U.S. patent application number 12/571081 was filed with the patent office on 2010-04-01 for systems and methods for monitoring and evaluating individual performance.
This patent application is currently assigned to ARCHINOETICS, LLC. Invention is credited to Jason Akagi, Fahrettin Olcay Cirit, J. Hunter Downs, III, Traci H. Downs, Erin Nishimura, Brendan F. P. O'Donnell, Evan D. Rapoport, J. Patrick Stautzenberger.
Application Number | 20100081889 12/571081 |
Document ID | / |
Family ID | 38832880 |
Filed Date | 2010-04-01 |
United States Patent
Application |
20100081889 |
Kind Code |
A1 |
Downs, III; J. Hunter ; et
al. |
April 1, 2010 |
SYSTEMS AND METHODS FOR MONITORING AND EVALUATING INDIVIDUAL
PERFORMANCE
Abstract
Systems, devices and methods for monitoring and evaluating
cognitive effectiveness are provided. In one exemplary embodiment,
a system of r monitoring cognitive effectiveness can include a
central network communicatively couple with a local computing
device, which can in turn be communicatively coupled with a
portable monitoring device. The portable monitoring device can be
located in proximity with a subject and configured to collect data
from the subject usable in determining a cognitive effectiveness
level. Adaptive methods for determining cognitive performance are
also provided.
Inventors: |
Downs, III; J. Hunter;
(Honolulu, HI) ; Downs; Traci H.; (Honolulu,
HI) ; Stautzenberger; J. Patrick; (Haleiwa, HI)
; Nishimura; Erin; (Honolulu, HI) ; Akagi;
Jason; (Waimanaio, HI) ; O'Donnell; Brendan F.
P.; (Honolulu, HI) ; Cirit; Fahrettin Olcay;
(Honolulu, HI) ; Rapoport; Evan D.; (Honolulu,
HI) |
Correspondence
Address: |
VIRTUAL LAW PARTNERS LLP
P.O. BOX 1329
MOUNTAIN VIEW
CA
94042
US
|
Assignee: |
ARCHINOETICS, LLC
Honolulu
HI
|
Family ID: |
38832880 |
Appl. No.: |
12/571081 |
Filed: |
September 30, 2009 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
11424742 |
Jun 16, 2006 |
7621871 |
|
|
12571081 |
|
|
|
|
Current U.S.
Class: |
600/300 |
Current CPC
Class: |
A61B 5/16 20130101; G16H
40/67 20180101; G16H 50/50 20180101; G16H 40/63 20180101; A61B
5/7232 20130101; A61B 5/4809 20130101; G16H 50/20 20180101 |
Class at
Publication: |
600/300 |
International
Class: |
A61B 5/00 20060101
A61B005/00 |
Claims
1. A method of monitoring individual performance, comprising:
measuring an objective data parameter corresponding to a first
aspect of a subject; collecting a subjective data parameter
corresponding to a second aspect of the subject by way of a user
interface; and determining a cognitive effectiveness level of the
subject based at least on the objective and subjective data
parameters; wherein the subjective data parameter is a data
parameter representative of a subject's manual indication of a
subjective mental state, the subjective mental state including a
wake state.
2. A method of monitoring individual performance using an algorithm
for determining cognitive effectiveness, comprising: measuring at
least one objective data parameter corresponding to at least a
first aspect of a subject; collecting at least one subjective data
parameter corresponding to at least a second aspect of the subject
by way of at least one user interface; and adapting the algorithm
for the subject based on at least on the objective data parameter
and the at least one subjective data parameter for the subject.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This is a continuation of U.S. patent application Ser. No.
11/424,742, filed Jun. 6, 2006, assigned to the assignee of the
present application, and incorporated herein by reference.
FIELD OF THE INVENTION
[0002] The present invention relates generally to the remote
monitoring of individual performance with distributed systems and
evaluating the monitored performance.
BACKGROUND
[0003] Recently, societal trends have indicated that a sizable
portion of the population have been and will continue to suffer
from sleep deprivation. Although much research has been performed
on the causes of and potential remedies for sleep deprivation,
research into the effects of sleep deprivation on the cognitive
performance of an individual has been relatively more limited.
