U.S. patent application number 14/304758 was filed with the patent office on 2015-04-16 for intelligent apparatus for providing personalized configuration of wheelchair tilt and recline.
The applicant listed for this patent is UNIVERSITY OF CENTRAL OKLAHOMA. Invention is credited to Jicheng FU.
Application Number | 20150106305 14/304758 |
Document ID | / |
Family ID | 52707980 |
Filed Date | 2015-04-16 |
United States Patent
Application |
20150106305 |
Kind Code |
A1 |
FU; Jicheng |
April 16, 2015 |
INTELLIGENT APPARATUS FOR PROVIDING PERSONALIZED CONFIGURATION OF
WHEELCHAIR TILT AND RECLINE
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.
Inventors: |
FU; Jicheng; (Edmond,
OK) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
UNIVERSITY OF CENTRAL OKLAHOMA |
Edmond |
OK |
US |
|
|
Family ID: |
52707980 |
Appl. No.: |
14/304758 |
Filed: |
June 13, 2014 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
61891600 |
Oct 16, 2013 |
|
|
|
Current U.S.
Class: |
706/11 ; 703/2;
706/12; 706/16 |
Current CPC
Class: |
A61G 7/057 20130101;
G06N 3/08 20130101; G16H 50/20 20180101; A61G 5/1056 20130101; A61G
5/1067 20130101; A61G 5/1075 20130101; G06N 3/10 20130101; A61G
2203/10 20130101; G06N 20/00 20190101; G06F 30/20 20200101 |
Class at
Publication: |
706/11 ; 706/12;
706/16; 703/2 |
International
Class: |
G06N 99/00 20060101
G06N099/00; G06F 17/50 20060101 G06F017/50; A61G 5/10 20060101
A61G005/10 |
Goverment Interests
STATEMENT AS TO RIGHTS TO INVENTIONS MADE UNDER FEDERALLY SPONSORED
RESEARCH OR DEVELOPMENT
[0002] This invention was made with Government support of OK-INBRE
(Oklahoma IDeA Network of Biomedical Research Excellence). The
OK-INBRE is a grant from the National Institute of General Medical
Sciences of the National Institutes of Health through Grant Number
8P20GM103447. The Government has certain rights in this invention.
Claims
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.
2. 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.
3. 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.
4. 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.
5. 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.
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.
7. 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.
8. 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.
9. The apparatus of claim 6, further comprising a user interface
configured to display recommended wheelchair tilt and recline
angles, or other position parameters.
10. 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.
11. (canceled)
12. 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.
13. 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.
14. (canceled)
15. 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.
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.
17. 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.
18. 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.
19. 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.
20. 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.
21. The apparatus of claim 20, wherein said goniometer can work
independently without relying on said artificial intelligence
engine.
22. 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.
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 .nu.=.alpha..sub.x, .alpha..sub.y,
.alpha..sub.z representing accelerations in three axes measured by
said accelerometer; utilizing the dot product property
.theta.=arccos(.nu..sub.1.nu..sub.2/|.nu..sub.1|.times.|.nu..sub.2|)
to calculate angle changes between at least two vectors, wherein
dynamic positioning of said computational device is calculated
relative to any reference physical orientation.
24. The method of claim 23, wherein said computational device is a
mobile device configured to measure incline angles.
Description
CROSS REFERENCES TO RELATED APPLICATIONS
[0001] This application claims the benefit of U.S. Provisional
Patent Application No. 61/891,600 filed on Oct. 16, 2013 in the
name of Jicheng Fu, which is expressly incorporated herein by
reference in its entirety.
FIELD OF THE INVENTION
[0003] The present invention relates to providing personalized
configuration of physical supports for the human body. More
particularly, the present invention relates to optimization of
physical supports and human positioning to achieve a desired
outcome, including reducing the risk of pressure ulcers in general,
and reducing the risk of pressure ulcers for people with spinal
cord injury (SCI).
Reference to Computer Program Listing Appendices on CD-R
[0004] This application includes herewith a transmittal under 37
C.F.R .sctn.1.52(e) of a Computer Program Listing Appendix on
Compact Disk (CD), where the transmittal comprises duplicate
compact discs (CDs), totaling two (2) CDs, respectively labeled
"Copy 1" and "Copy 2". Each disk contains the same files. The discs
are IBM-PC machine formatted and MICROSOFT WINDOWS Operating System
compatible, and include identical copies of the following list of
two (2) files, where each file has been saved as a document
viewable using MICROSOFT WORD. All of the materials on the compact
disk, including the computer program listings contained in the
following two (2) files, are incorporated herein by reference in
their entirety. The two files are: APPENDIX A--Listing 1 Mobile
(61,440 bytes) and APPENDIX B--Listing 2 Mobile-Cloud (366,592
bytes). The referenced listings were created on May 6, 2014.
[0005] The Table of Contents for APPENDIX A--Listing 1 Mobile is as
follows:
TABLE-US-00001 Contents ClsTrainer 2 FragmentAngleMeter 5
FragmentForm 10 FragmentFrequency 16 FragmentList 17 FragmentResult
17 InitActivity 19 InputData 20 Main 25 ResultTask 27
TtsIntentService 30
[0006] The Table of Contents for APPENDIX B--Listing 2 Mobile-Cloud
is as follows:
TABLE-US-00002 Contents App Engine Source Code (Cloud) 3 AngleData
3 ApplicationUser.java 3 BloodFlowCore.java 6 BloodFlowResult.java
9 DataManager 10 EMF 10 LinearUnit.java 11 MLP.java 12
NeuralConnection.java 43 Range.java 59 ResultEndpoint.java 61
UserEndpoint.java 62 App Engine Servlets (Cloud) 63
CheckAnglesServlet.java 63 DeleteUserServlet.java 64
SignInServlet.java 65 SignOutServlet.java 66 UpdateUserServlet.java
66 Google App Engine JSP/html Pages (Web) 69 admin.jsp 71 check.jsp
73 duration.jsp 74 editUser.jsp 75 index.html 77 optimal.jsp 79
profile.jsp 81 results.jsp 83 welcome.jsp 84 Android Source Code
(Mobile) 85 CloudEndpointUtils.java 88 Datastore.java 90
FragmentAngleAdjustment.java 94 FragmentAngleMeter 97
FragmentCheck.java 103 FragmentForm.java 104 FragmentFrequency.java
108 FragmentList.java 109 FragmentOptimal.java 109
FragmentResult.java 110 LoginActivity.java 111 MenuActivity.java
112
BACKGROUND OF THE INVENTION
[0007] Negative physiological conditions (e.g., attention deficit,
lower back pain, pressure ulcers) may be experienced by people who
are seated for long periods of time (e.g., long-haul truck drivers,
airline pilots). Pressure ulcers are often experienced by people
having compromised mobility (e.g., the elderly and infirm). People
inflicted with spinal cord injury (SCI) are particularly prone to
developing pressure ulcers. A pressure ulcer is any lesion caused
by unrelieved pressure that results in damage of underlying tissue.
Pressure ulcers may develop following a prolonged period of
compression of the tissue between a bony prominence and a
surface.
[0008] Unrelieved pressure may result in occlusion of capillaries
and lead to ischemia, which has historically been considered a
major factor leading to pressure ulcer formation. The cost of
treating an individual pressure ulcer ranges up to $40,000 but can
exceed $100,000 depending on severity of the wound. Up to 24% of
persons residing in nursing homes reportedly have developed a
pressure ulcer (also called bed sores). When a person is in a
seated position, his or her weight typically rests on a section of
the pelvic girdle called the ischial tuberosity (specifically, the
inferior, posterior portion of the ischium). There are two of these
bony swellings, left and right, sometimes called the sitting bones,
which are located on the posterior, inferior portion of the
ischium. The gluteus maximus muscle lies over it when a person is
standing; however, when he or she sits down, the muscle shifts to a
position that exposes the ischium tuberosity, which then bears the
majority of the weight, and pressure ulcers may occur.
[0009] Pressure ulcers are a frequently occurring healthcare
problem throughout the world. Pressure ulcers pose a significant
threat to the quality of life for people confined to wheelchairs,
such as persons inflicted with spinal cord injury (SCI). Pressure
ulcers once formed may lead to sepsis and early death. Pressure
ulcers remain the most common secondary condition associated with
SCI, and has been reported to occur in from 28% to 85% of patients
with SCI, often within a few days of injury. It is estimated that
more than half of the SCI population will develop at least one
pressure ulcer in their lifetimes. In Europe, treatment of illness
associated with pressure ulcers has been estimated to result in up
to 4% of total healthcare expenditure. The United States alone
spends about $1.4 billion annually on the treatment of pressure
ulcers for people with SCI. It has been estimated that the cost of
treating pressure ulcers is 2.5 times the cost of preventing them.
Considering the numbers and cost of treatment, pressure ulcers are
an important public health problem. There is an urgent and growing
need to develop effective modes of prevention and treatment.
[0010] Although wheelchair tilt and recline functions are typically
used for pressure ulcer prevention, present approaches cannot
determine at what angles wheelchair tilt and recline provide
effective prevention of pressure ulcers. Clinicians typically
recommend uniform guidance to all patients. However, clinical
evidence clearly shows that the SCI individuals demonstrate a wide
variety of requirements. Consequently, universal guidelines cannot
satisfy all the needs. Hence, personalized configuration of
wheelchair tilt and recline for each individual is more desirable
and beneficial.
[0011] It is known that traditional statistical methods can be used
to model biomedical problems. However, statistical methods are
found less capable of finding patterns, dealing with data that may
contain noise, or analyzing non-linear and dependent data.
Artificial intelligent techniques on machine learning, on the other
hand, have played increasingly important role in bioinformatics for
classifying and mining data. Such techniques can capture patterns
based on examples (i.e., training data) even though the underlying
nature, principles, and/or probability distributions may not be
clear. It is, therefore, an object of the present invention to use
an artificial intelligent module (artificial neural network in the
current implementation) to provide a method and apparatus with
which to discover patterns driven by individual human conditions,
operational environments, and outcome or use objectives. It is a
further objective of the present invention to provide personalized
guidance and configuration control for seating support, adjustment,
and positioning including wheelchair tilt and recline usage for
people confined to wheelchairs and who may be inflicted with
SCI.
SUMMARY OF THE INVENTION
[0012] Pressure ulcers (PUs) impose a significant threat to the
quality of life of wheelchair users. Prolonged unrelieved seating
pressure has been identified as a major causative factor of PUs.
Wheelchair tilt and recline (TR) functions are two of the most
desirable features on a wheelchair for relieving seating pressure.
