the medical condition, the times being obtained from one of a group of the contents of the unstructured text and a date and a time of an unstructured text entry of the unstructured text, and the geographic locations being obtained from one of a group of the contents of the unstructured text, a geographic location of a provider of the unstructured text supplied by a global positioning system device, a network location, and a corresponding user profile;
retrieving environmental information corresponding to the times and the geographic locations of the occurrences of the medical condition;
determining, via machine learning executing on the computing device, correlations between the medical condition information and the retrieved environmental information for the at least one subject;
identifying environmental factors affecting the medical condition based on the determined correlations; and
when the identified environmental factors that are correlated with aggravating symptoms of the medical condition are discovered, performing, by the computing device, at least one action including at least one from a group of alerting the at least one subject suffering from the medical condition to take medication and contacting a respective physician of the at least one subject.

2. The computer-implemented method of claim 1 , wherein the unstructured text is from at least one of a group of social media, email, and medical documents.

3. The computer-implemented method of claim 1 , wherein the environmental information includes weather information.

4. The computer-implemented method of claim 3 , wherein the identifying environmental factors comprises:
identifying weather conditions affecting the medical condition based on the determined correlations.

5. The computer-implemented method of claim 1 , wherein the determining correlations comprises:
determining correlations between medical condition information and retrieved environmental information aggregated for a plurality of subjects.

6. A computer program product comprising:
one or more computer readable storage media collectively having computer readable program code embodied therewith for execution on a processing system, the computer readable program code being configured to be executed by the processing system to:
analyze unstructured text to identify medical condition information of a plurality of occurrences of a medical condition for at least one subject, the analyzing further comprising:
create, by a first stage of an annotator executing on a computing device, annotations covering sections of the unstructured text, and
perform by subsequent stages of the annotator:
reading annotations created by earlier stages of the annotator, and
performing at least one from a group of adding to and modifying the annotations created by the earlier stages of the annotator, wherein
the annotator includes one or more predefined dictionaries and rules that identify medical symptoms, degree of the medical symptoms, dates and a location;

obtain times and geographic locations corresponding to the plurality of occurrences of the medical condition, the times being obtained from one of a group of the contents of the unstructured text and a date and a time of an unstructured text entry of the unstructured text, and the geographic locations being obtained from one of a group of the contents of the unstructured text, a geographic location of a provider of the unstructured text supplied by a global positioning system device, a network location, and a corresponding user profile;
retrieve environmental information corresponding to the times and the geographic locations of the occurrences of the medical condition;
determine, via machine learning executing on the processing system, correlations between the medical condition information and the retrieved environmental information for the at least one subject;
identify environmental factors affecting the medical condition based on the determined correlations; and
when the identified environmental factors that are correlated with aggravating symptoms of the medical condition are discovered, performing, by the computing device, at least one action including at least one from a group of alerting the at least one subject suffering from the medical condition to take medication and contacting a respective physician of the at least one subject.

7. The computer program product of claim 6 , wherein the unstructured text is from at least one of a group of social media, email, and medical documents.

8. The computer program product of claim 6 , wherein the environmental information includes weather information.

9. The computer program product of claim 8 , wherein the identify environmental factors comprises:
identify weather conditions affecting the medical condition based on the determined correlations.

10. A processing device comprising:
at least one processor;
a memory; and
a communication bus connecting the at least one processor with the memory, wherein the memory has stored therein instructions, which when executed by the at least one processor cause the processing device to perform a method comprising:
analyzing unstructured text to identify medical condition information of a plurality of occurrences of a medical condition for at least one subject, the analyzing further comprising:
create, by a first stage of an annotator, annotations covering sections of the unstructured text, and
perform by subsequent stages of the annotator:
reading annotations created by earlier stages of the annotator, and
performing at least one from a group of adding to and modifying the annotations created by the earlier stages of the annotator, wherein
the annotator includes one or more predefined dictionaries and rules that identify medical symptoms, degree of the medical symptoms, dates and a location;

obtaining times and geographic locations corresponding to the plurality of occurrences of the medical condition, the times being obtained from one of a group of the contents of the unstructured text and a date and a time of an unstructured text entry of the unstructured text, and the geographic locations being obtained from one of a group of the contents of the unstructured text, a geographic location of a provider of the unstructured text supplied by a global positioning system device, a network location, and a corresponding user profile;
retrieving environmental information corresponding to the times and the geographic locations of the occurrences of the medical condition;
determining, via machine learning, correlations between the medical condition information and the retrieved environmental information for the at least one subject; and
identifying environmental factors affecting the medical condi

Assignees

• IBM

Inventors

• Fink Patrick W
• Mcneil Kristin E
• Parker Philip E
• Werts David B

Classifications

• G06F16/35 Primary

  Clustering; Classification · CPC title

• G16H50/80

  for detecting, monitoring or modelling epidemics or pandemics, e.g. flu · CPC title

• G16H40/67

  for remote operation · CPC title

• G06F19/3418 Primary

  Physics · mapped topic

• G06F19/324

  Physics · mapped topic