# US 2023/0273682 A1 · Determinations of Characteristics from Biometric Signals > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2023/0273682 A1 - Document ID: US20230273682A1 - Status: Published application - Kind: A1 - Application number: 18/043,316 - Filed: 2020-09-08 - Published: 2023-08-31 - Earliest claimed: 2020-09-08 - Assignees: Hewlett-Packard Development Company, L.P. - Applicants: Hewlett-Packard Development Company, L.P. - Inventors: Jishang Wei; Rafael Ballagas; Keanu Spies - Web page: https://patentgrep.com/patent/US20230273682A1 ## CPC - A61B 5/7267: {involving training the classification device} - G06F 3/015: {Input arrangements based on nervous system activity detection, e.g. brain waves [EEG] detection, electromyograms [EMG] detection, electrodermal response detection} - A61B 5/165: {Evaluating the state of mind, e.g. depression, anxiety} - A61B 5/6803: {Head-worn items, e.g. helmets, masks, headphones or goggles} - A61B 5/7405: {using sound} - A61B 5/742: {using visual displays (displays for heart-related electrical signals, e.g. ECG, A61B5/339)} - G06N 3/0442: characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] - G06N 3/045: Combinations of networks - G06N 3/0455: Auto-encoder networks; Encoder-decoder networks - G06N 3/0464: Convolutional networks [CNN, ConvNet] - G06N 3/09: Supervised learning - G16H 40/60: for the operation of medical equipment or devices - G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems - G16H 50/70: for mining of medical data, e.g. analysing previous cases of other patients - A61B 5/021: Measuring pressure in heart or blood vessels - A61B 5/02416: {using photoplethysmograph signals, e.g. generated by infrared radiation (A61B5/14552 takes precedence)} - A61B 5/02438: {with portable devices, e.g. worn by the patient} - A61B 5/0245: by using sensing means generating electric signals, {i.e. ECG signals} - A61B 5/0533: Measuring galvanic skin response - A61B 5/0816: {Measuring devices for examining respiratory frequency (measuring frequency of electric signals G01R23/00)} - A61B 5/163: {by tracking eye movement, gaze, or pupil change} - A61B 5/318: Heart-related electrical modalities, e.g. electrocardiography [ECG] - A61B 5/389: Electromyography [EMG] - G06N 3/044: Recurrent networks, e.g. Hopfield networks - A61B: DIAGNOSIS; SURGERY; IDENTIFICATION (usefulness limited to only animals A61D) - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA ## Abstract An example system includes a plurality of biometric sensors to generate a plurality of signals. The system includes a feature engine to generate a plurality of feature vectors from the plurality of signals. The system includes a classifier engine to generate a plurality of decision vectors based on the plurality of feature vectors. The system includes an attention engine to weight the plurality of feature vectors and the plurality of decision vectors and determine a characteristic of a user based on the weighted feature and decision vectors. ## Claims ### Claim 1 A system comprising: a plurality of biometric sensors to generate a plurality of signals; a feature engine to generate a plurality of feature vectors from the plurality of signals; a classifier engine to generate a plurality of decision vectors based on the plurality of feature vectors; and an attention engine to: weight the plurality of feature vectors and the plurality of decision vectors; and determine a characteristic of a user based on the weighted feature and decision vectors. ### Claim 2 (depends on 1) The system of claim 1, wherein the feature engine includes a feature extraction engine to generate a first feature vector from a first signal and an autoencoder engine to generate a second feature vector from the first signal. ### Claim 3 (depends on 2) The system of claim 2, wherein the classifier engine is to generate a first decision vector based on the first feature vector without generating a decision vector based on the second feature vector. ### Claim 4 (depends on 1) The system of claim 1, further comprising a dimensionality engine to reduce a dimensionality of a feature vector to produce a feature vector with reduced dimensionality, wherein the attention engine is to weight the feature vector with reduced dimensionality. ### Claim 5 (depends on 1) The system of claim 1, further comprising a head-mounted display, wherein the system is to alter an audio or video output by the head-mounted display based on the determined characteristic of the user. ### Claim 6 A method, comprising: measuring a first biometric signal and a second biometric signal from a user of a head-mounted display; generating a first decision vector based on the first biometric signal; generating a second decision vector based on the second biometric signal; generating a third decision vector based on the first and second biometric signals; weighting the first, second, and third decision vectors; determining a characteristic of the user based on the weighted first, second, and third decision vectors; modifying audio or video content based on the determined characteristic; and delivering the modified audio or video content to the user of the head-mounted display. ### Claim 7 (depends on 6) The method of claim 6, wherein determining the characteristic of the user includes determining a cognitive load of the user. ### Claim 8 (depends on 7) The method of claim 7, wherein modifying the audio or video content comprises modifying the audio or video content to cause an increase or decrease in the cognitive load of the user toward a predetermined cognitive load. ### Claim 9 (depends on 6) The method of claim 6, further comprising determining a latent space representation of the first biometric signal, and weighting the latent space representation, wherein determining the characteristic comprises determining the characteristic based on the weighted latent space representation. ### Claim 10 (depends on 6) The method of claim 6, wherein weighting the first, second, and third decision vectors comprises weighting each of the first, second, and third decision vectors based on values of that vector. ### Claim 11 A non-transitory computer-readable medium comprising instructions that, when executed by a processor, cause the processor to: generate a first feature vector based on a first biometric signal; generate a second feature vector based on a second biometric signal; compute a decision vector based on the first feature vector using a classifier; calculate a first weighted vector based on the decision vector and a second weighted vector based on the second feature vector; and determine a characteristic of a user based on the first and second weighted vectors. ### Claim 12 (depends on 11) The computer-readable medium of claim 11, wherein the instructions to generate the second feature vector include instructions to cause the processor to use an autoencoder to generate the second feature vector based on the second biometric signal, and wherein the instructions to calculate the second weighted vector include instructions to cause the processor to calculate a reduced dimensionality vector from the second feature vector and apply a weight to the reduced dimensionality vector. ### Claim 13 (depends on 11) The computer-readable medium of claim 11, wherein the instructions to compute the decision vector include instructions to cause the processor to compute a first decision vector based on the first feature vector, compute a second decision vector based on the second feature vector, compute a third decision vector based on the first feature vector and the second feature vector, and compute a fourth decision vector based on the first decision vector and the second decision vector. ### Claim 14 (depends on 13) The computer-readable medium of claim 13, wherein the instructions to calculate the first weighted vector include instructions to cause the processor to calculate the first weighted vector based on the first decision vector, wherein the computer-readable medium further comprises instructions to cause the processor to calculate a third weighted vector based on the second decision vector, calculate a fourth weighted vector based on the third decision vector, and calculate a fifth weighted vector based on the fourth decision vector, and wherein the instructions to determine the characteristic of the user include instructions to cause the processor to determine the characteristic based on the first, second, third, fourth, and fifth weighted vectors. ### Claim 15 (depends on 11) The computer-readable medium of claim 11, wherein the instructions to compute the decision vector include instructions to cause the processor to compute a first decision vector using a first neural network with a first architecture and a second decision vector using a second neural network with a second architecture different from the first architecture, wherein the instructions to calculate the first weighted vector include instructions to cause the processor to calculate the first weighted vector based on the first decision vector, wherein the computer-readable medium further comprises instructions to cause the processor to calculate a third weighted vector based on the second decision vector, and wherein the instructions to determine the characteristic of the user include instructions to cause the processor to determine the characteristic based on the first, second, and third weighted vectors.