Tilt refers to a change of the seat angle orientation while
maintaining the seat-to-back angle and recline refers to a change
of the seat-to-back angle. Despite the importance of TR functions,
the majority of the wheelchairs do not offer a built-in mechanism
to measure TR angles. Wheelchair users tend to adjust TR angles
based on their own perceptions. However, research shows that
wheelchair users rarely adjust enough tilt or recline angles to
relieve seating pressure, which has been recognized as a major
causative factor of pressure ulcers. The reasons for the
ineffective usage of wheelchair TR functions are in part due to the
lack of a convenient way to measure wheelchair TR angles, and in
part due to the lack of a practical way to monitor whether
wheelchair users follow the clinical guidelines of wheelchair TR
usage.
[0013] The present invention enables a user to obtain a set of
favorable tilt and recline combinations derived from the user's
specific profile that can help reduce the risk of pressure ulcers.
A profile may include information comprising the user's age,
gender, height, weight, body mass index, level of injury,
completeness of injury, duration of injury, age at onset of injury
(e.g., SCI, stroke, amputation), whether he/she smokes, drinks
alcohol, exercises, and/or has pressure ulcer history. An overall
picture of the user's favorable tilt and recline settings are
presentable, along with choices to adjust seating positions. Users
are also presented with the best TR functions that can most
effectively reduce risk of pressure ulcers.
[0014] The present invention provides a unique way to effectively
use wheelchair TR functions, even though most wheelchairs in use
today do not offer a built-in mechanism for measuring TR angles.
Smart mobile devices (smartphones and tablets) are configurable
using the methods of the present invention to accurately measure TR
angles. For example, the advanced computational process of the
present invention enables users to conveniently measure wheelchair
TR angles by simply positioning a smartphone configured in
accordance with the present invention in their pockets (e.g., shirt
pocket). Further, through the combined use of mobile and cloud
computing, the methods of the present invention enable automatic
transmission of wheelchair TR usage information to "cloud-based"
storage and remote analysis.
[0015] Cloud computing is often defined as the practice of using a
network of remote servers hosted on the Internet to store, manage,
and process data, rather than a local server or a personal
computer. Cloud storage involves storing data on multiple virtual
servers that are generally hosted by third parties. The term
"cloud" as used herein generally refers to cloud computing, cloud
storage, and the World Wide Web. Hence, through the use of cloud
computing and web-based approaches, heathcare providers and
researchers can effectively monitor whether the TR guidelines are
properly carried out in wheelchair users' natural settings (e.g.,
home, office, community, etc.). The present invention works for
both power and manual wheelchairs, provided they are equipped with
tilt or tilt and recline functions. Hence, functionality of both
new and exising wheelchairs can be significantly improved through
use of the present invention.
[0016] The present invention may be implemented in multiple
non-limiting versions, including a local device (e.g., smartphone)
version, a mobile-cloud version, and a web-cloud version. In the
local version, all the functionality may be implemented locally in
a smartphone, or similar mobile device. This embodiment may be
preferable for use by individuals with limited data transfer and
bandwidth capacity. However, a fully localized embodiment implies
that the same artificial intelligent module will have to be
implemented multiple times for different mobile operating systems,
such as Android, iOS, and Windows. Alternative embodiments may be
preferable for users where data transfer and bandwidth capacity is
sufficient. The artificial intelligent module and data storage may
be extracted from the local version and implemented as a cloud
computing model in the Internet cloud. In this embodiment, only one
implementation of the intelligent module is needed, and both the
artificial intelligent module and the data storage may be accessed
from a mobile device or a web-based user interface. The smartphone
application as well as the web application are responsible for
collecting user's information, requesting guidance on wheelchair
tilt and recline usage or other position parameters, displaying
results to the users, and using a goniometer implemented on the
local device to measure wheelchair tilt and recline angles.
[0017] In a broad aspect, the apparatus of the present invention
accepts input that may comprise an individual's demographic
information, neurological attributes, physical history, operational
environment, and outcome or use objectives, then returning guidance
and/or control parameters directed to positioning and adjustment of
physical supports for the human body.
[0018] In another aspect, the present invention may be embodied as
a specific purpose mobile device comprising a computational
framework, artificial neural network, a goniometer, and minimum
functionality necessary for configuration and control of
positioning and adjustment directed to seating supports for the
human body.
[0019] In another aspect, the present invention provides functions
and analytical processes capable of finding patterns, dealing with
data that may contain noise, or analyzing non-linear and dependent
data.
[0020] In yet another aspect, the present invention may be embodied
as a specific purpose device integrated into a powered seating
apparatus, where the device comprises a computational framework,
artificial neural network, a goniometer, and minimum functionality
necessary for configuration and control of seating support
configuration.
[0021] In another aspect, the present invention may comprise a
computational framework, artificial neural network, and application
instruction set operable on mainstream general purpose mobile
devices including "smartphones" (e.g. iPhone 5, Samsung Galaxy),
tablets computers (e.g., iPad), Google glass, iWatch, etc.,
collectively "smart devices," running operating systems such as
Android, iOS, and MS-Windows, where such devices include at least
an accelerometer.
[0022] In another aspect, the artificial neural network in the
present invention 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.
[0023] In another broad aspect, the apparatus of the present
invention accepts input comprising an individual's demographic
information, neurological attributes, and pressure ulcer history
and provides guidance or control parameters directed to: (1) the
favorable wheelchair tilt and recline settings; (2) the optimal
wheelchair tilt and recline angles that may most effectively reduce
pressure ulcer risks; and (3) the measurement of tilt and recline
angles by implementing a goniometer.
[0024] In another aspect, the present invention may be configured
to provide optimal duration and frequency to perform wheelchair
tilt and recline functions in response to input comprising an
individual's demographic information, neurological attributes, and
pressure ulcer history.
[0025] In another aspect, the present invention may be configured
to measure wheelchair tilt and recline angles (i.e., a goniometer),
periodically remind the wheelchair user of performing wheelchair
tilt and recline, record wheelchair tilt and recline usage
information, including the time when the wheelchair user performs
the tilt and recline functions, the angles of the tilt and recline,
the duration on which the user maintains the tilt and recline
position, and the derived frequency, i.e., how often the wheelchair
user repositions himself/herself by means of wheelchair tilt and
recline.
[0026] In another aspect, the goniometer can work independently of
the artificial neural network and intelligent module, and operable
on mainstream general purpose mobile devices including
"smartphones" (e.g. iPhone 5, Samsung Galaxy), tablets computers
(e.g., iPad), Google glass, iWatch, etc., collectively "smart
devices," running operating systems such as Android, iOS, and
MS-Windows, where such devices include at least an
accelerometer.
[0027] In another broad aspect, the present invention may be
embodied as a specific purpose mobile device comprising a
computational framework, artificial neural network, a goniometer,
and minimum functionality necessary for configuration and control
of wheelchair tilt and recline.
[0028] In yet another aspect, the present invention may be embodied
as a specific purpose device integrated into a powered wheelchair,
where the device comprises a computational framework, artificial
neural network, a goniometer, and minimum functionality necessary
for configuration and control of wheelchair tilt and recline.
[0029] In another aspect, the present invention may comprise a
computational framework, artificial neural network, and application
instruction set operable on mainstream general purpose mobile
devices including "smartphones" (e.g. iPhone 5, Samsung Galaxy),
tablets computers (e.g., iPad), Google glass, iWatch, etc.,
collectively "smart devices," running operating systems such as
Android, iOS, and Windows, where such devices include an
accelerometer.
[0030] In another broad aspect, the present invention combines
mobile computing and artificial intelligence techniques,
incorporating an artificial intelligence (AI) module in an
application instruction set operable on a mobile device.
[0031] In another aspect, the AI module may be trained with
clinical research data on clinically recommended tilt and recline
angles, and other position parameters.
[0032] In another aspect, smart device users may input into the
user interface of the present invention their demographic,
neurological, and pressure ulcer history information, and
recommended wheelchair tilt and recline angles will be determined
favorable for the individual to reduce risk of pressure ulcers.
[0033] In another aspect of the present invention, the user
interface of the present invention may be configured to display the
recommended wheelchair tilt and recline angles, or other position
parameters.
[0034] In another aspect of the present invention, recommended
wheelchair tilt and recline angle may be output from a mobile
embodiment of the present invention to a control function operating
in a powered wheelchair or other powered mobility device.
[0035] In another aspect of the present invention, recommended
wheelchair tilt and recline angle may be transferred wirelessly to
a controller operational to adjust configuration orientation of a
powered wheelchair or other powered mobility device.
[0036] In a yet another broad aspect of the present invention, an
artificial intelligent module is provided comprising 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.
[0037] In another aspect, the input/output behavior of a neuron is
defined by its internal activation function, which accumulates the
input signals and then calculates the outputs.
[0038] In another aspect, a learning process proceeds in iterations
by tuning the weights of connections using a training algorithm
(e.g., the back-propagation algorithm).
[0039] In another aspect, a user registration component is
provided, which allows users to create their own profiles to record
their demographic information (e.g., gender, weight, height, etc.),
neurological information (e.g., level of injury, completeness of
injury, etc.), and pressure ulcer history (i.e., whether he/she
once developed pressure ulcers).
[0040] In another aspect of the present invention, the output
includes (1) a range of tilt and recline angles that are favorable
for pressure reduction for the user; and (2) the optimal tilt and
recline angles that are most effective in reducing the risk of
pressure ulcers.
[0041] In another aspect, the present invention may be configured
to provide optimal frequency and duration to perform wheelchair
tilt and recline functions, including guidance such as "perform
tilt and recline every 15 minutes (i.e., frequency) and maintain
the tilt and recline setting for at least 3 minutes (i.e.,
duration)."
[0042] In another aspect of the present invention, a goniometer is
provided, which uses an accelerometer sensor in a smart device
(e.g., smartphone or tablet) to measure angles of wheelchair tilt
and recline.
[0043] In another aspect of the present invention, a 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 or
other powered mobility device to be rotated to a precise angular
position.
[0044] In another aspect of the present invention, a goniometer may
be configured to periodically remind the wheelchair user of
performing wheelchair tilt and recline, and record wheelchair tilt
and recline usage information, including the time when the
wheelchair user performs the tilt and recline functions, the angles
of the tilt and recline, the duration on which the user maintains
the tilt and recline position, and the derived frequency, i.e., how
often the wheelchair user repositions himself/herself by means of
wheelchair tilt and recline.
[0045] In another aspect of the present invention, the goniometer
utilizes advanced math and physics methods to establish a model of
the mobile device, which is able to accurately measure wheelchair
TR angles no matter how the user positions the mobile device. As a
result, the wheelchair user can place the smartphone into his/her
pocket while accurately measuring the tilt and recline angles.