[0004] Much of the research into the effect of sleep loss on
cognitive effectiveness has been directed towards algorithmic
techniques for correlating sleep and wake states with cognitive
effectiveness levels. U.S. Pat. No. 6,241,686 (hereinafter
"Balkin") entitled "System and Method for Predicting Human
Cognitive Performance Using Data from an Actigraph" and U.S. Pat.
No. 6,579,233 (hereinafter "Hursh") entitled "System and Method for
Evaluating Task Effectiveness Based on Sleep Pattern" each propose
algorithmic techniques for correlating sleep loss with cognitive
effectiveness and making past or future predictions of cognitive
performance based, generally, on measured sleep and wake states and
are fully incorporated by reference herein.
[0005] In addition to the algorithmic techniques, these and other
references propose systems for measuring the sleep and wake states
of an individual, performing analysis of this measured data and
presenting the results to a user. However, these approaches lack
the capability to adapt sufficiently to variations between
individual users, fail to present data to users in a fully
comprehensive manner, and suffer from inefficient data collection
and analysis techniques.
[0006] Accordingly, improved devices, systems and methods for
monitoring and evaluating individual performance are needed.
SUMMARY
[0007] Improved devices, systems and methods for monitoring and
evaluating individual performance and the like, are provided in
this section by the way of exemplary embodiments. These embodiments
are examples only and are not intended to limit the invention.
[0008] In one exemplary embodiment, a method of monitoring
individual performance is provided that includes measuring a
objective data parameter corresponding to a first aspect of a
subject, collecting a subjective data parameter corresponding to a
second aspect of the subject by way of a user interface, and
determining a cognitive effectiveness level of the subject based at
least on the objective and subjective data parameters.
[0009] The determined cognitive effectiveness level can be one of a
past, current or future cognitive effectiveness level. Also, the
cognitive effectiveness level can be determined with a cognitive
effectiveness determination algorithm.
[0010] The objective data parameter can be any of multiple types of
parameters. In one embodiment the objective data parameter can be
representative of the subject's physical motion, such as an
actigraph data parameter, while in another embodiment the objective
data parameter can be a parameter representative of a blood oxygen
level of the subject. In yet another exemplary embodiment, the
objective data parameter can be a near infrared spectroscopy (NIRS)
signal parameter.
[0011] The subjective data parameter can be any of multiple types
of parameters. In one embodiment, the subjective data parameter can
be a data parameter representative of a subject's manual indication
of a subjective mental state, such as a wake state and the like. In
another exemplary embodiment, the subjective data parameter can be
a response to a test input by the subject.
[0012] In another exemplary embodiment, the cognitive effectiveness
level of the subject can be determined based on at least the
objective data parameter, and the determined cognitive
effectiveness level can be adjusted based upon the collected
subjective data parameter. For instance, the subjective data
parameter can correspond to a wake state and the boundary of the
wake state can be adjusted based upon the collected subjective data
parameter and a new cognitive effectiveness level can be determined
therefrom.
[0013] In another exemplary embodiment, the cognitive effectiveness
level of the subject can be determined by inputting the objective
and subjective data parameters into a cognitive effectiveness
determination algorithm.
[0014] In another exemplary embodiment, the method can further
include collecting an individual characteristic data parameter
corresponding to a third aspect of the subject and determining the
cognitive effectiveness level of the subject based at least on the
objective, subjective and individual characteristic data
parameters. The individual characteristic data parameter can be a
gender of the subject, a racial origin of the subject, a genetic
characteristic of the subject and the like.
[0015] In another exemplary embodiment, the cognitive effectiveness
level can be determined with a cognitive effectiveness
determination algorithm and the cognitive effectiveness
determination algorithm can be adapted to the traits of the
subject. The adaptation can include altering a population-level
based cognitive effectiveness determination algorithm into a
subject-based cognitive effectiveness determination algorithm.
[0016] Other devices, systems, methods, features and advantages of
the invention will be or will become apparent to one with skill in
the art upon examination of the following figures and detailed
description. It is intended that all such additional systems,
methods, features and advantages be included within this
description, be within the scope of the invention, and be protected
by the accompanying claims. It is also intended that the invention
is not limited to require the details of the example
embodiments.
