# US 2023/0056100 A1 · Systems and methods for dynamic biometric control of iot devices

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2023/0056100 A1
- Document ID: US20230056100A1
- Status: Published application
- Kind: A1
- Application number: 17/404,617
- Filed: 2021-08-17
- Published: 2023-02-23
- Earliest claimed: 2021-08-17
- Inventors: Robin H. Stewart; Qiliang Li
- Web page: https://patentgrep.com/patent/US20230056100A1

## CPC

- G05B 13/0265: {the criterion being a learning criterion}
- G05B 15/02: electric
- A61B 5/0022: {Monitoring a patient using a global network, e.g. telephone networks, internet}
- A61B 5/165: {Evaluating the state of mind, e.g. depression, anxiety}
- A61B 5/4836: {Diagnosis combined with treatment in closed-loop systems or methods (A61B5/0036 takes precedence)}
- A61B 5/486: {Biofeedback (using electroencephalography [EEG] A61B5/375)}
- A61B 5/681: {Wristwatch-type devices}
- A61B 5/7267: {involving training the classification device}
- G06N 20/00: Machine learning
- G16Y 30/00: IoT infrastructure
- A61B 5/02055: {Simultaneously evaluating both cardiovascular condition and temperature}
- 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 5/14532: {for measuring glucose, e.g. by tissue impedance measurement}
- A61B 5/14551: {for measuring blood gases}
- A61B 5/6823: {Trunk, e.g., chest, back, abdomen, hip}
- A61B 5/6824: {Arm or wrist}
- A61B 5/6826: {Finger}
- A61B 2503/12: Healthy persons not otherwise provided for, e.g. subjects of a marketing survey
- A61B 2560/0223: of calibration, e.g. protocols for calibrating sensors
- G05B 2219/2642: Domotique, domestic, home control, automation, smart house
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G)
- 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

Systems and methods of dynamic IoT device regulation and control can aid in shifting a user's emotional state from a first state of mind to a preferred second state of mind, using the user's biomarker response to device settings. Particularly, an IoT device controller may be embedded within a wearable device that is wirelessly connected to a computing device and one or more IoT devices. Initially, each wearable device can be calibrated, wherein a matrix of sensed user biomarker responses can be generated. In some embodiments, the system continuously monitors user biomarkers to detect which physiological state exists. When the user enters into the first physiological/psychological state, the system can adjust each IoT device to align with the second state. When the system detects that the user biomarker response has not shifted, the system can continuously adjust IoT settings based upon a learning algorithm having monitored user biomarkers as input.

## Claims

### Claim 1

A method of dynamic biometric regulation of one or more IoT devices, comprising:
calibrating a dynamic IoT device setting module with user biomarker response to one or more IoT device settings;
monitoring one or more user biomarkers using the dynamic IoT device setting module;
storing the one or more user biomarkers in a user profile matrix;
detecting a first physiological state based upon the user profile matrix;
adjusting, in response to the existence of the first physiological state, one or more IoT device settings to settings associated with a second physiological state;
sensing user biomarker response to the adjusted settings;
determining if the sensed user biomarkers align with the second physiological state; and
adjusting, in response to nonalignment of the sensed biomarkers with the second physiological state, the one or more IoT devices based upon a learning algorithm until alignment of the sensed biomarkers with the second physiological state occurs.

### Claim 2 (depends on 1)

The method of claim 1, wherein calibrating a dynamic IoT device setting module comprises,
receiving user input defining the first physiological state and the second physiological state;
retrieving a first set of model biomarkers associated with the first physiological state and a second set of model biomarkers associated with the second physiological state from a model database;
retrieving one or more IoT device settings associated with the first physiological state and with the second physiological state from a model database;
storing user input, the first set of model biomarkers, the second set of model biomarkers, and the retrieved IoT device settings in the user profile matrix;
adjusting one or more IoT device settings to the retrieved one or more IoT device settings associated with the second physiological state;
monitoring user biomarker response to the adjusted one or more IoT device settings;
detecting if the monitored user biomarker response aligns with the second set of model biomarkers associated with the second physiological state;
adjusting, in response to nonalignment of the monitored user biomarkers, one or more IoT devices based upon the retrieved one or more IoT device settings associated with the second physiological state until the monitored user biomarker response aligns with the second physiological state; and
storing user biomarker response in the user profile matrix.

### Claim 3 (depends on 1)

The method of claim 1, wherein monitoring one or more user biomarkers comprises,
retrieving the user's temperature from a thermistor;
retrieving the user's blood pressure from a heart rate sensing unit having a deep learning algorithm;
retrieving the user's blood oxygen level from a pulse oximeter blood oxygen sensor;
retrieving the user's pulse from an optical heart sensor;
retrieving the user's blood glucose levels from a non-invasive glucose monitor; and
repeating the retrieving until a stop command is issued.

