Computer system for group crisis-state detection and intervention

Abstract

The disclosed technology provides a system and a computer implemented method for crisis state detection and intervention of a person or group of persons, the method comprising: providing a computer system designed to detect and intervene non-normal, elevated crisis operating states; using one or more sensors that ascertains a crisis state via physical, behavioral, or cognitive indicators; deducing, with computational hardware, the operational state of a user or users from one or more sensors; and administering an immediate, dual intervention of a sensory form to de-escalate the crisis operating state of a person or group of persons.

What is claimed is: 
     
         1 . A system for continuous group anxiety-crisis state detection and intervention, the system comprising:
 a plurality of user devices collectively configured to:
 continuously detect, by a device sensor, biometric conditions specific to individual users of a group of users, each of the biometric conditions identifying a current health state of each user in the group of users; 
 receive, by a processor of each of the plurality of user devices, user input indicating at least one of (i) user-reviews of a workplace and (ii) user selection or interaction with a crisis intervention mobile application installed on the plurality of user devices, wherein the crisis intervention mobile application is configured to provide (i) real-time interventions to the user through at least one of a graphical user interface (GUI) display and at least one feedback device and (ii) an artificial intelligence (AI) coach that is trained to provide feedback to the user that is customized for the user and based on the current health state of the user, wherein the AI coach is only provided to the user device based on the user opting into sharing their continuously-detected biometric conditions with a computer system; 
 transmit, to the computer system, the continuously-detected biometric conditions and the user input; 
 perform, by the at least one feedback device of each of the plurality of user devices, one or more interventions generated by the computer system, wherein the one or more interventions comprise at least one of group interventions and individualized interventions; 
   environmental sensors positioned in one or more physical environments relative to locations of the group of users, the environmental sensors being separate from the plurality of mobile devices and the device sensor of each of the plurality of mobile devices, wherein the environmental sensors are configured to:
 continuously detect group conditions of the group of users in the one or more physical environments; 
 transmit, to the computer system, the continuously-detected group conditions; and 
   the computer system in network communication with the plurality of user devices and the environmental sensors, wherein the computer system is configured to detect a group anxiety-crisis state of the group of users and determine the one or more interventions to improve a current group health state of the group of users from the group anxiety-crisis state to a normal group health state of the group of users, wherein the computer system is configured to perform operations comprising:
 continuously receiving data from a plurality of data sources, the plurality of data sources comprising at least the plurality of user devices and the environmental sensors, wherein the data includes the continuously-detected biometric conditions, the user input, and the continuously-detected group conditions; 
 determining, based on processing the continuously-received data using a machine learning model, the current group health state of the group of users, wherein the machine learning model was trained to correlate different types of information about the group of users to different types of group health states; 
 determining that the current group health state of the group of users corresponds to the group anxiety-crisis state of the group of users based on the current group health state of the group of users exceeding a threshold group health state indicator for the group of users; 
 generating, in response to determining that the current group health state of the group of users corresponds to the group anxiety-crisis state of the group of users, the one or more interventions, wherein generating the one or more interventions comprises generating:
 (i) group intervention instructions that are configured to be automatically executed by one or more acting devices, the intervention instructions configured to produce actions on the one or more acting devices to lower the group of users collectively from the group anxiety-crisis state to the normal group health state, wherein the one or more acting devices include at least a portion of the plurality of user devices, one or more control devices configured to control aspects of the one or more physical environments, or combinations thereof, and 
 (ii) individualized intervention instructions that, when executed, cause another portion of the plurality of user devices associated with users that opted into sharing the respective continuously-detected biometric conditions with the computer system to present the AI coach in the respective GUI display, the AI coach having been trained by the computer system to provide suggestions indicating individualized actions that a user at each of the another portion of the plurality of user devices should perform to improve the current group health state of the group of users from the group anxiety-crisis state to the normal group health state; 
 
   transmitting, to the one or more acting devices and the another portion of the user devices, the group intervention instructions and the individualized intervention instructions, respectively, to be automatically executed; and   while the group intervention instructions and the individualized intervention instructions are executed by the respective one or more acting devices and the another portion of the user devices:
 iteratively (i) determining the current group health state of the group of users and (ii) modifying at least one of the group intervention instructions and the individualized intervention instructions based on processing the continuously-received data using the machine learning model, and 
 returning the iteratively modified group intervention instructions to at least one of the one or more acting devices and the another portion of the user devices until the computer system determines that the current group health state of the group of users corresponds to the normal group health state based on the current group health state being less than another threshold group health state indicator for the group of users. 
   
     
     
         2 . The system of  claim 1 , wherein the current group health state of the group of users comprises a physical environment score, a social environment score, and a group biometrics score. 
     
     
         3 . The system of  claim 2 , wherein the computer system is configured to perform operations further comprising:
 transmitting, to the plurality of user devices, instructions that, when executed by the plurality of user devices, cause the plurality of user devices to present the physical environment score, the social environment score, and the group biometrics score in the respective GUI displays.   
     