BRIEF DESCRIPTION OF SEVERAL VIEWS OF THE DRAWINGS
[0017] The details of the invention, both as to its structure and
operation, maybe gleaned in part by study of the accompanying
figures, in which like reference numerals refer to like parts. The
components in the figures are not necessarily to scale, emphasis
instead being placed upon illustrating the principles of the
invention. Moreover, all illustrations are intended to convey
concepts, where relative sizes, shapes and other detailed
attributes may be illustrated schematically rather than literally
or precisely.
[0018] FIG. 1 is a block diagram depicting an exemplary embodiment
of a performance monitoring system.
[0019] FIG. 2 is a block diagram depicting an exemplary embodiment
of a portable monitoring device.
[0020] FIGS. 3A-B are exemplary graphs of cardiac and respiratory
cycles, respectively, derived from data collected by an exemplary
embodiment of the portable monitoring device.
[0021] FIGS. 3C-D are exemplary graphs of spectral data derived
from cardiac cycle data collected by an exemplary embodiment of the
portable monitoring device.
[0022] FIG. 4 is a functional block diagram depicting another
exemplary embodiment of a performance monitoring system.
[0023] FIG. 5 is a flow diagram depicting an exemplary method of
determining a cognitive effectiveness level.
[0024] FIG. 6 is a flow diagram depicting an exemplary method of
providing behavioral guidance.
DETAILED DESCRIPTION
[0025] Provided herein are improved systems and methods for
monitoring and evaluating individual performance including, but not
limited to, individual cognitive performance. FIG. 1 is a block
diagram depicting an exemplary embodiment of performance monitoring
system 100. Performance monitoring system 100 is a distributed
system capable of monitoring the performance of one or more
individual subjects at different physical locations. System 100
preferably includes a central network 102 in communication with one
or more local computing devices 104-1 through 104-N, preferably by
way of a public or private communication path 103, such as the
internet. It should be noted that N can be any number desired.
Local computing devices 104-1 through 104-N are in communication
with a portable monitoring device 105-1 through 105-N by way of a
communication path 106-1 through 106-N, respectively, as depicted
here. Each communication path 106 is preferably a wireless
communication path, although wired communication can be used as
well. Each portable monitoring device 105 is preferably physically
coupled to an individual subject 107 to be monitored. Each of
communication paths 103 and 106 can be bidirectional to allow for
data transfer between each of the devices within system 100.
Furthermore, a single local computing device 104 can also
communicate with multiple portable monitoring devices 105 if
desired.
[0026] In one exemplary embodiment, system 100 operates to monitor
cognitive effectiveness of the various subjects 107-1 through
107-N. For instance, portable monitoring device 105-1 can collect
data from corresponding subject 107-1 and communicate that data to
the corresponding local computing device 104-1 by way of
communication path 106-1. Local computing device 104-1 can then
assemble the data reported from portable monitoring device 105-1,
process and format the assembled data and communicate the resulting
data to central network 102 by way of communication path 103.
Central network 102 can then further analyze the data and output a
cognitive effectiveness level for subject 107-1 back to local
computing device 104-1 or any other desired location.
[0027] In the embodiment depicted here, portable monitoring device
105 is preferably configured to collect one or more data parameters
that correspond to one or more physiological aspects of subject
107. This data parameter, or set of data parameters, is preferably
capable of being used to determine (or predict) a past, present or
future cognitive effectiveness level. The collected data parameter
can be any data parameter usable to determine cognitive
effectiveness including, but not limited to, activity data,
sleep/wake data, event data, objective data, and subjective data
(e.g., data generated from direct user input).
[0028] In one exemplary embodiment, these data parameters can be
representative of the subject's sleep/wake state at a given moment
in time, i.e., whether the subject is awake or asleep at a given
moment in time. These data parameters can be classified as being
objectively or subjectively derived. These data parameters can then
be used to determine a cognitive effectiveness level, which is an
estimated quantification of the relative amount of cognitive
ability, such as decision-making ability, which subject 107 can
expect to have at a given moment in time--before, during or after
collection of the data parameter.
