Identification of emotional states using physiological responses

• US 9,833,184 B2
• Filed: 10/26/2007
• Issued: 12/05/2017
• Est. Priority Date: 10/27/2006
• Status: Active Grant

• Current Assignee: Adidas AG
• Original Assignee: Adidas AG
• Inventors: Derchak, P. Alexander, Myers, Lance Jonathan
• Primary Examiner(s): JIAN, SHIRLEY XUEYING
• Application Number:
  US11/925,109
• Publication Number:
  US 20080221401A1
• Time in Patent Office: 3,693 Days
• Field of Search: 600 31, 600544, 600485, 600483, 600300-301
• CPC Class Codes: A61B 5/16 Devices for psychotechnics using teaching or educational appliances G09B1/00 - G09B7/00; Testing reaction times ; Devices for evaluating the psychological state
• CPC Class Codes: A61B 5/165 Evaluating the state of mind, e.g. depression, anxiety
• CPC Class Codes: A61B 5/6805 Vests
• CPC Class Codes: A61B 5/7267 involving training the classification device
• CPC Class Codes: G06F 2203/011 Emotion or mood input determined on the basis of sensed human body parameters such as pulse, heart rate or beat, temperature of skin, facial expressions, iris, voice pitch, brain activity patterns
• CPC Class Codes: G06F 2203/014 Force feedback applied to GUI
• CPC Class Codes: G16H 40/63 for local operation
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

The invention provides a method for automatically identifying emotional states of a person in real-time based on physiological responses by exposing a person to a stimulus, measuring the person'"'"'s physiological response to the stimulus, and comparing the measured physiological response to a known baseline physiological. The deviation of the measured physiological response from the baseline physiological response is determinative of the emotional state.

Claims

1. A method for automatically identifying a likely emotional state in real-time comprising:

exposing a person to a stimulus;

measuring the person'"'"'s physiological responses to the stimulus using a physiological monitoring device;

comparing the measured physiological responses to baseline physiological responses using a processor, wherein;

the baseline physiological responses of the person are measured when exposed to a stimulus known as likely to elicit a selected emotional state, and

the stimulus used to elicit the person'"'"'s baseline physiological responses and the stimulus used to elicit the person'"'"'s measured physiological responses are the same type;

recording and storing the baseline physiological responses as an acceptable range in an electronic database, wherein the baseline physiological responses are in association with particular emotional states;

the processor determining the degree to which the person is engaged with, concentrated on, or focused on the stimulus; and

the processor identifying a likely emotional state in dependence on the pattern of deviations of the measured physiological responses from the acceptable range of the baseline physiological responses and in dependence on the degree to which the person is engaged with, concentrated on, or focused on the stimulus,

wherein the measured physiological responses and baseline physiological responses include both standard deviation and instability values, and

wherein the measured physiological responses and baseline physiological responses comprise respiratory responses.

2. The method of claim 1, wherein the measured respiratory responses comprise a measure of breath to breath variability.

3. The method of claim 2, wherein the measure of breath to breath variability comprises a sample entropy.

4. The method of claim 1, wherein the measured respiratory responses comprise a respiratory rate, a minute ventilation rate, a tidal volume, a tidal volume instability, an inspiratory flow rate, an expiratory flow rate, an inspiratory pause, an expiratory pause, a presence of cough, or presence of apnea or hypopnea, or a combination thereof.

5. The method of claim 1, wherein the measured physiological responses further comprise a cardiac response, a posture/activity response, a temperature response, a galvanic skin response, an EEG response, an EOG response, an EMG response, or a vocal response, or a combination thereof.

6. The method of claim 5, wherein the cardiac responses comprise an ECG response, an RSA response, a heart rate response, or a heart rate variability response, or a combination thereof.

7. The method of claim 1, wherein the emotional state comprises neutrality, indifference, fear, anxiety, excitement, happiness, sadness, attentiveness, anticipation, boredom, anger, pleasure, or calmness, or a combination thereof.