     
         4 . The system of  claim 3 , wherein, for the users associated with the another portion of the plurality of user devices that opted into sharing the continuously-detected biometric conditions with the computer system:
 the current group health state of the group of users further comprises an individual biometrics score, and   transmitting, to the another portion of the plurality of user devices, the instructions that, when executed by the another portion of the plurality of user devices, cause the another portion of the plurality of user devices to further present the individual biometrics score in the respective GUI displays.   
     
     
         5 . The system of  claim 1 , wherein the environmental sensors are configured to detect environmental conditions in the one or more physical environments that may impact the current group health state of the group of users. 
     
     
         6 . The system of  claim 1 , wherein the computer system is configured to continuously train the AI coach to improve the suggestions based at least in part on (i) the continuously-detected biometric conditions specific to users in the group of users that opted into sharing their continuously-detected biometric conditions and (ii) user feedback indicating user opinions about efficacy of the AI coach-provided suggestions in lowering the user from the group anxiety-crisis state to the normal group health state. 
     
     
         7 . The system of  claim 1 , wherein the machine learning model is a deep learning model. 
     
     
         8 . The system of  claim 1 , wherein the operations further comprise:
 determining, based on processing the continuously-received data using the machine learning model, that the current group health state of the group of users remains in the group anxiety-crisis state after a predetermined amount of time that the group intervention instructions and the individualized intervention instructions are executed at the respective one or more acting devices and the another portion of user devices; and   transmitting instructions to a computing device of a human coach, wherein the instructions, when executed at the computing device, prompt the human coach to intervene and help one or more users in the group of users reach the normal group health state.   
     
     
         9 . The system of  claim 1 , wherein the operations further comprise:
 determining periodic operating state measurements of the group of users based on the continuously-received data;   calibrating at least one of the group anxiety-crisis state, the normal group health state, or the threshold group health state indicator based on processing the periodic operating state measurements of the group of users using the machine learning model; and   updating at least one of the group anxiety-crisis state or the normal group health state based on the calibrating.   
     
     
         10 . The system of  claim 1 , wherein the operations further comprise:
 generating and transmitting, in response to determining that the current group health state corresponds to the normal group health state, positive reinforcement intervention instructions to the one or more acting devices,   wherein the positive reinforcement intervention instructions, when executed, cause at least the portion of the plurality of user devices amongst the one or more acting devices to present, in the crisis intervention mobile application at the respective GUI displays, positive reinforcement and a loyalty program for which respective users accrue points and rewards for engaging in positive behavior with other users in the group in the workplace.   
     
     
         11 . The system of  claim 1 , wherein the intervention instructions are automatically executed by the one or more acting devices without user input or user authorization. 
     
     
         12 . The system of  claim 1 , wherein the operations further comprise:
 anonymizing the continuously-detected biometric conditions and the user input to generate anonymized user data, wherein the continuously-detected biometric conditions associated with users that opted into sharing the respective continuously-detected biometric conditions are excluded from the continuously-detected biometric conditions that are anonymized to generate the anonymized user data;   determining, based on processing the anonymized user data using the machine learning model, the current group health state of the group of users; and   generating, based on processing the anonymized user data, the group intervention instructions.   
     
     
         13 . The system of  claim 12 , wherein anonymizing the continuously-detected biometric conditions and the user input comprises aggregating the continuously-detected biometric conditions and the user input to a group-state metric, wherein the group-state metric is used to determine whether the current group health state exceeds the threshold group health state indicator for the group of users. 
     
     
         14 . The system of  claim 1 , wherein the at least one feedback device comprises a haptic feedback device. 
     
     
         15 . The system of  claim 1 , wherein the at least one feedback device comprises a sensory feedback device. 
     
     
         16 . The system of  claim 1 , wherein the one or more acting devices are configured to uniformly execute the intervention instructions for the group of users in the workplace. 
     
     
         17 . The system of  claim 1 , the operations further comprising:
 retrieve, from a database, historic data corresponding to the group of users and the workplace; and   determine, based on processing the continuously-received data and the historic data using the machine learning model, the current group health state of the group of users.   
     
     
         18 . The system of  claim 1 , wherein the group intervention instructions, when automatically executed by the one or more acting devices, cause the portion of the plurality of user devices to present a group AI coach in the respective GUI displays, the group AI coach having been trained by the computer system to provide uniform, group suggestions indicating group actions that the group of users should perform to improve the current group health state of the group of users from the group anxiety-crisis state to the normal group health state. 
     
     
         19 . The system of  claim 1 , wherein the group intervention instructions, when automatically executed by the one or more acting devices, cause the one or more control devices configured to control aspects of the one or more physical environments to adjust one or more settings in the respective one or more physical environments. 
     
     
         20 . The system of  claim 1 , wherein the AI coach is associated with a module of the crisis intervention mobile application and configured to interface with other modules of the crisis intervention mobile application, wherein the other modules each provide different types of interventions for helping the group of users achieve the normal group health state of the group of users.