[0029] FIG. 2 is a block diagram depicting an exemplary embodiment
of portable monitoring device 105. In this embodiment, portable
monitoring device 105 can include a sensor 120 configured to
collect data and a communication port 122 configured to communicate
the collected data to another device, preferably local computing
device 104. Portable monitoring device 105 can also include a data
processing unit 124, a bus 125, a memory 126, a power supply 127, a
subjective data input port 128, and/or a display 129.
[0030] Portable monitoring device 105 can collect data of any
desired physiological aspect in any desired manner. For instance,
in one exemplary embodiment, portable monitoring device 105 can
collect physiological data representative of the motion level of
subject 107 with sensor 120 configured as an actigraph. The
collected motion level data can then be correlated to a current
sleep/wake state of subject 107 with an appropriate algorithmic
technique, such as the Cole-Kripke algorithm (described in
"Automatic Sleep/Wake Identification from Wrist Actigraphy"
published in Sleep, vol. 15, pp. 461-469 (1992)) and other
algorithms, not limited to those based on the number of
zero-crossings detected by the actigraph. Subsequently, the
sleep/wake state can be used to determine a cognitive effectiveness
level of subject 107 using an appropriate algorithmic technique,
such as the techniques described in the Hursh or Balkin references
above.
[0031] In another exemplary embodiment, portable monitoring device
105 can collect other physiological data representative of
physiological aspects of subject 107 including, but not limited to
biometric data such as heart rate, breathing rate, blood pressure,
skin temperature, galvanic skin response, blood oxygenation level,
brain and/or nervous system electrical activity level and the like.
In one exemplary embodiment, sensor 120 is configured as an
infrared sensor, such as a near infrared spectroscopy (NIRS)
sensor. In this embodiment, NIRS sensor 120 can collect
physiological data, such as blood oxygenation levels, breath rate,
heart rate and other data that can then be used to determine a
cognitive effectiveness level. An embodiment that implements NIRS
sensor 120 will be described in more detail below.
[0032] Portable monitoring device 105 is preferably configured to
be wearable on subject 107, although device 105 can also be placed
in proximity with subject 107 without making actual physical
contact, if desired. Portable monitoring device 105 can also be
implantable if desired. When configured to be wearable, portable
monitoring device 105 can be worn around the wrist, ankle, neck, or
head, or worn on any other part of the body as desired.
[0033] As mentioned above, portable monitoring device 105 is
preferably configured to communicate with local computing device
104 by way of communication port 122. This communication can occur
in real-time as data is collected, at predetermined intervals, as a
result of prompting by local computing device 104, any combination
thereof or otherwise. Communication port 122 can be configured to
communicate externally in any desired manner. In one exemplary
embodiment, port 122 is configured to communicate wirelessly using
IEEE 802.11, Bluetooth, ZigBee, any mobile communication standard
(e.g., WCDMA, GSM etc.), infrared communication and the like. In
another exemplary embodiment, port 122 is configured as a wired
port, such as a universal serial bus (USB) port, an IEEE 1394
FireWire port, and the like.
[0034] As mentioned above, portable monitoring device 105 can
include subjective data input port 128 and display 129. Subjective
data input port 128 can be any device used to input data including,
but not limited to, a button, switch, rollerball, trackball,
wheel/button combination, touchscreen, touchpanel, keyboard, any
combination thereof and the like. Furthermore, display 129 can be
any type of display including, but not limited to a cathode ray
tube (CRT), liquid crystal display (LCD), light emitting device
(LED), any combination thereof and the like. Also, port 128 and
display 129 can be combined such as in a touchscreen display. Of
course, one of skill in the art will appreciate that certain
configurations are preferable given the preferred configuration of
device 105 as being portable. Thus, the configuration for port 128
and display 129 preferably takes into account weight, size and
space in the actual implementation.
[0035] As will be described in more detail below, port 128 and
display 129 can be used to enter subjective data. The subjective
data can indicate subject 107's current mental state, including a
sleep/wake state, subject 107's self-appraised cognitive
effectiveness level, answers input by subject 107 to cognitive
effectiveness tests, such as tests of reaction time, comprehension
and/or memory, as well as other types of subjective data that will
be appreciated by one of ordinary skill in the art.