### Claim 4 (depends on 3)

The method of claim 3, wherein storing one or more user biomarkers in a user profile matrix comprises,
storing the user temperature, blood pressure, blood oxygen level, blood glucose levels, and pulse in the user profile matrix.

### Claim 5 (depends on 1)

The method of claim 1, wherein detecting a first physiological state based upon the user profile matrix comprises,
retrieving a first set of model biomarkers associated with the first physiological state from the user profile matrix;
comparing the monitored one or more user biomarkers with the first set of model biomarkers;
detecting a match of the monitored one or more user biomarkers with the first set of model biomarkers; and
setting a first flag, when the match occurs.

### Claim 6 (depends on 1)

The method of claim 1, wherein adjusting the one or more IoT device settings associated with a second physiological state comprises,
retrieving the one or more IoT device settings associated with the second physiological state from the user profile matrix; and
modifying the setting of the one or more IoT devices to match the retrieved one or more IoT device settings.

### Claim 7 (depends on 6)

The method of claim 6, wherein modifying the setting of the one or more IoT devices comprises:
adjusting a thermostat to align with one or more temperature settings of the second physiological state from the user profile matrix;
changing a displayed image on an electronic surface to align with the second physiological state;
changing a broadcasted audio to align with the second physiological state;
changing a broadcasted video to align with the second physiological state;
adjusting a light switch dimmer to align with the second physiological state; and
updating user profile matrix.

### Claim 8 (depends on 1)

The method of claim 1, wherein determining if the sensed biomarkers align with the second physiological state comprises,
retrieving a second set of model biomarkers associated with the second physiological state from the user profile matrix;
comparing the sensed user biomarker response with the second set of model biomarkers;
detecting a match of the sensed user biomarker response with the second set of model biomarkers; and
setting a second flag, when the match occurs.

### Claim 9 (depends on 1)

The method of claim 1, wherein adjusting one or more IoT devices based upon a learning algorithm until alignment of the sensed biomarkers with the second physiological state occurs comprises,
retrieving a model set of user biomarker responses associated with the second physiological state from a datastore;
extracting a set of one or more abridged matrices from the user profile matrix;
selecting a first matrix from the extracted set;
processing the first matrix using a Convolution Neural Network (CNN);
extracting a second matrix from the CNN processed matrix;
detecting the similarity between the second matrix and the retrieved model set of user biomarker responses using a pattern recognition technique;
projecting a next set of one or more IoT device settings based upon the similarity;
modifying the one or more IoT devices settings to match the next set;
sensing user biomarker response to the modified IoT device settings;
determining if the sensed user biomarkers align with the model set of user biomarker responses; and
repeating the projecting, modifying, sensing, and determining step until sensed user biomarkers align with the second physiological state.

### Claim 10 (depends on 9)

The method of claim 9, wherein the pattern recognition technique comprises Principal Component Analysis (PCA).

### Claim 11 (depends on 9)

The method of claim 9, wherein the pattern recognition technique comprises Linear Discriminate Analysis (LDA).

### Claim 12

A dynamic physiological regulation system comprising:
a memory; and
a processor coupled to the memory, the processor operable to:
calibrate a dynamic IoT device setting module with user biomarker response to one or more IoT device settings defined within a user profile matrix;
monitor one or more user biomarkers;
store the one or more user biomarkers in a user profile matrix;
detect a first physiological state based upon the user profile matrix;
adjust, in response to the existence of the first physiological state, one or more IoT device settings to settings associated with a second physiological state;
sense user biomarker response to the adjusted settings;
determine if the sensed user biomarkers align with the second physiological state; and
adjust, in response to nonalignment of the sensed biomarkers with the second physiological state, the one or more IoT devices based upon a learning algorithm until alignment of the sensed biomarkers with the second physiological state occurs.

### Claim 13 (depends on 12)

The dynamic physiological regulation system of claim 12, wherein the processor, for calibrating a dynamic IoT device setting module operable to:
receive user input defining the first physiological state and the second physiological state;
retrieve a first set of model biomarkers associated with the first physiological state and a second set of model biomarkers associated with the second physiological state from a model database;
retrieve one or more IoT device settings associated with the first physiological state and with the second physiological state from a model database;
store user input, the first set of model biomarkers, the second set of model biomarkers, and the retrieved IoT device settings in the user profile matrix;
adjust one or more IoT device settings to the retrieved one or more IoT device settings associated with the second physiological state;
monitor user biomarker response to the adjusted one or more IoT device settings;
detect if the monitored user biomarker response aligns with the second set of model biomarkers associated with the second physiological state;
adjust, in response to nonalignment of the monitored user biomarkers, one or more IoT devices based upon the retrieved one or more IoT device settings associated with the second physiological state until the monitored user biomarker response aligns with the second physiological state; and
store user biomarker response in the user profile matrix.

