# US 2023/0263441 A1 · Sensor device-based detection of trauma events and responses > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2023/0263441 A1 - Document ID: US20230263441A1 - Status: Published application - Kind: A1 - Application number: 17/677,083 - Filed: 2022-02-22 - Published: 2023-08-24 - Earliest claimed: 2022-02-22 - Assignees: AT&T Intellectual Property I, L.P. - Applicants: AT&T Intellectual Property I, L.P. - Inventors: Amee Sheth; Laura Nordyke; Jay Shah - Web page: https://patentgrep.com/patent/US20230263441A1 ## CPC - A61B 5/165: {Evaluating the state of mind, e.g. depression, anxiety} - G06N 3/09: Supervised learning - A61B 5/1128: {using image analysis (A61B5/1127 takes precedence)} - A61B 5/6803: {Head-worn items, e.g. helmets, masks, headphones or goggles} - A61B 5/7267: {involving training the classification device} - A61B 5/7275: {Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor} - A61B 5/7465: {Arrangements for interactive communication between patient and care services, e.g. by using a telephone network (telemetry of measured physiological signal A61B5/0002)} - A61B 5/7475: {User input or interface means, e.g. keyboard, pointing device, joystick} - G06N 3/0464: Convolutional networks [CNN, ConvNet] - G06N 20/20: Ensemble learning - G06N 3/0475: Generative networks - G06N 3/092: Reinforcement learning - G06N 3/094: Adversarial learning - G06N 20/10: using kernel methods, e.g. support vector machines [SVM] - G16Y 40/10: Detection; Monitoring - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G16Y: INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR THE INTERNET OF THINGS [IoT] ## Abstract A processing system including at least one processor may obtain data associated with a user, the data associated with the user including at least one of: visual data captured via at least one camera associated with the user or audio data captured via at least one microphone associated with the user, detect at least one trauma event of at least one defined trauma event type in at least one of the visual data or the audio data via at least one classification model, determine, responsive to detecting the at least one trauma event, a stress score based upon at least a portion of the data associated with the user in accordance with a stress prediction model, and generate an alert in response to the stress score exceeding a threshold. ## Claims ### Claim 1 A method comprising: obtaining, by a processing system including at least one processor, data associated with a user, wherein the data associated with the user includes at least one of: visual data captured via at least one camera associated with the user or audio data captured via at least one microphone associated with the user; detecting, by the processing system, at least one trauma event of at least one defined trauma event type in at least one of: the visual data or the audio data via at least one classification model; determining, by the processing system responsive to the detecting of the at least one trauma event, a stress score based upon at least a portion of the data associated with the user in accordance with a stress prediction model; and generating, by the processing system, an alert in response to the stress score exceeding a threshold. ### Claim 2 (depends on 1) The method of claim 1, further comprising: identifying a plurality of stressor indicators within the data associated with the user, wherein the determining the stress score comprises determining the stress score based upon the plurality of stressor indicators in accordance with the stress prediction model. ### Claim 3 (depends on 1) The method of claim 1, wherein the at least one defined trauma event type is one of a plurality of defined trauma event types. ### Claim 4 (depends on 1) The method of claim 1, wherein the at least one classification model is trained to detect trauma events of the at least one defined trauma event type. ### Claim 5 (depends on 1) The method of claim 1, wherein the data associated with the user further comprises at least one of: location data of the user; environmental data of an environment of the user; or biometric data of the user. ### Claim 6 (depends on 1) The method of claim 1, wherein the data associated with the user further comprises calendar data of the user. ### Claim 7 (depends on 6) The method of claim 6, wherein the calendar data includes at least one of: a quantity of time at work for the user within a designated time period; a commute time of the user within the designated time period; or at least one personal event of the user. ### Claim 8 (depends on 1) The method of claim 1, wherein the data associated with the user further comprises a quantity of time on screen for the user within a designated time period. ### Claim 9 (depends on 1) The method of claim 1, wherein the user is of a designated category of one of: a first responder; a medical professional; or a professional vehicle operator. ### Claim 10 (depends on 1) The method of claim 1, wherein the at least the portion of the data associated with the user is from at least one designated time period in relation to the at least one trauma event. ### Claim 11 (depends on 10) The method of claim 10, wherein the at least one designated time period precedes the at least one trauma event, includes the at least one trauma event, or follows the at least one trauma event. ### Claim 12 (depends on 1) The method of claim 1, wherein the stress prediction model comprises a machine learning model trained on a plurality of sets of data associated with a plurality of users. ### Claim 13 (depends on 12) The method of claim 12, wherein each of the plurality of sets of data is labeled with one of: a respective stress score; or a respective label of excess stress or non-excess stress. ### Claim 14 (depends on 1) The method of claim 1, wherein the stress prediction model comprises a regression model. ### Claim 15 (depends on 1) The method of claim 1, wherein the stress prediction model comprises a formula-based model that weights the plurality of stressor indicators. ### Claim 16 (depends on 1) The method of claim 1, wherein the at least one camera associated with the user comprises at least one of: a body-worn camera; or passenger. ### Claim 17 (depends on 1) The method of claim 1, wherein the processing system comprises at least one server, or an endpoint device of the user. ### Claim 18 (depends on 1) The method of claim 1, wherein the alert is provided to at least one of: the user via an endpoint device of the user; or a designated authorized entity. ### Claim 19 A non-transitory computer-readable medium storing instructions which, when executed by a processing system including at least one processor, cause the processing system to perform operations, the operations comprising: obtaining data associated with a user, wherein the data associated with the user includes at least one of: visual data captured via at least one camera associated with the user or audio data captured via at least one microphone associated with the user; detecting at least one trauma event of at least one defined trauma event type in at least one of the visual data or the audio data via at least one classification model; determining, responsive to the detecting of the at least one trauma event, a stress score based upon at least a portion of the data associated with the user in accordance with a stress prediction model; and generating an alert in response to the stress score exceeding a threshold. ### Claim 20 An apparatus comprising: a processing system including at least one processor; and a computer-readable medium storing instructions which, when executed by the processing system, cause the processing system to perform operations, the operations comprising: obtaining data associated with a user, wherein the data associated with the user includes at least one of: visual data captured via at least one camera associated with the user or audio data captured via at least one microphone associated with the user; detecting at least one trauma event of at least one defined trauma event type in at least one of the visual data or the audio data via at least one classification model; determining, responsive to the detecting of the at least one trauma event, a stress score based upon at least a portion of the data associated with the user in accordance with a stress prediction model; and generating an alert in response to the stress score exceeding a threshold.