Automated selection and titration of sensory stimuli to induce a target pattern of autonomic nervous system activity

Abstract

Systems and techniques are disclosed for aspects of processing physiological signals in connection with the selection and output of sensory stimuli. In an example, a system for modulating the cognitive state or physiological activity of a human subject may be used in a variety of settings, such as for offering exposure therapy or other forms of guided psychological treatment, based on the presentation and control of a sensory stimulus to a human subject. In another example, a system for generating or training a physiological response model may be established to classify or identify a cognitive response state from sensory stimulus, such as media content, based on the identification of a physiologic or cognitive response state produced from physiological response signals during exposure to the sensory stimulus.

What is claimed is: 
     
         1 . A system, comprising:
 at least one processor; and   at least one memory device comprising instructions, which when executed by the processor, causes the processor to perform operations that:   select a sensory stimulus for delivery to a human subject, using a decision algorithm adapted to predict an expected physiological response signal to the sensory stimulus;   monitor the human subject to obtain physiological response signals induced by delivery of the sensory stimulus;   evaluate the physiological response signals induced by the delivery of the sensory stimulus; and   control the delivery of the sensory stimulus, based on the evaluated physiological response signals relative to a desired outcome.   
     
     
         2 . The system of  claim 1 , wherein the system is adapted for modulating a cognitive state of the human subject, and wherein the desired outcome is a desired cognitive state. 
     
     
         3 . The system of  claim 1 , wherein the system is adapted for modulating physiological activity of the human subject, and wherein the desired outcome is a desired physiological response. 
     
     
         4 . The system of  claim 3 , wherein the instructions further perform operations that:
 select a scene of media content, for presentation to the human subject, using the decision algorithm within a trained classification model, the trained classification model adapted to predict an expected physiological response to a stimulus in the scene of media content;   obtain the physiological response signals from monitoring of the human subject during presentation of the scene of media content;   evaluate the physiological response signals in response to the presentation of the scene of media content; and   control the presentation of the scene of media content, based on the evaluated physiological response signals relative to the expected physiological response.   
     
     
         5 . The system of  claim 4 , the instructions further cause the processor to perform operations that:
 identify, based on the evaluated physiological response signals, a scaled response score, the scaled response score providing a measurement of the response to the stimulus from the presentation of the scene of media content;   wherein the presentation of the scene of media content is controlled based on a comparison of the scaled response score to a goal response score for the human subject.   
     
     
         6 . The system of  claim 5 , wherein the comparison of the scaled response score to the goal response score is based on a comparison of at least one of: a target response magnitude occurring from presentation of the scene of media content, a target response decay occurring from presentation of the scene of media content, or a target response score occurring from presentation of the scene of media content. 
     
     
         7 . The system of  claim 4 , the instructions further to cause the processor to perform operations that:
 obtain prior physiological response signals being obtained from before the presentation of the scene of media content; and   wherein evaluating the physiological response signals is based on comparing the prior physiological response signals with the physiological response signals obtained during presentation of the scene of media content.   
     
     
         8 . The system of  claim 4 , wherein the scene of media content is selected based on a cognitive state goal for the human subject, and wherein the cognitive state goal defines a quantified cognitive state that is associated with a plurality of physiological measurements. 
     
     
         9 . The system of  claim 4 , wherein the classification model is trained based on a plurality of scenes of media content having corresponding physiological responses, and wherein the classification model is untrained on the scene of media content provided in the presentation to the human subject. 
     
     
         10 . The system of  claim 4 , wherein the scene of media content is selected from a media library based on a plurality of scene parameters corresponding to stimulus measurements, the scene parameters provided from among: intensity, duration, excitatory or sedative valence, or subject characteristics of the scene of media content. 
     
     
         11 . The system of  claim 4 , wherein the physiological response signals are based on a plurality of signals obtained from monitoring the human subject with a respective plurality of sensors. 
     
     
         12 . A system for training a media content model, comprising:
 at least one processor; and   at least one memory device comprising instructions, which when executed by the processor, causes the processor to perform operations that:
 capture physiological response signals from a human subject, the physiological response signals produced by exposing the human subject to a sensory stimulus; 
 identify, from the physiological response signals, a response state of the human subject; and 
 train a model to associate the response state with a scene of media content. 
   
     
     
         13 . The system of  claim 12 , wherein the response state is a physiologic response state, and wherein the model is trained for analyzing a physiologic response of the human subject. 
     
     
         14 . The system of  claim 12 , wherein the response state is a cognitive response state, wherein the model is trained for analyzing a cognitive response of the human subject. 
     
     
         15 . The system of  claim 14 , the instructions further to cause the processor to perform operations that:
 train the model to associate the cognitive response state with a type of treatment.   
     
     
         16 . The system of  claim 15 , the instructions further to cause the processor to perform operations that:
 monitor the human subject during presentation of the scene of media content, to identify the physiological response signals induced by the sensory stimulus;   identify, from the physiological response signals, the cognitive response state of the human subject; and   train the model to associate the cognitive response state with the scene of media content.   
     
     
         17 . The system of  claim 16 , wherein the scene of media content is provided from a video clip, with the instructions further to cause the processor to perform operations that identify the scene of media content from a portion of the video clip; and
 wherein the operations are performed with at least a second scene of the video clip, to train the model to associate a second identified cognitive response state with the second scene of the video clip.   
     
     
         18 . The system of  claim 16 , wherein the scene of media content is selected by a clinician as part of an identified treatment. 
     
     
         19 . The system of  claim 16 , wherein the association of the cognitive response state is established based on a plurality of scene parameters corresponding to the physiological response signals, the scene parameters provided from among: intensity, duration, excitatory or sedative valence, or subject characteristics of the scene of media content. 
     
     
         20 . The system of  claim 16 , wherein the model is trained based on response characteristics of the physiological response signals, the response characteristics of the physiological response signals defining one or more of: peak of the response, magnitude of the response, direction of the response, valence of the response, time course of the response, transition of the response, and total measurement of the response.