[0046] In another aspect of the present invention, a goniometer
uses voice alerts to guide the usage of wheelchair tilt and
recline. As a result, the wheelchair user can place the smartphone
into his/her pocket while measuring the tilt and recline
angles.
[0047] In another aspect of the present invention, a goniometer can
work independently without relying on the artificial neural network
and intelligent module.
[0048] In another aspect of the present invention, the network
structure and weights of the artificial neural network are
determined offline by using clinical research data on clinically
recommended tilt and recline angles, or other position
parameters.
[0049] In another aspect of the present invention, the artificial
neural network is fully configurable through adjusting the network
structure and weights.
[0050] In another aspect of the present invention, the artificial
neural network operable in the AI module can be replaced by other
artificial intelligence techniques, namely, any classification,
clustering, and regression techniques.
[0051] In another broad aspect, the present invention is operable
in a mobile-to-cloud configuration, where the AI module is
implemented in a cloud computing platform, and the use of
cloud-computing ("the cloud") will enable smart devices running on
different operating systems to share the same AI module in the
cloud.
[0052] In another aspect of the present invention, where the AI
module is operable in the cloud, the smart device will be
responsible for at least collecting user's information, requesting
guidance on wheelchair tilt and recline usage or other position
parameters, displaying results to the users, and using an
implemented goniometer to measure wheelchair tilt and recline
angles, balancing workload between mobile and cloud and simplifying
maintenance and upgrade.
[0053] In another aspect of the present invention, where the AI
module is operable in the cloud, the smart device may output
adjustment parameters to a control device operational in a powered
seating apparatus (e.g. powered wheelchair).
[0054] In another broad aspect, the present invention provides
actionable aural guidance to achieve recommended tilt and recline
settings suitable to a particular wheelchair user based on his/her
own profile.
[0055] In another aspect, the present invention enables
measurement, display, and auditory notification of tilt and recline
angles in near real-time as a user adjusts tilt and recline
settings on a wheelchair.
[0056] In another aspect, the present invention provides remote
monitoring and analytics as to whether or not wheelchair users
follow recommended tilt and recline guidance.
[0057] In another broad aspect of the present invention, a
goniometer measures current user positioning angles and contrasts
those angles with clinical guidance to generate control parameters
that cause the seating position of a powered seating apparatus to
be altered to a precise angular position.
[0058] In another broad aspect of the present invention, a method
is provided 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 .nu.=.alpha..sub.x, .alpha..sub.y,
.alpha..sub.z representing accelerations in three axes measured by
said accelerometer; utilizing the dot product property
.theta.=arccos(.nu..sub.1.nu..sub.2/|.nu..sub.1|.times.|.nu..sub.2|)
to calculate angle changes between at least two vectors; where
dynamic positioning of said computational device is calculated
relative to any reference physical orientation.
[0059] In another aspect, the present invention includes the
computational device implemented on a mobile device configured to
measure incline angles which may include tilt and recline
angles.
BRIEF DESCRIPTION OF THE DRAWINGS
[0060] FIG. 1 is a non-limiting diagram showing a process flow
directed to an embodiment of the present invention using a
smartphone implementation (i.e., the local version). Users must
create their profiles before they can use the system. A profile may
include information comprising the user's age, gender, height,
weight, body mass index, level of injury, completeness of injury,
duration of injury, age at onset of SCI, whether he/she smokes,
drinks alcohol, exercises, and/or has pressure ulcer history.
[0061] FIG. 2 is a non-limiting diagram showing a process flow
directed to an embodiment of the present invention using the
mobile-to-cloud implementation. Users must register to create their
profiles before they can login. A profile may include information
comprising the user's age, gender, height, weight, body mass index,
level of injury, completeness of injury, duration of injury, age at
onset of SCI, whether he/she smokes, drinks alcohol, exercises,
and/or has pressure ulcer history.
[0062] FIG. 3 is a non-limiting diagram presenting the function of
"retrieve wheelchair tilt & recline usage". Specifically, users
can obtain a set of favorable incline angles including tilt and
recline combinations that can help reduce the risk of pressure
ulcers.
[0063] FIG. 4a is a non-limiting diagram showing a screen shot of a
smartphone implementation.
[0064] FIG. 4b is a non-limiting diagram showing a screen shot of a
web-based implementation.
[0065] FIG. 4c is a non-limiting diagram showing the structure of a
sample artificial neural network.
[0066] FIG. 5 is a non-limiting diagram showing top-level code
structure for a smart mobile device application (i.e., the local
version).
[0067] FIG. 6 is a non-limiting diagram showing top-level code
structure for web-cloud configuration.
[0068] FIG. 7 is a non-limiting diagram showing top-level code
structure for mobile-to-cloud configuration using the Android
operating system.
[0069] FIG. 8(a) is a non-limiting diagram showing a class diagram
for the Google App Engine (GAE, i.e., cloud) configuration where
the classes are used to compute personalized guidance on wheelchair
tilt and recline usage, and interact with the mobile and web
applications.
[0070] FIG. 8(b) is a non-limiting diagram showing a class diagram
for Google App Engine (cloud) configuration of the present
invention where the classes are used to store the tilt and recline
usage information (the time when the user performs the tilt and
recline functions, the angles of the tilt and recline, etc.)
[0071] FIG. 9 is a non-limiting diagram showing a class diagram for
a mobile configuration using the Android operating system
(complementing FIG. 7).
[0072] FIG. 10a is a non-limiting diagram showing a screen shot of
a smartphone implementation providing a user interface to access
system functions.
[0073] FIG. 10b is a non-limiting diagram showing a screen shot of
a web-based implementation providing a user interface to access
system functions.
[0074] FIG. 11a is a non-limiting diagram showing a screen shot of
a smartphone implementation providing a user interface to enter
demographic attributes.
[0075] FIG. 11b is a non-limiting diagram showing a screen shot of
a web-based implementation providing a user interface to enter
demographic attributes.
[0076] FIG. 12a is a non-limiting diagram showing a screen shot of
a smartphone implementation providing a user interface to display
favorable tilt and recline angles.
[0077] FIG. 12b is a non-limiting diagram showing a screen shot of
a web-based implementation providing a user interface to display
favorable tilt and recline angles.
[0078] FIG. 12c is a non-limiting diagram showing a screen shot of
a smartphone implementation providing a user interface to display
the best tilt and recline angles for the user.
[0079] FIG. 12d is a non-limiting diagram showing a screen shot of
a web-based implementation of the present invention providing a
user interface to display the best tilt and recline angle for the
user.
[0080] FIG. 13 is a non-limiting sequence diagram showing the
process for determining proper adjustment of tilt and recline
settings as determined by the present invention, and aided by
actionable aural guidance provided by the present invention.
[0081] FIG. 14 is a non-limiting diagram showing an exemplary
screenshot for "1: Set the target tilt and recline angles (e.g., 15
tilt/110 recline)" as the first step depicted in FIG. 13.
[0082] FIG. 15 is a non-limiting diagram showing an exemplary
screenshot for "5: Alert the user to stay still for 5 seconds" as
the fifth step depicted in FIG. 13.
[0083] FIG. 16 is a non-limiting diagram showing an exemplary
screenshot of the display on the user interface while the user
adjusts the tilt angle as the seventh step depicted in FIG. 13.
[0084] FIG. 17 is a non-limiting diagram showing an exemplary
screenshot of the display on the user interface while the user
adjusts the recline angle as the tenth step depicted in FIG.
13.
[0085] FIG. 18 is a non-limiting diagram showing an exemplary
screenshot of the display on the user interface while the
smartphone application uses voice alerts to tell the user that the
target recline angle has been reached as the eleventh step depicted
in FIG. 13.
[0086] FIG. 19 is a non-limiting diagram showing the top level
architecture of the mobile-cloud implementation of the present
invention.
DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS OF THE INVENTION
[0087] In brief: FIG. 1 is a non-limiting diagram showing the
process flow directed to use of a smartphone implementation of the
present invention (i.e., the local version). Users must create
their profiles before they can use the system. A profile may
include information comprising the user's age, gender, height,
weight, body mass index, level of injury, completeness of injury,
duration of injury, age at onset of SCI, whether he/she smokes,
drinks alcohol, exercises, and/or has pressure ulcer history. With
a valid profile, the user has the options to update his/her
profile, retrieve recommendations for wheelchair tilt & recline
usage, and use the goniometer implemented in the smartphone to
measure wheelchair tilt/recline angles.
[0088] FIG. 2 is a non-limiting diagram showing the process flow
directed to use of the mobile-to-cloud implementation of the
present invention. Users must register to create their profiles
before they can login. A profile may include information comprising
the user's age, gender, height, weight, body mass index, level of
injury, completeness of injury, duration of injury, age at onset of
SCI, whether he/she smokes, drinks alcohol, exercises, and/or has
pressure ulcer history. In addition, the user needs to choose a
user name and password. If a user can provide a valid user name and
password, he/she can proceed to use the implemented smartphone
application. The user has the options to update his/her profile,
retrieve recommendations for wheelchair tilt & recline usage,
and use the goniometer implemented in the smartphone to measure
wheelchair tilt/recline angles.
[0089] FIG. 3 is a non-limiting diagram presenting the function of
"retrieve wheelchair tilt & recline usage." Specifically, users
can obtain a set of favorable incline angles including tilt and
recline combinations that can help reduce the risk of pressure
ulcers. An overall picture of a user's favorable tilt and recline
settings are presentable, along with choices to adjust seating
positions. Users are also presented with the best tilt and recline
settings that can most effectively reduce risk of pressure ulcers.
Users may select the option "retrieve optimal wheelchair tilt and
recline setting". The option of retrieving the optimal duration and
frequency to perform wheelchair tilt and recline may be selected.
Users may retrieve information directed to how often (i.e.,
frequency) they should perform wheelchair tilt and recline
functions and how long (i.e., duration) each time they should
maintain at that tilt and recline setting.
[0090] FIG. 4a is a non-limiting diagram showing a screen shot of a
smartphone implementation for the mobile-to-cloud version. A user
can choose "submit" if he/she is an existing user. Otherwise, the
user needs to register first.
[0091] FIG. 4b is a non-limiting diagram showing a screen shot of a
web-based implementation. A user can choose "submit" if he/she is
an existing user. Otherwise, the user needs to register first.
[0092] FIG. 4c is a non-limiting diagram showing the structure of
an artificial neural network. It consists of three layers, which
are input layer, hidden layer, and output layer arranged from left
to right.
[0093] FIG. 5 is a non-limiting diagram showing top-level code
structure for a smart device application (i.e., the local version).
