Cognitive stroke detection and notification

Patent metadata

Field	Value
Publication number	US-11139079-B2
Application number	US-201715450603-A
Country	US
Kind code	B2
Filing date	Mar 6, 2017
Priority date	Mar 6, 2017
Publication date	Oct 5, 2021
Grant date	Oct 5, 2021

Abstract

Embodiments of the invention include methods, systems, and computer program products for determining stroke onset. Aspects of the invention include determining a baseline behavioral model for a user and receiving real-time user data from a personal portable device. Aspects of the invention also include analyzing the real-time user data to determine the presence of an abnormal event. Aspects of the invention also include, based at least in part on a determination that the abnormal event is present, conducting a plurality of stroke analyses to generate a plurality of impairment characteristics. Aspects of the invention also include integrating the plurality of impairment characteristics, comparing the integrated plurality of impairment characteristics to the baseline behavioral model and outputting a stroke onset determination.

First claim

What is claimed is:

1. A computer program product for determining stroke onset, the computer program product comprising:
a computer readable storage medium readable by a processing circuit and storing program instructions for execution by the processing circuit for performing a method comprising:
receiving training data from a user, the training data comprising text entry data;
training a baseline behavioral model on the training data, wherein the baseline behavior model is further trained against false positive states using additional data corresponding to one or more states that mimic one or more symptoms of stroke;
determining a baseline text entry threshold based on the baseline behavioral model;
receiving real-time user data from a personal portable device, the real-time user data comprising real-time text entry data, the real-time text entry data comprising a text correction rate;
analyzing the real-time user data and the baseline text entry threshold to determine the presence of an abnormal event;
conducting a plurality of stroke analyses to generate a plurality of impairment characteristics based at least in part on a determination that the abnormal event is present;
integrating the plurality of impairment characteristics;
comparing the integrated plurality of impairment characteristics to the baseline behavioral model; and
outputting a stroke onset determination.

2. The computer program product according to claim 1 , wherein the baseline behavioral model is personalized to the user.

3. The computer program product according to claim 1 , wherein the real-time user data comprises real-time speech data.

4. The computer program product according to claim 3 , wherein analyzing the real-time user data to determine the presence of the abnormal event comprises:
analyzing the real-time speech data for comprehensibility, and
comparing the comprehensibility to a baseline comprehensibility.

5. The computer program product according to claim 1 , wherein analyzing the real-time user data to determine the presence of the abnormal event comprises:
calculating a personal text input profile for the user;
comparing the real-time text entry data to the personal text input profile for the user; and
determining the presence of a significant text entry deviation based upon the comparison.

6. The computer program product according to claim 1 , wherein the real-time data comprises real-time positional data.

7. The computer program product according to claim 6 , wherein analyzing the real-time user data to determine the presence of the abnormal event comprises:
calculating a personalized baseline movement model for the user;
determining a plurality of real-time walking characteristics of the user;
comparing the plurality of real-time walking characteristics of the user to the personalized baseline movement model for the user; and
determining the presence of the abnormal event based upon the comparison.

8. The computer program product according to claim 1 , wherein the method further comprises conducting a facial analysis.

9. The computer program product according to claim 8 , wherein the facial analysis comprises:
receiving baseline facial data from the personal portable device, wherein the personal portable device comprises a camera, and wherein the baseline facial data comprises an image of a face the user performing an action;
calculating a baseline facial symmetry and a symmetry threshold based upon the baseline facial data;
outputting a signal directing the user to perform the action;
receiving real-time facial data from the camera, wherein the real-time facial data comprises an updated user facial image;
calculate a real-time facial symmetry based upon the real-time facial data; and
compare the real-time facial symmetry to the baseline facial symmetry threshold.

10. The computer program product according to claim 1 , wherein the method further comprises conducting a real-time textual analysis.

11. The computer program product according to claim 1 , wherein the method further comprises conducting a real-time verbal analysis.

12. The computer program product according to claim 1 , wherein the method further comprises conducting a real-time motion analysis.

13. The computer program product according to claim 12 , wherein the method further comprises conducting a facial analysis and a real-time verbal analysis.

14. The computer program product according to claim 1 , wherein software is provided as a service in a cloud environment.

15. The computer program product according to claim 1 , wherein the method further comprises sending a notification to an emergency responder.

16. A computer-implemented method for determining stroke onset, the method comprising:
receiving training data from a user, the training data comprising text entry data;
training, by a processor, a baseline behavioral model on the training data, wherein the baseline behavior model is further trained against false positive states using additional data corresponding to one or more states that mimic one or more symptoms of stroke;
determining a baseline text entry threshold based on the baseline behavioral model;
receiving, to the processor, real-time user data from a personal portable device, the real-time user data comprising real-time text entry data, the real-time text entry data comprising a text correction rate;
analyzing, by the processor, the real-time user data and the baseline text entry threshold to determine the presence of an abnormal event;
conducting, by the processor, a plurality of stroke analyses to generate a plurality of impairment characteristics based at least in part on a determination that the abnormal event is present;
integrating, by the processor, the plurality of impairment characteristics;
comparing, by the processor, the integrated plurality of impairment characteristics to the baseline behavioral model; and
outputting a stroke onset determination.

17. The computer-implemented method of claim 16 , wherein the baseline behavioral model is personalized to the user.

18. A processing system for determining stroke onset, comprising:
a processor in communication with one or more types of memory, the processor configured to:
receive training data from a user, the training data comprising text entry data;
train a baseline behavioral model on the training data, wherein the baseline behavior model is further trained against false positive states using additional data corresponding to one or more states that mimic one or more symptoms of stroke;
determine a baseline text entry threshold based on the baseline behavioral model;
receive real-time user data from a personal portable device, the real-time user data comprising real-time text entry data, the real-time text entry data comprising a text correction rate;
analyze the real-time user data and the baseline text entry threshold to determine the presence of an abnormal event;
conduct a plurality of stroke analyses to generate a plurality of impairment characteristics based at least in part on a determination that the abnormal event is present;
integrate the plurality of impairment characteristics;
compare the integrated plurality of impairment characteristics to the baseline behavioral model; and
output a stroke onset determination.

19. The processing system of claim 18 , wherein the baseline behavioral model is personalized to the user.

Assignees

• IBM

Inventors

• Goldberg Itzhack
• Norel Raquel
• Rhrissorrakrai Kahn

Classifications

• G16H40/67

  for remote operation · CPC title

• G16H50/20 Primary

  for computer-aided diagnosis, e.g. based on medical expert systems · CPC title

• G16H50/50

  for simulation or modelling of medical disorders · CPC title

• A61B5/0077

  Devices for viewing the surface of the body, e.g. camera, magnifying lens · CPC title

• A61B5/4803

  Speech analysis specially adapted for diagnostic purposes · CPC title