# US 2015/0106305 A1 · Intelligent apparatus for providing personalized configuration of wheelchair tilt and recline

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

- Number: US 2015/0106305 A1
- Document ID: US20150106305A1
- Status: Published application
- Kind: A1
- Application number: 14/304,758
- Filed: 2014-06-13
- Published: 2015-04-16
- Earliest claimed: 2013-10-16
- Assignees: University of Central Oklahoma
- Applicants: University of Central Oklahoma
- Inventors: Jicheng FU
- Web page: https://patentgrep.com/patent/US20150106305A1

## CPC

- G06N 3/08: Learning methods
- A61G 5/1056: {Arrangements for adjusting the seat}
- G06F 30/20: Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F30/30)
- G06F 30/27: using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
- G06N 3/0499: Feedforward networks
- G06N 3/09: Supervised learning
- G06N 20/00: Machine learning
- G16H 50/20: for computer-aided diagnosis, e.g. based on medical expert systems
- A61G 5/1067: {adjusting the backrest relative to the seat portion}
- A61G 5/1075: {tilting the whole seat backwards (tilting forwards for standing-up or sitting-down aids A61G5/14)}
- A61G 7/057: Arrangements for preventing bed-sores or for supporting patients with burns, e.g. mattresses specially adapted therefor {(ventilation openings in mattresses A47C21/042; pneumatic or hydraulic mattresses A47C27/08)}
- A61G 2203/10: characterised by specific control means, e.g. for adjustment or steering
- G06N 3/10: Interfaces, programming languages or software development kits, e.g. for simulating neural networks
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- A61G: TRANSPORT, PERSONAL CONVEYANCES OR ACCOMMODATION SPECIALLY ADAPTED FOR PATIENTS OR DISABLED PERSONS (appliances for aiding patients or disabled persons to walk A61H3/00); OPERATING TABLES OR CHAIRS; CHAIRS FOR DENTISTRY; FUNERAL DEVICES (embalming corpses A01N1/00)
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G16H: HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA

## Abstract

A method and apparatus for providing personalized configuration of physical supports for the human body, comprising accepting input including an individual's demographic information, neurological attributes, physical history, operational environment, and outcome or use objectives, processing user input employing an artificial intelligence engine, and then returning guidance and/or control parameters directed to seating adjustment and positioning, including incline angles for wheelchair tilt and recline.

## Claims

### Claim 1

A method for providing personalized configuration of physical supports for the human body, comprising:
providing a mobile device removably mountable on a physical support, adapted to at least collect user information, request guidance on position parameters of said physical support, display results, and receive orientation data produced by an angle detection device to measure at least tilt and recline angles of said physical support;
providing an artificial intelligence module adapted to operate in an Internet cloud, including an artificial neural network (ANN) having a layered network structure in which processing units (i.e., neurons) are arranged in layers, where the neurons in adjacent layers can communicate with each other by sending and receiving signals through weighted connections;
accepting input including an individual's demographic information, neurological attributes, physical history, operational environment, and outcome or use objectives;
processing said input employing an artificial intelligence technique, and then returning at least one of guidance and control parameters directed to adjustment and positioning of said physical support, including at least one of tilt and recline angles;
wherein behavior of a neuron in said ANN is defined by its internal activation function, which accumulates input signals and then calculates outputs, and
wherein a learning process proceeds in iterations by tuning weights of connections using a training algorithm.

### Claim 2 (depends on 1)

The method of claim 1, wherein the method is implemented as a specific purpose mobile device comprising a computational framework, artificial neural network, a goniometer, and controls for positioning and adjustment settings directed to seating supports for the human body.

### Claim 3 (depends on 1)

The method of claim 1, further comprising executing functions and analytical processes capable of finding patterns within data that may contain noise, and analyzing non-linear and dependent data.

### Claim 4 (depends on 1)

The method of claim 1, wherein said artificial neural network is embodied as an artificial intelligence (AI) module trained with clinical research data directed to optimal positioning and adjustment of physical supports for the human body for a defined purpose or desired outcome.

### Claim 5 (depends on 1)

The method of claim 1, wherein users input into a user interface demographic, neurological, and pressure ulcer history information for an individual, and recommended tilt and recline angles are determined favorable to reduce risk of pressure ulcers.

### Claim 6

An apparatus for providing personalized configuration of physical supports for the human body, comprising:
a user interface for accepting input including at least one of an individual's demographic information, neurological attributes, physical history, operational environment, and outcome or use objectives;
an artificial intelligence module adapted to operate in an Internet cloud that evaluates user input and returns at least one of guidance and control parameters directed to at least one of seating adjustment and support positioning, including at least one of tilt and recline angles;
an artificial neural network (ANN) having a layered network structure, in which processing units (i.e., neurons) are arranged in layers, where the neurons in adjacent layers can communicate with each other by sending and receiving signals through weighted connections, and behavior of a neuron is defined by its internal activation function, which accumulates input signals and then calculates outputs;
a mobile device adapted to at least collect said user information, request guidance on support position parameters, display results, and receive input captured by an angle measuring device.