[0036] Local computing device 104 is configured to communicate with
portable monitoring device 105 and preferably includes a processing
unit and a user interface. Local computing device 104 is preferably
a personal desktop or notebook computer, but can also be a handheld
computing device such as a personal digital assistant (PDA), a
personal email device (e.g., a BLACKBERRY), a mobile phone, any
combination thereof and the like. Local computing device 104
preferably receives data collected by portable monitoring device
105 and processes it into a desired format. This processing can
include data conditioning, data filtering, encryption, and/or
compression. Local computing device 104 can be configured to send
the formatted data to central network 102.
[0037] Central network 102 preferably includes a processing unit
configured to process the formatted data from local computing
device 104 into a cognitive effectiveness level. A data analyst can
manually monitor and process data received at central network 102,
or all data monitoring and processing can be performed
automatically. Once the data has been processed into the desired
format, the data can be sent back to subject 107 for assessment or
can be sent to a third party (embodiments involving a third party
will be discussed in more detail below).
[0038] It should be noted that the division of data processing
between central network 102, local computing device 104 and
portable monitoring device 105 can be organized in any manner
desired. For instance, the bulk of data processing can be performed
at central network 102 if desired, with local computing device 104
acting as an intermediary to transfer data from portable monitoring
device 105 to central network 102. Alternatively, the bulk of data
processing can be performed on either portable monitoring device
105 or local computing device 104, with communication with central
network 102 for the purpose of data processing avoided
altogether.
[0039] For instance, in one exemplary embodiment, raw actigraphy
data can be collected in the form of a zero-crossing count. This
information can then be stored in memory on portable monitoring
device 105, where it can be time-stamped and transmitted at a
predetermined interval. Local computing device 104 can then receive
the data, compile it into tabular format, store a back-up copy of
the raw data, compress the data, condition the data, encrypt the
data, and transmit the modified actigraphy data along with any
other data, such as subjective data and other data entered into
local computing device 104 (such as travel information, event start
time, event end time, individual-specific pieces of data such as
age, sex or location, and the like) to central network 102 where
this data is input to the cognitive effectiveness determination
algorithm (such as the algorithm described in the Hursh or Balkin
references, which are incorporated above).
[0040] Similarly, the evaluation or assessment of the cognitive
effectiveness level can be performed in any manner desired. In one
exemplary embodiment, subject 107 can evaluate the cognitive
effectiveness data through a user interface located on local
computing device 104, or through display 129 located on portable
monitoring device 105. A software tool can be used to facilitate
presentation of the cognitive effectiveness data to subject 107.
One exemplary tool is the Fatigue Analysis Software Tool/Fatigue
Avoidance Scheduling Tool (FAST), described in published U.S.
Patent Application no. 2003/0018242, entitled "Interface for a
System and Method for Evaluating Task Effectiveness Based on Sleep
Pattern," which is fully incorporated by reference herein.
[0041] As mentioned above, portable monitoring device 105 can be
configured to monitor physiological aspects of subject 107 with an
NIRS sensor 120. In one exemplary embodiment, NIRS sensor 120 is
configured to measure the oxy- and deoxy-hemoglobin concentrations
in subject 107's blood with NIRS. From this NIRS signal, other
physiological signals can be obtained or derived. For instance, the
dominant measured NIRS signals can include signal oscillations from
each heart and respiratory cycle. These signal oscillations can
cause rises and drops in the NIRS signal, corresponding to each
heart beat and breath taken by the subject. These signal
oscillations are overlaid on the raw NIRS signal and can be
extracted using filtering. The heartbeat signal, extracted through
a band-pass filter, is depicted in the exemplary graph of FIG. 3A.
The respiratory signal, extracted through a low-pass filter from
the same data, is depicted in the exemplary graph of FIG. 3B, with
the same time window as in FIG. 3A. The signals of FIGS. 3A-B can
be used to determine the heart and respiratory rates, respectively,
of subject 107. This processing can occur in any desired processing
unit within system 100.