The code structure comprises the following modules: InitActivity,
ClsTrainer, Main, InputData, ResultTask, FragmentForm,
FragmentFrequency, FragmentAngleMeter, FragmentList, and
FragmentResult.
[0094] FIG. 6 is a non-limiting diagram showing top-level data flow
for a web-based configuration. The code structure for the web-based
configuration comprises the following modules: Index Page
(index.html), Register (SignInServlet), Sign in (SignInServlet),
User Welcome Page (welcome.jsp), Profile Page (profile.jsp), Update
Profile (UpdateUserServlet), Check Angle Page (check.jsp), Check
Angles (CheckAnglesServlet), Range of Angles Page (result.jsp),
Optimal Angle Page (optimal.jsp), Duration and Frequency Page
(duration.jsp), Admin User List Page (admin.jsp), Delete User
(DeleteUserServlet), Edit User Page (edituser.jsp), Edit User
(UpdateUserServlet), and Create New User (UpdateUserServlet).
[0095] FIG. 7 is a non-limiting diagram showing top-level code
structure for mobile-to-cloud configuration using the Android
operating system. The code structure includes Register, Sign In,
Main Menu Screen (MenuActivity), Profile Screen (FragmentForm),
Check Angle Page (FragmentCheck), Range of Angles Page
(FragmentResult), Optimal Angles Page (FragmentOptimal), Duration
and Frequency Page (FragmentFrequency), and Goniometer
(FragmentAngleAdjustment).
[0096] FIG. 8(a) is a non-limiting diagram showing a class diagram
for a GAE (cloud) configuration where the classes are used to
compute personalized guidance on wheelchair tilt and recline usage,
and interact with the mobile and web applications. The code
structure includes: ApplicationUser, BloodFlowCore,
BloodFlowResult, Range, UserEndpoint, CheckAnglesServlet,
SignInServlet, ResultEndpoint, UpdateUserServlet,
DeleteUserServlet, SignOutServlet, MLP, LinearUnit, NeuralEnd, and
NeuralConnection.
[0097] FIG. 8(b) is a non-limiting diagram showing a class diagram
for the GAE (cloud) configuration of the present invention where
the classes are used to store the tilt and recline usage
information (the time when the user performs the tilt and recline
functions, the angles of the tilt and recline, etc.) The code
structure includes: AngleData, DataManager, and EMF.
[0098] FIG. 9 is a non-limiting diagram showing a class diagram for
a mobile configuration using the Android operating system
(complementing FIG. 7). The code structure includes: LoginActivity,
MenuActivity, FragmentForm, FragmentCheck, FragmentResult,
FragmentOptimal, FragmentFrequency, FragmentAngleAdjustment,
FragmentAngleMeter, Datastore, UserEndpoint, ResultEndpoint,
BloodFlowCore, LoginActivity, MenuActivity, FragmentForm,
FragmentCheck, FragmentResult, FragmentOptimal, FragmentFrequency,
Datastore, UserEndpoint, ResultEndpoint, and BloodFlowCore.
[0099] FIG. 10a is a non-limiting diagram showing a screen shot of
a smartphone implementation providing a user interface to access
system functions. System responses are anticipated to at least user
touch and voice commands. Audio recitation and response for
visually impaired individuals is anticipated.
[0100] FIG. 10b is a non-limiting diagram showing a screen shot of
a web-based implementation providing a user interface to access
system functions. System responses are anticipated to at least user
touch and voice commands. Audio recitation and response for
visually impaired individuals is anticipated.
[0101] FIG. 11a is a non-limiting diagram showing a screen shot of
a smartphone implementation providing a user interface to enter
demographic attributes. System responses are anticipated to at
least user touch and voice commands. Audio recitation and response
for visually impaired individuals is anticipated.
[0102] FIG. 11b is a non-limiting diagram showing a screen shot of
a web-based implementation providing a user interface to enter
demographic attributes. System responses are anticipated to at
least user touch and voice commands. Audio recitation and response
for visually impaired individuals is anticipated.
[0103] FIG. 12a is a non-limiting diagram showing a screen shot of
a smartphone implementation providing a user interface to display
favorable tilt and recline angles. System responses are anticipated
to at least user touch and voice commands. Audio recitation and
response for visually impaired individuals is anticipated.
[0104] FIG. 12b is a non-limiting diagram showing a screen shot of
a web-based implementation providing a user interface to display
favorable tilt and recline angles. System responses are anticipated
to at least user touch and voice commands. Audio recitation and
response for visually impaired individuals is anticipated.
[0105] FIG. 12c is a non-limiting diagram showing a screen shot of
a smartphone implementation providing a user interface to display
the best tilt and recline angle for a user. System responses are
anticipated to at least user touch and voice commands. Audio
recitation and response for visually impaired individuals is
anticipated.
[0106] FIG. 12d is a non-limiting diagram showing a screen shot of
a web-based implementation providing a user interface to display
the best tilt and recline angle for a user. System responses are
anticipated to at least user touch and voice commands. Audio
recitation and response for visually impaired individuals is
anticipated.
[0107] FIG. 13 is a non-limiting diagram showing the process for
determining proper adjustment of tilt and recline settings as
determined by the present invention. Measurement, display, and
auditory notification of tilt and recline angles are accomplished
in substantially real-time as a user adjusts tilt and recline
settings on a wheelchair. Actionable aural guidance is provided to
enable the user to achieve recommended tilt and recline settings
suitable to the particular wheelchair user based on his or her
specific profile.
[0108] FIG. 14 is a non-limiting diagram showing an exemplary
screenshot of the user interface implemented as an element in the
process for determining proper adjustment of tilt and recline
settings as determined by the present invention. A screenshot for
"1: Set the target tilt and recline angles (e.g., 15 tilt/110
recline)" is shown as the first step depicted in FIG. 13.
[0109] FIG. 15 is a non-limiting diagram showing an exemplary
screenshot of the user interface implemented as an element in the
process for determining proper adjustment of tilt and recline
settings as determined by the present invention. Actionable aural
guidance is provided to ask the user to stay still for a period of
time (e.g., 5 seconds) so that the smart device application can
accurately measure the initial upright position. An exemplary
screenshot for "5: Alert the user to stay still for 5 seconds" is
shown as the fifth step depicted in FIG. 13.
[0110] FIG. 16 is a non-limiting diagram showing an exemplary
screenshot of the user interface implemented as an element in the
process for determining proper adjustment of tilt and recline
settings as determined by the present invention. A screenshot of
the display on the user interface while the user adjusts the tilt
angle is shown as the seventh step depicted in FIG. 13.
[0111] FIG. 17 is a non-limiting diagram showing an exemplary
screenshot of the user interface implemented as an element in the
process for determining proper adjustment of tilt and recline
settings as determined by the present invention. A screenshot of
the display on the user interface while the user adjusts the
recline angle, is shown as the tenth step depicted in FIG. 13.
[0112] FIG. 18 is a non-limiting diagram showing an exemplary
screenshot of the user interface implemented as an element in the
process for determining proper adjustment of tilt and recline
settings as determined by the present invention. A screenshot of
the display on the user interface is shown as the eleventh step
depicted in FIG. 13. Actionable aural guidance may be provided
concomitantly to tell the user that the target recline angle has
been reached.
[0113] FIG. 19 is a non-limiting diagram showing the top level
architecture of the mobile-cloud implementation of the present
invention. An artificial neural network is shown implemented in the
cloud, along with data processing and analysis. Researchers and
healthcare providers are able to remotely access patient data
through a secure and controlled interface.
[0114] In detail: Referring now to FIG. 1, a non-limiting schematic
illustration of one embodiment of the present invention 10 shows
one configuration of the process flow for a typical smartphone
implementation of the present invention (i.e., the local version)
10. Users must create their profiles 11 before they can use the
system (see FIG. 11a). A profile may include information comprising
the user's age, gender, height, weight, body mass index, level of
injury, completeness of injury, duration of injury, age at onset of
SCI, whether he/she smokes, drinks alcohol, exercises, and/or has
pressure ulcer history. The profile is stored locally in the
smartphone. Then, the user can proceed to use the implemented
smartphone application. The user has the options to update his/her
profile 14, retrieve recommendations for wheelchair tilt &
recline usage 15, and use the goniometer implemented in the
smartphone to measure wheelchair tilt/recline angles 16. Note that
the artificial neural networks (ANN) are implemented locally in the
smartphone. The ANNs can provide the set of favorable tilt and
recline settings and the best tilt and recline setting for
individual users based on their profiles.
[0115] Referring now to FIG. 2, a non-limiting schematic
illustration of one embodiment of the present invention 10 shows
one configuration of the process flow for the mobile-to-cloud
implementation of the present invention 10 (also see FIG. 19).
Users must register 21 to create their profiles before they can
login 22. The smartphone application provides the user interface
that allows the users to register (i.e., create their own profiles.
See FIG. 11a). A profile may include information comprising the
user's age, gender, height, weight, body mass index, level of
injury, completeness of injury, duration of injury, age at onset of
SCI, whether he/she smokes, drinks alcohol, exercises, and/or has
pressure ulcer history. In addition, the user needs to choose a
user name and password. The profile is then stored in the cloud
(see FIG. 19), i.e., the smartphone's communication capability (3G,
4G or WIFI) is used to transmit data to the cloud. If a user can
provide a valid user name and password 23, he/she can proceed to
use the implemented application operable and running on the
smartphone. The user has the options to update his/her profile 24,
retrieve recommendations for wheelchair tilt & recline usage
25, and use the goniometer implemented in the smartphone to measure
wheelchair tilt/recline angles 26. Note that under the
mobile-to-cloud configuration, the artificial neural network (ANN)
is implemented in the cloud. The ANN can provide the set of
favorable tilt and recline settings and the best tilt and recline
setting for individual users based on a user's profile. Under the
mobile-to-cloud configuration of the present invention 10, a single
artificial intelligent module may be configured to serve a
plurality of mobile users, who may use various mobile operating
systems, such as iOS, Android, Windows, etc. In the mobile side,
the users need to login 22 to the application by providing their
user names/passwords. Then, the users may use the application in
the same way as a user operating the local version of the present
invention 10. The difference is that the information on the
guidance of wheelchair tilt and recline usage is retrieved from the
cloud. This difference is made transparent to the users. Hence, the
users do not have to take care of the complex technical details
directed to data storage and computation.