### Claim 7 (depends on 6)

The apparatus of claim 6, further comprising a specific purpose device integrated into a powered wheelchair, where the device includes a computational framework, artificial neural network, a goniometer, and controls for actuating adjustment of at least one of wheelchair tilt and recline.

### Claim 8 (depends on 7)

The apparatus of claim 7, wherein said goniometer measures current wheelchair tilt and recline angle and contrasts those angles with guidance angles to generate control parameters that cause the tilt and recline angle of a powered wheelchair to be rotated to a precise angular position.

### Claim 9 (depends on 6)

The apparatus of claim 6, further comprising a user interface configured to display recommended wheelchair tilt and recline angles, or other position parameters.

### Claim 10 (depends on 9)

The apparatus of claim 9, wherein recommended tilt and recline angles may be output wirelessly to a control function operating in a powered wheelchair to adjust seating orientation of said powered wheelchair.

### Claim 11

(canceled)

### Claim 12 (depends on 16)

The apparatus of claim 16, wherein said goniometer uses an accelerometer sensor configured in said mobile device to measure angles relative to any physical orientation in which said mobile device is positioned, enabling measurement of at least one of tilt and recline angles.

### Claim 13 (depends on 16)

The apparatus of claim 16, wherein a user registration component is provided which allows users to create and review profiles comprising at least one of demographic information, neurological information, and pressure ulcer history.

### Claim 14

(canceled)

### Claim 15 (depends on 16)

The apparatus of claim 16, wherein output includes: a range of tilt and recline angles that are favorable for pressure reduction for the user; optimal tilt and recline angles that are most effective in reducing the risk of pressure ulcers; and the optimal frequency and duration to perform wheelchair tilt and recline functions.

### Claim 16

An apparatus for determining optimal positioning for wheelchair seating support orientation, comprising:
a user interface on a mobile device that accepts input including an individual's demographic information, neurological attributes, and injury history, said mobile device adapted to at least collect user information, request guidance on wheelchair tilt and recline usage or other position parameters, display results to the user, and receive input captured by a goniometer to measure wheelchair tilt and recline angles;
an artificial intelligence engine that provides at least one of guidance and control parameters directed to: (1) favorable wheelchair tilt and recline settings; (2) optimal wheelchair tilt and recline angles that may most effectively reduce pressure ulcer risks; (3) optimal duration and frequency to perform wheelchair tilt and recline functions; and (4) measurement of tilt and recline angles by implementing said goniometer;
an artificial intelligence module operable in an Internet cloud, including an artificial neural network (ANN) having a layered network structure, in which processing units (i.e., neurons) are arranged in layers, the neurons in adjacent layers adapted to communicate with each other by sending and receiving signals through weighted connections;
wherein behavior of a neuron in said ANN is defined by its internal activation function, which accumulates the input signals and then calculates the outputs, and
wherein a learning process proceeds in iterations by tuning weights of connections using a training algorithm.

### Claim 17 (depends on 16)

The apparatus of claim 16, wherein actionable aural guidance is provided to achieve recommended tilt and recline settings in substantially real-time as a user adjusts tilt and recline settings on a wheelchair.

### Claim 18 (depends on 17)

The apparatus of claim 17, wherein said goniometer uses aural alerts to enable the wheelchair user to respond to measured tilt and recline angles absent visible input.

### Claim 19 (depends on 16)

The apparatus of claim 16, wherein said goniometer may be configured to periodically remind the wheelchair user to perform wheelchair tilt and recline functions, and to record tilt and recline usage information.

### Claim 20 (depends on 16)

The apparatus of claim 16, wherein said goniometer implements a computational model of the mobile device to enable substantially accurate measurement of wheelchair tilt and recline angles relative to any physical orientation in which said mobile device is positioned.

### Claim 21 (depends on 20)

The apparatus of claim 20, wherein said goniometer can work independently without relying on said artificial intelligence engine.

### Claim 22 (depends on 16)

The apparatus of claim 16, wherein said wheelchair seat orientation can be determined for at least one of a manual wheelchair and a power wheelchair.

### Claim 23

A method for determining spatial orientation of a computational device configured with an accelerometer, comprising:
providing a positioning model of said computational device, said positioning model including a vector ν=αx, αy, αz representing accelerations in three axes measured by said accelerometer;
utilizing the dot product property θ=arccos(ν1·ν2/|ν1|×|ν2|) to calculate angle changes between at least two vectors,
wherein dynamic positioning of said computational device is calculated relative to any reference physical orientation.

### Claim 24 (depends on 23)

The method of claim 23, wherein said computational device is a mobile device configured to measure incline angles.