[0042] The heartbeat signal can be used to derive a heart rate
variability (HRV) signal. In one exemplary embodiment, the HRV
signal is derived through a peak detection algorithm and outlier
removal based on an estimated change in heart rate between
consecutive beats. The HRV signal is then spectrally analyzed using
a fast fourier transform (FFT). Spectral analysis of the HRV signal
has been shown to reveal indicators of sleep onset and fatigue
levels, after the removal of motion artifacts from the received
optical signal, preferably through artifact removal algorithms.
[0043] FIGS. 3C-D are exemplary graphs depicting spectral data from
the HRV signal derived from heart rate data. This spectral data
shows expected changes due to sleep onset and arousal. FIG. 3C
depicts an expected increase in the high frequency variance (HFV)
of the HRV signal due to the onset of sleep. FIG. 3D depicts the
expected peak in the low frequency variance (LFV) due to an arousal
from sleep. Respiratory rate can be derived from the NIRS signal as
well and can be used to reveal indicators of sleep onset and
arousals, if desired.
[0044] In addition to collecting objective data representing a
physiological aspect of subject 107, system 100 can be configured
to collect subjective data as well, and determine cognitive
effectiveness based on both the collected objective and subjective
data. In one exemplary embodiment, objective data collected by
sensor 120 is processed and correlated to a sleep/wake state of
subject 107 using a sleep/wake state determination algorithm, and
then processed to determine a cognitive effectiveness level using a
cognitive effectiveness determination algorithm. Portable
monitoring device 105 can be configured to allow subject 107 to
manually input an indication that subject 107 is awake through
subjective data input port 128. Subject 107 can be prompted to
enter subjective input by device 105 or subject 107 can enter
subjective input voluntarily. This subjective data can be used to
verify the sleep/wake state that was determined based on the
objective data. If a discrepancy exists, the bounds for the
sleep/wake state can be modified accordingly to yield a more
accurate cognitive effectiveness prediction.
[0045] System 100 can also be configured to collect subjective data
relating to the cognizant state of subject 107 at a given point in
time. This data can be collected through a cognitive effectiveness
test that evaluates subject 107's ability to perform tasks of
varying degrees of difficulty. Examples can include tests of
subject 107's reaction time, ability to comprehend relatively
complex information, short term memory, long term memory and the
like. These tests can be administered through the local computing
device 104, through the portable monitoring device 105, or through
other devices. The results of the test or tests can be reported to
system 100 and used verify the accuracy of subject 107's current
predicted cognitive effectiveness level or the test results can be
used to directly predict subject 107's cognitive effectiveness
level at a past, current or future point in time.
[0046] Also, correlation can be performed with the objective data
to adjust the sleep/wake determination algorithm to attempt to
eliminate future discrepancies. For instance, if the sleep/wake
determination algorithm incorrectly predicts a sleep state based on
certain objective data characteristics, the algorithm can be
modified, or tuned, to more accurately predict sleep/wake states in
the future. As a result, system 100 can be configured to
dynamically adapt to the characteristics of each individual subject
107. Preferably, this is an iterative process that continues over
time, in order to continually adapt the algorithm to the
characteristics of the individual subject 107.
[0047] For instance, in another exemplary embodiment, system 100
can be configured to implement an adaptive algorithm by first
measuring activity with actigraph sensor 120 and determining a
corresponding sleep/wake state based on the Cole-Kripke algorithm.
Sleep scores are assigned to epochs of uniform time period based on
the collected objective actigraph data. Subjective data, when
available, can be used to modify the bounds for sleep and wake
periods. The subjective data can be stored and used to individually
optimize the Cole-Kripke algorithm parameters and weights for a
given subject 107.
[0048] In another exemplary embodiment, a host of physiological
aspects and other individual-specific characteristics are used to
adaptively improve the accuracy of the cognitive effectiveness
predictions over time. FIG. 4 is a functional block diagram
depicting an exemplary embodiment of system 100 configured for
algorithmic adaptation. Here, system 100 includes a processing unit
401, which can be located anywhere throughout system 100 (if
implemented in portable monitoring device 105, processing unit 401
would be equivalent to processing unit 124). Here, processing unit
401 includes a cognitive effectiveness prediction unit 409 that is
configured to use a customizable population-level model for
predicting a cognitive effectiveness level 408. In this embodiment,
objective physiological data 402, subjective data 404, and/or
individual characteristic data 406 are each collected and input
into processing unit 401, for use by cognitive effectiveness
prediction unit 409.