[0116] Referring now to FIG. 3, a non-limiting diagram is shown
presenting the function of "retrieve wheelchair tilt & recline
usage" 30 of the present invention 10 (see also 15 in FIGS. 1 and
25 in FIG. 2). Specifically, users can obtain a set of favorable
incline angles including tilt and recline combinations 31 that can
help reduce the risk of pressure ulcers. An overall picture of a
user's favorable tilt and recline settings are presentable, along
with choices to adjust seating positions. Users are also presented
with the best tilt and recline settings 32 that can most
effectively reduce risk of pressure ulcers. Users may select the
option "retrieve optimal wheelchair tilt and recline setting". A
third option is retrieving the optimal duration and frequency to
perform wheelchair tilt and recline. Users may elect to retrieve
information directed to how often (i.e., frequency) they should
perform wheelchair tilt and recline functions and how long (i.e.,
duration) each time they should maintain at that tilt and recline
setting 33. For example, guidance may be provided such as "perform
tilt and recline every 15 minutes (i.e., frequency) and maintain
the tilt and recline setting for at least 3 minutes (i.e.,
duration).
[0117] The preferable output includes (1) a range of tilt and
recline angles that are favorable for pressure reduction for the
user 31; (2) the optimal tilt and recline angles that are most
effective in reducing the risk of pressure ulcers 32; and (3) the
optimal frequency and duration to perform wheelchair tilt and
recline functions 33.
[0118] Referring now to FIG. 4a, a non-limiting diagram is shown
presenting a screen shot of a user interface 40 in a smartphone
implementation of the present invention 10 (i.e., the
mobile-to-cloud version). A user can choose "submit" 41 if he/she
is an existing user. Otherwise, the user needs to register 42 (see
FIG. 11a) before proceeding. All users' profiles are stored in the
cloud (see FIG. 19). For an existing user, the smartphone
application sends his/her user name and password (collected in FIG.
4a) to the cloud application of the present invention to verify the
user's identity. Only valid users can use or gain access to the
system. System responses are anticipated to at least user touch and
voice commands.
[0119] Referring now to FIG. 4b, a non-limiting diagram is shown
presenting a screen shot 43 of a user interface in a web
implementation of the present invention 10. A user can choose "sign
in" 44 if he/she is an existing user. Otherwise, the user needs to
register 45 (see FIG. 11b) before proceeding. Under the web
implementation, all users' profiles are stored in the cloud. For an
existing user, the web application sends his/her user name and
password (collected in FIG. 4b) to the cloud application of the
present invention to verify the user's identity. Only valid users
can use or gain access to the system. System responses are
anticipated to at least user touch and voice commands.
[0120] Referring now to FIG. 4c, in a preferred implementation, the
best known artificial neural network (ANN) is implemented for the
present invention 10. ANN has a layered network structure 400, in
which the processing units (i.e., neurons) are arranged in layers.
The ANN in FIG. 4c consists of three layers, including the input
layer 401, the hidden layer 402, and the output layer 403. Neurons
in adjacent layers can communicate with each other by sending and
receiving signals through the weighted connections. The
input/output behavior of a neuron is defined by its internal
activation function, which accumulates the input signals and then
calculates the outputs. Once the network structure 400 is
determined, the learning process proceeds in iterations by tuning
the weights of connections using a training algorithm, such as the
well-known back-propagation algorithm.
[0121] The network structure and weights of the ANN in the
application are determined offline by using clinical research data
on clinically recommended tilt and recline angles. Specifically,
wheelchair users with spinal cord injury were recruited to
participate in the research. A testing condition includes a
five-minute sitting-induced ischemic period, i.e., the research
participant sits in the upright position with no tilt or recline
for 5 minutes, and a five-minute pressure relief period, i.e., the
research participant sits in a clinically recommended tilt and
recline setting for 5 minutes. The skin blood flow was measured
throughout the test so that we can know whether a tilt and recline
setting is favorable for increasing skin blood flow, which has been
widely used to determine the efficacy of wheelchair seating
conditions. Then, the skin blood flow data was used to train the
ANN to predict tilt and recline settings for individual wheelchair
users. Other position parameters may be incorporated as well, such
as the elevating leg-rest function of a power wheelchair. The ANN
in the invention is fully configurable through adjusting the
network structure 400 and weights. The ANN can be replaced by other
artificial intelligence techniques, namely, any classification,
clustering, and regression techniques, such as support vector
machine (SVM), C4.5 decision tree, random forest, etc. The present
invention will support such transparency in changing the AI
module.
[0122] Referring now to FIG. 5, a non-limiting diagram is shown
presenting a top-level code structure 50 for a smart device
application of the present invention 10 (i.e., the local version).
The code structure 50 comprises the following modules: InitActivity
51, ClsTrainer 52, Main 50A, InputData 53, ResultTask 54,
FragmentForm 55, FragmentFrequency 56, FragmentAngleMeter 57,
IntentService.java 571, FragmentList 58, and FragmentResult 59.
[0123] InitActivity.java: This class 51 shows the welcome screen
when the application is loading. It calls ClsTrainner 52 to train
the classifiers in the backend. Once it finishes initializing
classifiers, this activity class will transfer to the Main 50A
activity class.
[0124] ClsTrainner.java: This class 52 is used to initialize a
classifier and regression learner coded in the present invention.
The classifier can classify whether a given tilt and recline
setting is favorable for an individual with spinal cord injury
(SCI) to reduce the risk of pressure ulcer. The regression learner
can predict the extent of risk deduction for a given tilt and
recline setting. This class runs in the backend as a thread when
the application starts.
[0125] Main.java: The Main class 50A is the container for all the
fragment classes in this application. It provides the overall
layout of the application.
[0126] FragmentForm.java: This class 55 is used to provide the user
interface to input data 53. Users can update their profiles (FIG.
1, 14, FIG. 2, 24) here. It can also call the classifier and
regression modules to make new predictions with updated
profiles.
[0127] FragmentFrequency.java: This class 56 shows to the users the
optimal duration and frequency to perform the wheelchair tilt and
recline functions. It invokes the daemon thread that is running in
the backend to return the optimal duration and frequency to the
user interface (UI) thread.
[0128] FragmentList.java: This class 58 provides a list of
functions that is offered by the smartphone app. It redirects a
user to the appropriate functions based on the user's choice.
[0129] FragmentResult.java: This class 59 includes the template of
My Range, My Optimal, and My Test screens (shown on FIG. 10a) in
the application. It shows the up-to-date prediction results
obtained from the back-end thread.
[0130] InputData.java: This is a singleton class 53 that it has
only a single instance in the memory. It contains all the data in
this application. It acts as a data store in this application. The
trained functions (classifier and regression) as well as user
inputs are all stored in this class.
[0131] ResultTask.java: The ResultTask class 54 is running in the
backend as a daemon thread. Its functionality is to make
predictions based on a user's profile (FIG. 1, 14, FIG. 2, 24).
This class also answers other requests, such as whether a
particular tilt and recline setting is favorable for the user, and
returns the result to the UI thread.
[0132] FragmentAngleMeter.java: This class 57 provides the
goniometer function. It reads the accelerometer sensor in the
smartphone and calculates the current angle of the phone
orientation for the user. This class provides a novel algorithm to
measure wheelchair tilt and recline (TR) angles by using the
accelerometer in a smartphone. Specifically, the position of a
smartphone is modeled with a vector .nu.=.alpha..sub.x,
.alpha..sub.y, .alpha..sub.z, which represents accelerations in
three axes measured by the accelerometer. When the tilt or recline
stabilizes to a new angle, accelerations in three axes will change
due to the decomposition of the gravity along the new angle of the
phone. Then, we utilize the dot product property to calculate angle
changes between two vectors (positions):
.nu..sub.1.nu..sub.2=|.nu..sub.1|.times.|.nu..sub.2|.times.cos
.theta. (1)
Or equivalently,
.theta.=arccos(.nu..sub.1.nu..sub.2/|.nu..sub.1|.times.|.nu..sub.2|)
(2)
[0133] Hence, no matter how the smartphone is positioned, the TR
angle .theta. between two vectors can be measured. In addition,
this class employs the novel text-to-speech technique (see class
IntentService.java), which enables the system to use voice alerts
to guide wheelchair users for proper TR usage.
[0134] IntentService.java: This class 571 implements the Android
text-to-speech listener and initializes the text-to-speech function
for the subsequent usage.
[0135] Referring now to FIG. 6, a non-limiting diagram is shown
presenting a top-level data flow for a Web based configuration 60
of the present invention 10. The code structure for a Web based
configuration 60 comprises the following modules: Index Page 61
(index.html), Register 611 (SignInServlet), Sign in 612
(SignInServlet), User Welcome Page 62 (welcome.jsp), Profile Page
621 (profile.jsp), Update Profile 6211 (UpdateUserServlet), Check
Angle Page 622 (check.jsp), Check Angles 6221 (CheckAnglesServlet),
Range of Angles Page 623 (result.jsp), Optimal Angle Page 624
(optimal.jsp), Duration and Frequency Page 625 (duration.jsp),
Admin User List Page 63 (admin.jsp), Delete User 631
(DeleteUserServlet), Edit User Page 632 (edituser.jsp), Edit User
64 (UpdateUserServlet), and Create New User 65
(UpdateUserServlet).
[0136] Index Page 61 (index.html): Index page 61 is the first web
page that a user can access. It provides options for registered
users to sign in and for unregistered users to register.
[0137] Register 611 (SignInServlet): It is a Java Servlet that is
invoked by index.html and allows unregistered users to register and
create their own user names and passwords. A Java servlet is a
class that is used to extend the functionality of the cloud.
[0138] Sign in 612 (SignInServlet): It is a Java servlet used by
index.html when to sign in and register users given a username and
password.
[0139] User Welcome Page 62 (welcome.jsp): It is the welcome page
after a user successfully signs in the system.
[0140] Profile Page 621 (profile.jsp): This page allows users to
create their own profiles including their demographic attributes,
neurological information, and pressure ulcer history, etc.
[0141] Update Profile 6211 (UpdateUserServlet): It is a servlet
class that is invoked by profile.jsp to update the user's
profile.
[0142] Check Angle Page 622 (check.jsp): This page gives a user the
option to check whether a particular wheelchair tilt and recline
setting will be favorable for the individual user to reduce
pressure ulcer's risk.
[0143] Check Angles 6221 (CheckAnglesServlet): It is a servlet
class that is invoked by check.jsp to check whether a particular
wheelchair tilt and recline setting will be favorable for the
individual user to reduce pressure ulcer risk.
[0144] Range of Angles Page 623 (result.jsp): This page shows the
range of tilt and recline angles that are favorable for reducing
pressure ulcers' risk.
[0145] Optimal Angle Page 624 (optimal.jsp): This page shows the
optimal wheelchair tilt and recline settings that may most
effectively reduce risk of pressure ulcers.