### Claim 14 (depends on 12)

The dynamic physiological regulation system of claim 12, wherein the processor, for adjusting one or more IoT devices based upon a learning algorithm until alignment of the sensed biomarkers with the second physiological state occurs operable to:
retrieve input from a clinician monitored client node;
retrieve a set of model biomarkers associated with the second physiological state from a clinical datastore;
extract a set of one or more abridged matrices of biomarkers from the user profile matrix;
selecting a first matrix from the extracted set;
process the first matrix using a CNN;
extract a second matrix from the CNN processed matrix;
detect the difference between the second matrix and the set of model biomarkers using a pattern recognition technique;
derive an IoT device setting based upon the difference;
modify an associated one of the one or more IoT devices to match the derive IoT device setting;
sense user biomarker response to the modified IoT device setting;
determine if the sensed user biomarkers align with the second physiological state; and
repeating the deriving, modifying, sensing, and determining step until sensed user biomarkers align with the second physiological state.

### Claim 15

A non-transitory computer-readable medium including code for performing a method of dynamic physiological regulation, the method comprising:
calibrating a dynamic IoT device setting module with user biomarker response to one or more IoT device settings defined within a user profile matrix;
monitoring one or more user biomarkers;
storing the one or more user biomarkers in a user profile matrix;
detecting a first physiological state based upon the user profile matrix;
adjusting, in response to the existence of the first physiological state, one or more IoT device settings to settings associated with a second physiological state;
sensing user biomarker response to the adjusted settings;
determining if the sensed user biomarkers align with the second physiological state; and
adjusting, in response to nonalignment of the sensed biomarkers with the second physiological state, the one or more IoT devices based upon a learning algorithm until alignment of the sensed biomarkers with the second physiological state occurs.

### Claim 16 (depends on 15)

The computer-readable medium of claim 15, wherein calibrating a dynamic IoT device setting module comprises,
receiving user input defining the first physiological state and the second physiological state;
retrieving a first set of model biomarkers associated with the first physiological state and a second set of model biomarkers associated with the second physiological state from a model database;
retrieving one or more IoT device settings associated with the first physiological state and with the second physiological state from a model database;
storing user input, the first set of model biomarkers, the second set of model biomarkers, and the retrieved IoT device settings in the user profile matrix;
adjusting one or more IoT device settings to the retrieved one or more IoT device settings associated with the second physiological state;
monitoring user biomarker response to the adjusted one or more IoT device settings;
detecting if the monitored user biomarker response aligns with the second set of model biomarkers associated with the second physiological state;
adjusting, in response to nonalignment of the monitored user biomarkers, one or more IoT devices based upon the retrieved one or more IoT device settings associated with the second physiological state until the monitored user biomarker response aligns with the second physiological state; and
storing user biomarker response in the user profile matrix.

### Claim 17 (depends on 15)

The computer-readable medium of claim 15, wherein monitoring one or more user biomarkers comprises,
retrieving the user's temperature from a thermistor;
retrieving the user's blood pressure from a heart rate sensing unit having a deep learning algorithm;
retrieving the user's blood oxygen level from a pulse oximeter blood oxygen sensor;
retrieving the user's pulse from an optical heart sensor;
retrieving the user's blood glucose levels from a non-invasive glucose monitor; and
repeating the retrieving until a stop command is issued.

### Claim 18 (depends on 15)

The computer-readable medium of claim 15, wherein detecting a first physiological state based upon the user profile matrix comprises,
retrieving a first set of model biomarkers associated with the first physiological state from the user profile matrix;
comparing the monitored one or more user biomarkers with the first set of model biomarkers;
detecting a match of the monitored one or more user biomarkers with the first set of model biomarkers; and
setting a first flag, when the match occurs.

### Claim 19 (depends on 15)

The computer-readable medium of claim 15, wherein determining if the sensed biomarkers align with the second physiological state comprises,
retrieving a second set of model biomarkers associated with the second physiological state from the user profile matrix;
comparing the sensed user biomarker response with the second set of model biomarkers;
detecting a match of the sensed user biomarker response with the second set of model biomarkers; and
setting a second flag, when the match occurs.

### Claim 20 (depends on 15)

The computer-readable medium of claim 15, wherein adjusting one or more IoT devices based upon a learning algorithm until alignment of the sensed biomarkers with the second physiological state occurs comprises,
retrieving a set of model biomarkers associated with the second physiological state from a datastore;
extracting a set of one or more abridged matrices from the user profile matrix;
selecting a first matrix from the extracted set;
processing the first matrix using a CNN;
extracting a second matrix from the CNN processed matrix;
detecting the difference between the second matrix and the user profile matrix using a pattern recognition technique;
projecting a model set of one or more IoT device settings based upon the difference;
modifying the one or more IoT devices settings to match the model set;
sensing user biomarker response to the modified IoT device settings;
determining if the sensed user biomarkers align with the second physiological state; and
repeating the projecting, modifying, sensing, and determining step until sensed user biomarkers align with the second physiological state.