[0049] Objective physiological data 402 can be any data relating to
the measurable aspects of the body of subject 107, such as
motion/activity data, heart rate, breathing rate, blood pressure,
skin temperature, galvanic skin response, blood oxygenation level,
brain and/or nervous system electrical activity level and the like.
This data 402 is preferably obtained by sensor 120 within portable
monitoring device 105.
[0050] Subjective data 404 can be any data that is based on a
subjective assessment of subject 107. Subjective data 404 is
typically input by subject 107, but can also be input by another
individual possessing information on the state of subject 107.
Subjective data 404 can be data confirming whether subject 107 is
currently asleep or awake, data confirming whether subject 107 was
asleep or awake at a past point in time, data representative of
when subject 107 plans to sleep or wake in the future, data
representative of a current or past cognitive effectiveness level
of subject 107, data representative of a current or past feeling of
exhaustion, feedback data representative of subject 107's
evaluation of a previously output cognitive effectiveness level 408
and the like. This data 404 is preferably obtained by subjective
input into subjective data input port 128 in portable monitoring
device 105 or subjective input into local computing device 104.
[0051] Individual characteristic data 406 can be any data that is
representative of a broad individual characteristic. Examples can
include race, gender, pharmacological data, heritage, genetic
traits and the like. This data 406 is preferably obtained by input
into subjective data input port 128 in portable monitoring device
105, input into local computing device 104 by subject 104 or input
into central network 102 by another party with knowledge of the
individual characteristics of subject 107 freely disclosed by
subject 107.
[0052] Cognitive effectiveness prediction unit 409 is preferably
configured to predict a cognitive effectiveness level 408 based on
any or all of the input data 402-406 and, also, adapt itself to
improve the accuracy of predictions based on any or all of the
input data 402-406. In one exemplary embodiment, feedback provided
by subject 107 on the accuracy of a current cognitive effectiveness
level 408 can be input as subjective data 404 and used by cognitive
effectiveness prediction unit 409 to rescale or individually
customize the population-level model appropriately to bring further
predictions in-line with subject 107's self-assessment. In other
words, the population-level model (or algorithm) can be customized,
adapted, or individualized into a relatively more subject-based
model to improve the accuracy of predictions for a given subject
107. In doing so, the model is no longer optimized to produce
accurate predictions for any member of a given population, but is
instead optimized for a subset, or even an individual, within the
population.
[0053] Furthermore, adjustments to the population-level model as
well as data 402-406 can be shared with central network 102 and
used to correlate changes to the population-level model with
specific input data 402-406 to improve the accuracy of the
population-level model for other individuals. For instance, based
on adjustments made for male and female subjects 107, the
population-level model can be customized to take into account the
gender of a new subject 107 using system 100 for the first
time.
[0054] In one exemplary embodiment, cognitive effectiveness
prediction unit 408 is configured to use a boosted decision tree
technique to adapt the population-level model to a subject-based
model for a given subject 107. FIG. 5 is a flow diagram depicting
one exemplary method 500 where cognitive effectiveness prediction
unit 408 uses a boosted decision tree technique. At 502, objective
data 402 (and/or optionally individual characteristic data 406) is
collected for subject 107. At 504, subjective data 404 is collected
with regard to subject 107. Preferably, subjective data 404
includes self-assessed cognitive effectiveness levels entered by
subject 107. At 506, a decision tree is generated that predicts the
self-assessed cognitive effectiveness levels included in subjective
data 404 based on objective data 402 and/or the output of a
population-level model.