[0146] Duration and Frequency Page 625 (duration.jsp): This page
illustrates the optimal duration and frequency to perform
wheelchair tilt and recline functions. For example, the user should
perform wheelchair tilt and recline functions every 15 minutes
(i.e., frequency) and each time the user should maintain that
setting for 3 minutes (i.e., duration).
[0147] Admin User List Page 63 (admin.jsp): This is a page designed
for administrators, who will maintain users, including "add",
"edit", and "delete" users.
[0148] Delete User 631 (DeleteUserServlet): It is a Java servlet
used by admin.jsp when an administrator attempts to delete an
application user.
[0149] Edit User Page 632 (edituser.jsp): This is a web page that
invokes Servlets to add a new user or update an existing user.
[0150] Edit User 64 (UpdateUserServlet): It is a Java servlet used
by admin.jsp when an administrator attempts to edit a user's
information.
[0151] Create New User 65 (UpdateUserServlet): The same
UpdateUserServlet can also be used to create a new user.
[0152] Referring now to FIG. 7, a non-limiting diagram is shown
presenting a top-level control flow of the present invention 10 for
mobile-to-cloud configuration using the Android operating system.
The code structure 70 includes: Login Screen 71, Register 72,
Datastore 721, Sign In 73, User Menu Screen 701 (MenuActivity),
Profile Screen 74 (FragmentForm), Check Angle Page 75
(FragmentCheck), Range of Angles Page 76 (FragmentResult), Optimal
Angles Page 77 (FragmentOptimal), Duration and Frequency Page 78
(FragmentFrequency), and Goniometer 79
(FragmentAngleAdjustment).
[0153] Login Screen 71 (LoginActivity): It is the starting Android
activity that calls register and signin methods and redirects user
to the MenuActivity 701 if the user name and password are verified
successfully. Activity is an Android term that represents a
function that a user can perform.
[0154] Register 72: It invokes the Datastore class
(Datastore.register function) that interacts with the Google App
Engine datastore to store new user's information (see FIG. 19).
[0155] Datastore 721: This class interacts with the Google App
Engine datastore service and is used by both the mobile endpoints
and java servlets.
[0156] Sign In 73: It invokes the Datastore class (Datastore.signin
function) that interacts with the Google App Engine datastore to
validate the user's information (see FIG. 19).
[0157] User Menu Screen 701 (MenuActivity): It is the main activity
that shows the main menu of the system. It consists of the
currently selected fragment and a navigation list for changing
fragments. A fragment is an Android term that represents a portion
of the user interface.
[0158] Profile Screen 74 (FragmentForm): It is a fragment that
consists of the input fields for user information. Once the button
at the bottom of the fragment is pressed, the given information is
then updated 741 to the datastore in the cloud (see FIG. 19).
[0159] Check Angle Page 75 (FragmentCheck): It is a fragment that
determines if the given tilt and recline angles 751 are in the
ranges provided by the artificial neural network (see FIG. 19).
[0160] Range of Angles Page 76 (FragmentResult): It is a fragment
that displays a list of ranges provided by the artificial neural
network (see FIG. 19). These ranges are favorable tilt and recline
combinations that can help reduce the risk of pressure ulcers.
[0161] Optimal Angles Page 77 (FragmentOptimal): It is a fragment
that displays the optimal angles of wheelchair tilt and recline
provided by the artificial neural network (see FIG. 19).
[0162] Duration and Frequency Page 78 (FragmentFrequency): It is a
fragment used to check the duration and frequency that the user
should perform wheelchair tilt and recline functions. For example,
the user should perform wheelchair tilt and recline functions in
every 15 minutes (i.e., frequency) and each time the user should
maintain that position for 3 minutes (i.e., duration).
[0163] Goniometer 79 (FragmentAngleAdjustment): It is a fragment
used to display the current angle of the phone. It reads the
accelerometer sensor in the smartphone and calculates the current
angle of the phone orientation for the user. A desired angle can be
set by using the device's menu button. The background of this
fragment will turn greener the closer the current angle is to the
desired angle.
[0164] Referring now to FIG. 8a, a non-limiting diagram is shown
presenting a class diagram 80 for GAE (cloud) configuration of the
present invention 10 where the classes are used to compute
personalized guidance on wheelchair tilt and recline, and interact
with the mobile and web applications. The code structure includes:
ApplicationUser 81, BloodFlowCore 82, BloodFlowResult 83, Range 84,
UserEndpoint 85, CheckAnglesServlet 86, SignInServlet 87,
ResultEndpoint 88, UpdateUserServlet 89, DeleteUserServlet 810,
SignOutServlet 811, MLP 812, LinearUnit 816, NeuralEnd 817, and
NeuralConnection 818.
[0165] ApplicationUser 81: consists of all user fields and
represents the entity structure stored in the Google App Engine
(GAE) datastore.
[0166] BloodFlowCore 82: contains methods for interacting with the
WEKA API, which is an open source data mining platform and
returning the BloodFlowResult object. This is where the artificial
neural network is built and angles are returned.
[0167] BloodFlowResult 83: contains all output results needed and
eventually displayed to the user, including a list of tilt and
recline ranges, the optimal angles, and duration and frequency.
[0168] Range 84: is a class used to hold one set of tilt and
recline ranges.
[0169] UserEndpoint 85: this Endpoint class manipulates
ApplicationUser entities in the datastore by calling the Datastore
class methods. Endpoint classes are located in the GAE source code
and are annotated to be generated into an API to be used with
Android.
[0170] CheckAnglesServlet 86: is a servlet class that checks
whether a particular wheelchair tilt and recline setting will be
favorable for the individual user to reduce pressure ulcer's
risk.
[0171] SignInServlet 87: is a Java servlet used when to sign in and
register users given a username and password.
[0172] ResultEndpoint 88: this endpoint creates a BloodFlowResult
object to store results from the runBloodFlowCore method. Endpoint
classes are located in the GAE source code and are annotated to be
generated into an API to be used with Android.
[0173] UpdateUserServlet 89: is a Java servlet used when an
administrator attempts to edit a user's information.
[0174] DeleteUserServlet 810: is a Java servlet used when an
administrator attempts to delete an application user.
[0175] SignOutServlet 811: This class provides the sign out
function in the web application.
[0176] MLP.java 812: The MLP class is customized by adding
getNumWeights( ), importWeights( ), and exportWeights( ) methods.
These methods allow us to reconstruct ANN if the network structure
and weights are provided.
[0177] MLP 812, LinearUnit 816, NeuralEnd 817, and NeuralConnection
818 are obtained from WEKA, which is an open source platform for
data mining. These classes are used to model the artificial neural
network. LinearUnit 816, NeuralEnd 817, and NeuralConnection 818
are used without any customizations.
[0178] Referring now to FIG. 8b, a non-limiting diagram is shown
presenting a class diagram 80 for GAE (cloud) configuration of the
present invention 10 where the classes are used to store the tilt
and recline usage information (the time when the user performs the
tilt and recline functions, the angles of the tilt and recline,
etc.) The code structure includes: AngleData 813, DataManager 814,
and EMF 815.
[0179] AngleData 813: is the data type class that models tilt and
recline angle data, which is sent from the mobile client.
[0180] DataManager 814: is the class that handles the communication
between the client and Google datastore.
[0181] EMF 815: EntityManagerFactory helps communication between
the Google datastore and the application.
[0182] Referring now to FIG. 9, a non-limiting diagram is shown
presenting a class diagram 90 for a mobile configuration of the
present invention 10 using the Android operating system
(complementing FIG. 7). The code structure includes: LoginActivity
91, MenuActivity 92, FragmentForm 921, FragmentCheck 922,
FragmentResult 923, FragmentOptimal 924, FragmentFrequency 925,
FragmentAngleAdjustment 926, FragmentList 927, Datastore 93,
UserEndpoint 94, ResultEndpoint 95, and BloodFlowCore 96.
[0183] LoginActivity 91: it is the starting Android activity that
calls register and signin methods and redirects user to the
MenuActivity 92 if the user name and password are verified
successfully. Activity is an Android term that represents a
function that a user can perform.
[0184] MenuActivity 92: it is the main activity that shows the main
menu of the system. It consists of the currently selected fragment
and a navigation list for changing fragments. A fragment is an
Android term that represents a portion of the user interface.
[0185] FragmentForm 921: It is a fragment that consists of the
input fields for user information. Once the button at the bottom of
the fragment is pressed, the given information is then updated to
the datastore in the cloud. A fragment is an Android term that
represents a portion of the user interface.
[0186] FragmentCheck 922: It is a fragment that determines if the
given tilt and recline angles are in the ranges provided by the
artificial neural network.
[0187] FragmentResult 923: It is a fragment that displays a list of
ranges provided by the artificial neural network. These ranges are
favorable tilt and recline combinations that can help reduce the
risk of pressure ulcers.
[0188] FragmentOptimal 924: It is a fragment that displays the
optimal angles of wheelchair tilt and recline provided by the
artificial neural network.
[0189] FragmentFrequency 925: It is a fragment used to check the
duration and frequency that the user should perform wheelchair tilt
and recline functions. For example, the user should perform
wheelchair tilt and recline functions every 15 minutes (i.e.,
frequency) and each time the user should maintain that setting for
3 minutes (i.e., duration).
[0190] FragmentAngleAdjustment 926: It is a fragment used to
display the current angle of the wheelchair (tilt or recline). It
reads the accelerometer sensor in the smartphone and calculates the
current angle of the phone orientation for the user. A desired
angle can be set by using the device's menu button. The background
of this fragment will turn greener the closer the current angle is
to the desired angle.
[0191] FragmentList 927: is a fragment that provides a list of
functions that is offered by the smartphone app. It redirects a
user to the appropriate functions based on the user's choice.
[0192] Datastore 93: this class is used by the mobile endpoints to
interact with the Google App Engine datastore to manipulate
data.
[0193] UserEndpoint 94: this Endpoint class manipulates
ApplicationUser entities in the datastore by calling the Datastore
class methods. Endpoint classes are located in the GAE source code
and are annotated to be generated into an API to be used with
Android.
[0194] ResultEndpoint 95: this endpoint creates a BloodFlowResult
object to store results from the runBloodFlowCore method. Endpoint
classes are located in the GAE source code and are annotated to be
generated into an API to be used with Android.
[0195] BloodFlowCore 96: contains methods for interacting with the
WEKA API, which is an open source data mining platform and
returning the BloodFlowResult object. This is where the artificial
neural network is built and angles are returned.
[0196] FIG. 10a is a non-limiting diagram showing a screen shot of
a smartphone implementation of the present invention 10 providing a
user interface 101 to access system functions. Both the local
mobile version and the mobile-to-cloud version may have the same
interface as shown in FIG. 10a. System responses are anticipated
and implemented to include at least user touch and voice commands.