8. The method of claim 1, wherein the emotional state comprises a measure of the degree to which the person is engaged with, concentrated on, or focused on the stimulus.

9. The method of claim 1, wherein identifying the likely emotional state further depends on the measured physiological responses being significantly similar to the baseline physiological responses, the baseline physiological response and the measured physiological response being defined as significantly similar when the likelihood that the similarity is due to chance is less than a selected threshold as defined through a measure of statistical significance.

10. The method of claim 1, wherein identifying a likely emotional state occurs based on conscious and subconscious responses of the person.

11. The method of claim 1, wherein the plurality of emotional states comprises fear, no fear, and excitement.

12. A monitoring system for automatically identifying an emotional state of a person in real-time, comprising:

a processor;

a physiological monitoring device operatively coupled to the processor for measuring the person'"'"'s physiological response, the device being configured for monitoring an ambulatory subject; and

a computer readable memory operatively coupled to the processor and configured with instructions for causing the processor to compare measured physiological responses obtained using the physiological monitoring device to baseline physiological responses, wherein;

the baseline physiological responses of the person are measured when exposed to a stimulus known as likely to elicit a selected emotional state, and

the stimulus used to elicit the person'"'"'s baseline physiological responses and the stimulus used to elicit the person'"'"'s measured physiological responses are the same type;

the processor recording and storing the baseline physiological responses as an acceptable range in an electronic database in the computer, wherein the baseline physiological responses are associated with particular emotional states;

the processor determining the degree to which the person is engaged with, concentrated on, or focused on the stimulus; and

the processor identifying a likely emotional state in dependence on the pattern of deviations of the measured physiological responses from the acceptable range of the baseline physiological responses, in dependence on the degree to which the person is engaged with, concentrated on, or focused on the stimulus, and in dependence on the measured physiological responses being significantly similar to the baseline physiological responses, the baseline physiological response and the measured physiological response being defined as significantly similar when the likelihood that the similarity is due to chance is less than a selected threshold as defined through a measure of statistical significance of both standard deviation and instability,

wherein the measured physiological responses and baseline physiological responses comprise respiratory responses.

13. The monitoring system of claim 12, wherein the processor correlates a time of exposure to the stimulus and the type of stimulus with the measured physiological response.

14. A method for automatically identifying an emotional state of a person in real-time, comprising:

exposing a person to a stimulus;

measuring the person'"'"'s physiological responses to the stimulus with a physiological monitoring device configured to measure physiological responses comprising a respiratory and a cardiac response;

classifying the measured physiological responses into classes with a processor including a classifier trained to recognize patterns of physiological responses based on discrete periods of physiological data collected from a distribution of a population of people, and occurring in a plurality of emotional states, wherein training the classifier comprises measuring physiological responses, identifying discrete periods of physiological responses to the stimulus, selecting features to be used to identify each class, and grouping the selected features into groups that characterize the emotional states; and

identifying a likely emotional state with the processor in dependence on the classes into which the measured responses have been classified if the measured response matches one of the plurality of emotional states or falls within a predetermined deviational range of one of the plurality of emotional states according to both standard deviation and instability, wherein the emotional state comprises a measure of the degree to which the person is engaged with, concentrated on, or focused on the stimulus.

15. The method of claim 14 wherein a pattern of measured physiological responses comprises deviations of the measured responses from baseline physiological responses, the baseline physiological responses characteristic of an emotional state.

16. The method of claim 14, further comprising scaling determined feature vectors, applying dimension-reduction analysis to the determined feature vectors, or both scaling and applying dimension-reduction analysis to the determined feature vectors.

17. The method of claim 14, wherein classifying further comprises:

extracting determined feature vectors from the measured physiological responses, the determined feature vectors being those determined during classifier training; and

applying the trained classifier to the extracted feature vectors.

18. The method of claim 17 further comprising scaling the extracted feature vectors, or applying dimension-reduction analysis to the extracted feature vectors, or both scaling and applying dimension-reduction analysis to the extracted feature vectors.