[0055] At 508, new cognitive effectiveness levels are output from
the boosted decision tree based on newly acquired objective data
402 (and/or optionally individual characteristic data 406). At 510,
new subjective data 404 is entered with regard to subject 107,
preferably including subject 107's self-assessed cognitive
effectiveness levels. At 512, discrepancies between subject 107's
self-assessed cognitive effectiveness levels and the predicted
cognitive effectiveness level output by the decision tree are
identified and a higher weight is assigned to those instances. At
514, a new decision tree is generated to predict the reweighted
data. At 516, this new tree is combined with previous trees by
weighted voting with the weighting based on the overall prediction
error of each tree. At 518, the accuracy of the combination is
evaluated and if further modifications are necessary, the method
reverts to 512. If not, the method proceeds to 520 and waits for
new types of objective data 402 (or individual data 406) to be
acquired or changes in the base level population model, in which
case the method reverts to 506.
[0056] In addition to providing a predicted level of cognitive
effectiveness, system 100 can be configured to provide behavioral
guidance to subject 107 based on the predicted cognitive
effectiveness level. The behavioral guidance can be provided in
real-time and can inform subject 107 as to the effect current
and/or future behavior will have on subject 107's cognitive
effectiveness level. As mentioned above, portable monitoring device
105 can include display 129. Portable monitoring device 105 can
generate a behavioral guidance message with processing unit 124 (or
receive a behavioral guidance message by way of communication port
122) and present the message to subject 107 on display 129. A
behavioral guidance message can be generated at pre-determined
intervals during certain times of the day, can be generated as a
result of prompting by subject 107, can be generated as a result of
variations in data collected by sensor 120, or in any other desired
manner.
[0057] Preferably, the conditions under which a behavioral guidance
message should be generated are stored in system 100 as a set of
behavioral guidance parameters. The behavioral guidance parameters
can include time parameters, date parameters, predicted cognitive
effectiveness level parameters, objective data parameters,
subjective data parameters, individual characteristic data
parameters, interval parameters (i.e., how often should a
behavioral guidance message be generated and is this interval fixed
or variable dependent on the time of day, date, severity of
collected objective or subjective data, etc.) and the like.
[0058] In one exemplary embodiment, a behavioral guidance message
informs the user as to what consequences sleep time will have on
cognitive effectiveness for the next day. For instance, if a set of
behavioral guidance parameters are used that specify that should
system 100 detect that subject 107 is still awake past a
predetermined time of day, for instance, 10 o'clock p.m., system
100 can be configured to generate a behavioral guidance message and
display it on portable monitoring device 105. The behavioral
guidance message can contain information such as an average
cognitive effectiveness level for a specific event or time period
in the future subject 107 can expect to achieve if subject 107 goes
to sleep within 30 minutes (e.g., 90% of maximum), 60 minutes
(e.g., 85% of maximum) and 90 minutes (e.g., 80% of maximum) from
the current time, and wakes at a normal time the following day. In
such an instance, subject 107 preferably sees a quantified
representation of the consequences of continuing to remain awake,
through the ever-decreasing predicted cognitive effectiveness
levels. Updated behavioral guidance messages can be continually
generated at predetermined intervals until system 100 detects that
subject 107 has gone to sleep.
[0059] FIG. 6 is a flow diagram depicting an exemplary embodiment
of a method 600 of providing behavioral guidance, which can be
performed with system 100. At 602, one or more behavioral guidance
parameters are set. Next, at 604, system 100 monitors the current
conditions, which can include collected data (e.g., objective data
402, subjective data 404, and/or individual characteristic data
406) as well as generally known data (e.g., time, day, etc.). Then,
at 606, system 100 interprets the monitored conditions in light of
the behavioral guidance parameters and, at 608, determines whether
the monitored conditions are such that a behavioral guidance
message should be generated. If not, then system 100 proceeds to
610 to determine if new behavioral guidance parameters are to be
set. If so, system 100 proceeds back to 602 and if not, system 100
proceeds back to 604. If a behavioral guidance message should be
generated, then system 100 can proceed from 608 to 612 where the
behavioral guidance message is generated. After which, system 100
can proceed to 610 to determine if new behavioral guidance
parameters are to be set.
[0060] While the invention is susceptible to various modifications
and alternative forms, a specific example thereof has been shown in
the drawings and is herein described in detail. It should be
understood, however, that the invention is not to be limited to the
particular form disclosed, but to the contrary, the invention is to
cover all modifications, equivalents, and alternatives falling
within the spirit of the disclosure.
* * * * *