Audio recitation and response for visually impaired individuals may
be provided by the present invention 10. User touch, voice
activation and audio recitation functions are generally
programmable and operable on industry standard smart devices, such
as various device models of iPhone, iPad, Samsung Galaxy, and HP
tablets, running operating systems such as Android, iOS, and
Windows, where such devices include an accelerometer.
Implementation on any such mobile device having the minimum
function set as described herein is anticipated.
[0197] FIG. 10b a non-limiting diagram showing a screen shot of a
web-based implementation of the present invention 10 providing a
user interface 102 to access system functions. System responses are
anticipated and provided in the present invention 10 to include at
least user touch and voice commands. Audio recitation and response
for visually impaired individuals may also be provided by the
present invention 10.
[0198] FIG. 11a is a non-limiting diagram showing a screen shot of
a smartphone implementation of the present invention 10 providing a
user interface 110 to enter demographic attributes. System
responses are anticipated and provided in the present invention 10
to include at least user touch and voice commands. Audio recitation
and response for visually impaired individuals is anticipated and
provided by the present invention 10. User touch, voice activation
and audio recitation functions are generally programmable and
operable on industry standard smart devices, such as various device
models of iPhone, iPad, Samsung Galaxy, and HP tablets, running
operating systems such as Android, iOS, and Windows, where such
devices include an accelerometer. Any such device having the
minimum function set as described herein is anticipated.
[0199] FIG. 11b is a non-limiting diagram showing a screen shot of
a web-based implementation of the present invention 10 providing a
user interface 112 to enter demographic attributes. System
responses are anticipated and provided in the present invention 10
to include at least user touch and voice commands. Audio recitation
and response for visually impaired individuals is anticipated and
provided by the present invention 10.
[0200] FIG. 12a is a non-limiting diagram showing a screen shot of
a smartphone implementation of the present invention 10 providing a
user interface 120 to display favorable tilt and recline angles.
System responses are anticipated and provided in the present
invention 10 to include at least user touch and voice commands.
Audio recitation and response for visually impaired individuals is
anticipated and provided by the present invention 10. User touch,
voice activation and audio recitation functions are generally
programmable and operable on industry standard smart devices, such
as various device models of iPhone, iPad, Samsung Galaxy, and HP
tablets, running operating systems such as Android, iOS, and
Windows, where such devices include an accelerometer.
Implementation on any such device having the minimum function set
as described herein is anticipated.
[0201] FIG. 12b is a non-limiting diagram showing a screen shot of
a web-based implementation of the present invention 10 providing a
user interface 122 to display favorable tilt and recline angles.
System responses are anticipated in the present invention 10 to
include at least user touch and voice commands. Audio recitation
and response for visually impaired individuals is anticipated and
provided by the present invention 10.
[0202] FIG. 12c is a non-limiting diagram showing a screen shot of
a smartphone implementation of the present invention 10 providing a
user interface 124 to display the best tilt and recline angle for
the user. System responses are anticipated and provided in the
present invention 10 to include at least user touch and voice
commands. Audio recitation and response for visually impaired
individuals is anticipated and provided by the present invention
10. User touch, voice activation and audio recitation functions are
generally programmable and operable on industry standard smart
devices, such as various device models of iPhone, iPad, Samsung
Galaxy, and HP tablets, running operating systems such as Android,
iOS, and Windows, where such devices include an accelerometer.
Implementation on any such device having the minimum function set
as described herein is anticipated.
[0203] FIG. 12d is a non-limiting diagram showing a screen shot of
a web-based implementation of the present invention 10 providing a
user interface 126 to display the best tilt and recline angle for
the user. System responses are anticipated in the present invention
10 to include at least user touch and voice commands. Audio
recitation and response for visually impaired individuals is
anticipated and provided by the present invention 10.
[0204] FIG. 13 is a non-limiting diagram showing the measurement
and notification process 130 for determining proper adjustment of
tilt and recline settings as determined by the present invention
10. Measurement, display, and auditory notification of tilt and
recline angles are accomplished in substantially real-time as a
user adjusts tilt and recline settings on a wheelchair. Actionable
aural guidance is provided to enable the user to achieve
recommended tilt and recline settings suitable to the particular
wheelchair user based on his or her specific profile.
[0205] The present invention 10 can benefit all wheelchair users,
who use a wheelchair with either a tilt or both tilt and recline
functions. Both power and manual wheelchair users can benefit from
this and other functions of the present invention 10. Healthcare
providers and researchers will benefit from the present invention
10, as well. If they use the tilt and recline guidance provided by
the present invention 10, the guidance will be automatically
provided as inputs to the measurement and notification process 130
implemented in source code and operable on a mobile device. If the
health providers and researchers do not use the personalized
guidance, the present invention 10 will allow them to input
alternative tilt and recline (TR) guidelines (see FIG. 14) to the
measurement and notification process 130 so that the wheelchair
users can follow those guidelines.
[0206] As shown in FIG. 13, in step 1 the wheelchair user uses the
goniometer to set the target tilt and recline angles (e.g.,
15.degree. tilt/110.degree. recline) and then click the "Submit"
button (see FIG. 14). If the wheelchair only has the tilt function,
the user only needs to provide the tilt angle.
[0207] In step 2, the goniometer asks the wheelchair user to adjust
the wheelchair to the upright position (i.e., no tilt or recline).
As shown in FIG. 13, the goniometer will use the novel voice alert
technique of the present invention 10 to guide the user. For
example, the voice alert may recite the non-limiting script "Please
make sure that your wheelchair is in the upright position. Touch
anywhere on the screen when you are ready!"
[0208] In step 3, the wheelchair user adjusts the wheelchair to the
upright position following the voice guidance.
[0209] In step 4, the wheelchair user touches the screen of the
smartphone after the wheelchair has been adjusted to the upright
position.
[0210] In step 5, the goniometer asks the user to sit still so that
the goniometer can record the initial position of the smartphone.
This step is needed to ensure the precision of angle
calculation.
[0211] Voice alert is used to guide the user. For example, the
voice alert may recite the non-limiting script "Please do not move
your phone for five seconds." As shown in FIG. 15, the goniometer
may also show the message on the screen.
[0212] In step 6, the goniometer may be configured to ask the user
to adjust the tilt angle by using a voice alert. For example, the
voice alert may recite the non-limiting script "You may now adjust
your position. Please adjust your tilt to 15 degrees."
[0213] In step 7, the wheelchair user starts to adjust the tilt
angle as instructed by the voice alert. In the meantime, the
goniometer will measure and display the current tilt angle on the
screen of the smartphone as shown in FIG. 16.
[0214] In step 8, if the target tilt angle has been reached, the
goniometer may be configured to ask the wheelchair user to stop
with the voice alert. For example, the voice alert may recite the
non-limiting script "Please stop!"
[0215] In step 9, the goniometer may be configured to ask the
wheelchair user to adjust the recline angle by using the voice
alert. For example, the voice alert may recite the non-limiting
script "Please adjust your Recline to 110 degrees."
[0216] In step 10, the wheelchair user starts to adjust the recline
angle. In the meantime, the goniometer will measure and display the
current recline angle on the screen of the smartphone as shown in
FIG. 17.
[0217] In step 11, if the target recline angle has been reached,
the goniometer of the present invention may be configured to use an
aural instruction where the user may be asked with the voice alert
to stop. For example the voice alert may recite the non-limiting
script "Please stop! You are now in your target position." In the
meantime, the goniometer will also show the final angle and the
stop message on the screen of the smartphone as shown in FIG. 18.
Note that 90.degree. of recline represents no recline. Hence, for
15.degree. tilt and 110.degree. recline, the final angle should be
15.degree.+(110.degree.-90.degree.)=35.degree.. The present
invention considers the lag that occurs when the user hears the
voice alert and then stops adjusting the wheelchair position. The
present invention calculates the anticipated time to reach the
target angle based on the angular speed of wheelchair positioning
adjustment. It alerts the user to stop ahead of the anticipated
time to compensate the lag.
[0218] FIG. 14 is a non-limiting diagram showing an exemplary
screenshot of the user interface 140 implemented as an element in
the process for determining proper adjustment of tilt and recline
settings as determined by the present invention 10. A screenshot
for "1: Set the target tilt and recline angles (e.g., 15 tilt/110
recline)" is shown as the first step depicted in FIG. 13. User
instructions and alerts displayed may be accompanied by aural
instructions.
[0219] FIG. 15 is a non-limiting diagram showing an exemplary
screenshot of the user interface 150 implemented as an element in
the process for determining proper adjustment of tilt and recline
settings as determined by the present invention 10. A screenshot
for "5: Alert the user to stay still for 5 seconds" is shown as the
fifth step depicted in FIG. 13. User instructions and alerts
displayed may be accompanied by aural instructions.
[0220] FIG. 16 is a non-limiting diagram showing an exemplary
screenshot of the user interface 160 implemented as an element in
the process for determining proper adjustment of tilt and recline
settings as determined by the present invention 10. A screenshot of
the display on the user interface while the user adjusts the tilt
angle is shown as the seventh step depicted in FIG. 13. User
instructions and alerts displayed may be accompanied by aural
instructions.
[0221] FIG. 17 is a non-limiting diagram showing an exemplary
screenshot of the user interface 170 implemented as an element in
the process for determining proper adjustment of tilt and recline
settings as determined by the present invention 10. A screenshot of
the display on the user interface while the user adjusts the
recline angle is shown as the tenth step depicted in FIG. 13.
[0222] FIG. 18 is a non-limiting diagram showing an exemplary
screenshot of the user interface 180 implemented as an element in
the process for determining proper adjustment of tilt and recline
settings as determined by the present invention 10. A screenshot of
the display on the user interface 180 is shown as the eleventh step
depicted in FIG. 13. This screenshot occurs when the user has
adjusted to the target recline setting. Hence, the wheelchair has
been in the target tilt and recline setting. To let user know that
the target setting has been reached, the actionable aural guidance
is provided to alert the user.
[0223] FIG. 19 is a non-limiting diagram showing the top level
architecture of the mobile-cloud implementation of the present
invention. An artificial neural network is shown implemented in the
cloud, along with data processing and analysis. Researchers and
healthcare providers are able to remotely access patient data
through a secure and controlled interface. The present invention 10
includes a mobile subsystem 191 and a cloud subsystem 192.
Specifically, a mobile computing-based subsystem 191 is provided,
which uses mobile devices (e.g., smartphones) to manage personal
profile, retrieve personalized guidance on wheelchair tilt and
recline (TR) usage, measure wheelchair 193 TR angles, and transmit
TR usage data. Smartphones provide an ideal platform for
implementing the present invention 10 due to the ubiquity of
smartphones, their ever-increasing power, and rich set of sensors,
such as the accelerometer. The present invention 10 provides a
novel algorithm to measure wheelchair 193 TR angles (incline
angles) by using the accelerometer in a smartphone. Specifically,
the position of a smartphone is modeled with a vector
.nu.=.alpha..sub.x, .alpha..sub.y, .alpha..sub.z, which represents
accelerations in three axes measured by the accelerometer. When the
tilt or recline stabilizes to a new angle, accelerations in three
axes will change due to the decomposition of the gravity along the
new angle of the phone. Then, the present invention utilizes the
dot product property to calculate angle changes between two vectors
(positions):
.nu..sub.1.nu..sub.2=|.nu..sub.1|.times.|.nu..sub.2|.times.cos
.theta. (1)
Or equivalently,
.theta.=arccos(.nu..sub.1.nu..sub.2/|.nu..sub.1|.times.|.nu..sub.2|)
(2)
[0224] Hence, no matter how the smartphone is positioned, the TR
angle .theta. between two vectors can be measured. In addition, the
mobile subsystem 191 employs the novel text-to-speech technique,
which enables the system to use voice alerts to guide wheelchair
users for proper TR usage.
[0225] The present invention 10 provides a cloud computing-based
subsystem 192 that can provide personalized guidance on wheelchair
tilt and recline usage using the artificial neural network, and
process, store, and analyze wheelchair 193 TR usage data. This
subsystem employs the cloud computing paradigm, which can provide
virtually unlimited resources for computation and data storage.
Based on the longitudinal TR usage data, the present invention 10
may be used to provide operational applications for mobile devices
to evaluate whether wheelchair users adjust enough TR angles to
relieve seating pressure and whether they frequently reposition
themselves by performing TR functions. The present invention 10 may
be used to provide a novel machine-learning approach to analyze
historical data of an individual wheelchair user, and assess his or
her pressure ulcer (PU) risks correspondingly.
[0226] The present invention 10 may use the Google App Engine (GAE)
as the cloud computing platform. GAE is managed by Google and
provides a platform for developing and hosting web applications.
Note that other techniques may be used to replace GAE. Essentially,
there are currently three options: (1) continue to use commercial
cloud computing platforms, such as Google App Engine, Microsoft
Azure, Amazon EC2, etc.; (2) set up a dedicated private cloud
computing platform; or (3) use a traditional web server as the data
management and computation platform. Other options may emerge in
the future and are anticipated as possible web development and
hosting solutions to support implementation of various features of
the present invention.
[0227] The combination of mobile and cloud computing can yield a
balanced and integrated system, in which the mobile subsystem 191
will collect user's information, display personalized guidance on
TR usage, control the sensor, measure wheelchair TR angles, and
transmit TR usage data to the cloud, while the cloud subsystem 192
will handle the subsequent data management and analysis. Therefore,
the present invention 10 provides a practical way to improve
wheelchair 193 TR usage and capture longitudinal TR usage data
among wheelchair users
[0228] The mobile application of the present invention 10 may be
implemented for any mobile operating system, including the
mainstream mobile operating systems, such as Google Android, Apple
iOS, and Microsoft Windows. To use the mobile application provided
by the present invention 10, the user needs to download it from an
accessible public source where it may be made available, such as
Google Play, Apple Store, or Windows App Store depending on the
mobile operating systems they use.
Additional Embodiments of the Present Invention
[0229] Preferred embodiments of the present invention may comprise
generating personalized adjustment parameters directed to
positioning and control of seating configurations in both
commercial and private automotive vehicles, including trucks and
passenger cars. Outcome objectives may reflect both safety and
comfort. A smartphone implementation providing a user interface to
display at least current position and shape parameters and send
related control parameters to powered, adjustable seats is
anticipated. System responses are anticipated to at least user
touch and voice commands. Audio recitation and response is
anticipated. User touch, voice activation and audio recitation
functions are generally programmable and operable on industry
standard smart devices, such as various device models of iPhone,
iPad, Samsung Galaxy, HP tablets, Google glass, iWatch, etc,
running on operating systems such as Android, iOS, and Windows,
where such devices include an accelerometer. Any such mobile device
having the minimum function set as described herein is anticipated.
Implementation using on-board devices installed as vehicle
equipment is also anticipated.
[0230] Preferred embodiments of the present invention may comprise
generating personalized adjustment parameters directed to
positioning and control of seating configurations in aircraft
including both crew and passenger seating. Outcome objectives may
reflect both safety and comfort. A smartphone implementation
providing a user interface to display at least current position and
shape parameters and send related control parameters to powered,
adjustable seats is anticipated. System responses are anticipated
to at least user touch and voice commands. Audio recitation and
response is anticipated. User touch, voice activation and audio
recitation functions are generally programmable and operable on
industry standard smart devices, such as various device models of
iPhone, iPad, Samsung Galaxy, HP tablets, Google glass, iWatch,
etc, running on operating systems such as Android, iOS, and
Windows, where such devices include an accelerometer. Any such
mobile device having the minimum function set as described herein
is anticipated. Implementation using on-board devices installed as
vehicle equipment is also anticipated.
[0231] Preferred embodiments of the present invention may comprise
generating personalized adjustment parameters directed to
positioning and control of seating configurations in furniture,
including tilt and recline angle, seat and back shape, firmness and
support. Outcome objectives may reflect both safety and comfort. A
smartphone implementation providing a user interface to display at
least current position and shape parameters and send related
control parameters to powered, adjustable seats is anticipated.
System responses are anticipated to at least user touch and voice
commands. Audio recitation and response is anticipated. User touch,
voice activation and audio recitation functions are generally
programmable and operable on industry standard smart devices, such
as various device models of iPhone, iPad, Samsung Galaxy, HP
tablets, Google glass, iWatch, etc., running on operating systems
such as Android, iOS, and Windows, where such devices include an
accelerometer. Any such mobile device having the minimum function
set as described herein is anticipated. Implementation using
on-board devices installed as furniture components is also
anticipated.
[0232] Preferred embodiments of the present invention may comprise
generating personalized adjustment parameters directed to
positioning and control of support and comfort configurations in
both commercial and private sleep platforms for healthcare,
hospitality and in-home applications. A smartphone implementation
providing a user interface to display at least current position and
shape parameters, and send related control parameters to powered,
adjustable seats is anticipated. System responses are anticipated
to at least user touch and voice commands. Audio recitation and
response is anticipated. User touch, voice activation and audio
recitation functions are generally programmable and operable on
industry standard smart devices, such as various device models of
iPhone, iPad, Samsung Galaxy, HP tablets, Google glass, iWatch,
etc., running on operating systems such as Android, iOS, and
Windows. Any such mobile device having the minimum function set as
described herein is anticipated. Implementation using on-board
devices installed as sleep-platform equipment components is also
anticipated.
[0233] Preferred embodiments of the present invention may comprise
using the goniometer functions implemented in a mobile device for
generating and recording personalized parameters directed to
measuring and scoring joint range of motion and flexibility by
clinicians including at least physical therapists, orthopedists,
physical medicine clinicians and sports medicine practitioners.
Goniometric measurements provided using the mobile version of the
present invention may be used as outcome measures (e.g., after a
course of treatment), as an exam finding to aid in the diagnosis of
a condition, and to determine level of fitness for a specific
purpose. System responses are anticipated to at least user touch
and voice commands. Audio recitation and response is anticipated.
User touch, voice activation and audio recitation functions are
generally programmable and operable on industry standard smart
devices, such as various device models of iPhone, iPad, Samsung
Galaxy, HP tablets, Google glass, iWatch, etc., running on
operating systems such as Android, iOS, and Windows. Any such
mobile device having the minimum function set as described herein
is anticipated.
[0234] Those skilled in the art will appreciate that in some
embodiments of the invention, the functional modules of the Web
implementation, as well as the personal and the integrated
communication devices, may be implemented as pre-programmed
hardware or firmware elements (e.g., application specific
integrated circuits (ASICs), electrically erasable programmable
read-only memories (EEPROMs), etc.), or other related components.
Mobile communication devices that can use the present invention may
include but are not limited to any of the "smart" phones or tablet
computers equipped with digital displays, wireless communication
connection capabilities such as iPhones and iPads available from
Apple, Inc., as well as communication devices configured with the
Android operating system available from Google, Inc and with the
Windows operating system available from Microsoft. In addition, it
is anticipated that the new communication devices and operating
systems will become available as more capable replacements of the
forgoing listed communication devices, and these may use the
present invention as well.
[0235] In other embodiments, the functional modules of the
mobile-to-cloud implementation may be implemented by an arithmetic
and logic unit (ALU) having access to a code memory which holds
program instructions for the operation of the ALU. The program
instructions could be stored on a medium which is fixed, tangible
and readable directly by the processor, (e.g., removable diskette,
CD-ROM, ROM, or fixed disk), or the program instructions could be
stored remotely but transmittable to the processor via a modem or
other interface device (e.g., a communication adapter) connected to
a network over a transmission medium. The transmission medium may
be either a tangible medium (e.g., optical or analog communications
lines) or a medium implemented using wireless techniques (e.g.,
microwave, infrared or other transmission schemes).
[0236] The program instructions stored in the code memory can be
compiled from a high level program written in a number of
programming languages for use with many computer architectures or
operating systems. For example, the program may be written in
assembly language suitable for use with a pixel shader, while other
versions may be written in a procedural programming language (e.g.,
"C") or an object oriented programming language (e.g., "C++" or
"JAVA").
[0237] In other embodiments, cloud computing may be implemented on
a web hosted machine or a virtual machine. A web host can have
anywhere from one to several thousand computers (machines) that run
web hosting software, such as Apache, OS X Server, or Windows
Server. A virtual machine (VM) is an environment, usually a program
or operating system, which does not physically exist but is created
within another environment (e.g., Java runtime). In this context, a
VM is called a "guest" while the environment it runs within is
called a "host." Virtual machines are often created to execute an
instruction set different than that of the host environment. One
host environment can often run multiple VMs at once.
[0238] While specific embodiments of the present invention have
been described and illustrated, it will be apparent to those
skilled in the art that numerous modifications and variations can
be made without departing from the scope of the invention as
defined in the appended claims. It is understood that the words
that have been used are words of description and illustration,
rather than words of limitation. Although the invention has been
described with reference to particular means, materials and
embodiments, the invention is not intended to be limited to the
particulars disclosed; rather, the invention extends to all
functionally equivalent structures, methods and uses, such as are
within the scope of the appended claims.
* * * * *