Upload
lamduong
View
217
Download
0
Embed Size (px)
Citation preview
Approved for public release: distribution unlimited.
This material is based upon work supported by the Assistant Secretary of Defense for Research and
Engineering under Air Force Contract No. FA8721-05-C-0002 and/or FA8702-15-D-0001. Any opinions,
findings, conclusions or recommendations expressed in this material are those of the author(s) and do
not necessarily reflect the views of the Assistant Secretary of Defense for Research and Engineering.
© 2016 Massachusetts Institute of Technology.
Delivered to the US Government with Unlimited Rights, as defined in DFARS Part 252.227-7013 or 7014
(Feb 2014). Notwithstanding any copyright notice, U.S. Government rights in this work are defined by
DFARS 252.227-7013 or DFARS 252.227-7014 as detailed above. Use of this work other than as
specifically authorized by the U.S. Government may violate any copyrights that exist in this work.
Predict, restore and improve health and performance
through continuous, personalized technologies
Population Informatics
Molecular Analysis
Human Systems Monitoring
Tactical
Clinical Work/Home
Environments
Focus AreasCharacteristics
Individualized
Persistent Ubiquitous
http://www.passbiology.co.nz/biology-level-3/homeostasis/homeostasis_negative_feedbacks1315428467622.png?attredirects=0
• Multiple, nested control
systems that regulate living
systems
• Keeps organisms “healthy” in
face of a changing external
environment
• Autonomic
• Operates at multiple spatial
and temporal scales
• Limits on dynamic range
(sensors) and ability to
compensate (effectors) for
extremes
Thermo-regulation
• Degraded health and performance can
be improved with external sensors and
effective response
• Wearable sensors provide objective,
digital data that the system is out of
homeostatic bounds (e.g. temperature,
glucose, etc.)
• Analytics provide state estimates and
recommended course of action
• Response changes the system and/or
environment to restore homeostasis
Sensor Effector
Human System
WearableSensor
Response
Analytics
Assess
Effective response must be taken to restore health, make sensor data actionable
Measurement
Technologies/Sensors
Physiological Data Across
Time & Size Scales
Blood Pressure Sensor
Blood Pressure
ECG Heart Rate & Cardio Health
Microphone Speech Parameters
Thermometer Temperature
EEG Brain Activity
Wrist-worn Sensors
Heart Rate & Activity
Metabolic Monitor
Macronutrient Oxidation
Ultrasound Imaging and Tissue Properties
Load Cell & IMU
Gait, Load, Balance
Prediction
Domains
Psych
Health
CBA Exposure
Infectious Disease
Musculo-skeletal
Injury
Cognitive & Physical
Fatigue
Thermal
Strain
Predictive Analytics
Key applications revolve around known degraders of health, measurable signals, and effective
responses that resolve the threat
Psych Health
CBA Exposure
Thermal Strain Infectious Disease
Musculoskeletal Injury
Cognitive & Physical Fatigue
• Sense neuro-behavioral changes
to infer cognitive states
• Respond with therapy and/or Rx
• Sense molecular and/or
physiological indicators of
exposure
• Respond with decon/isolation/Rx
• Sense physiological signal and
infer core temperature
• Respond with
rest/hydration/clothing/cooling
• Sense molecular and/or physiological
indicators of infection
• Respond with isolation/Rx
• Sense neuro-behavioral changes to
quantify fatigued states
• Sense and estimate energy expenditure
• Respond with load balancing, rest
• Sense changes in gait
• Sense bone health
• Respond with reduced load, rest, surgery
The National Research Action Plan* highlights the critical need for diagnostic and rehabilitative biomarkers and screening tools for PTSD, TBI and associated co-morbidities
* Improving Access to Mental Health Services for Veterans, Service Members and Military Families, Aug, 2012
Detecting Moderate/Severe Depression
0 0.2 0.4 0.6 0.8 1.00
0.2
0.4
0.6
0.8
1
Probability of False Alarm
Pro
bab
ilit
y o
f D
ete
cti
on
AVEC 2013
AVEC 2014
AVEC 2013
AVEC 2014
AVECInternational Audio/Visual Emotion
Challenge and Workshop
Affect and Depression
2013 Features:
• Articulatory coordination
• Phonetic timing
2014 Features:
• Facial muscle coordination
• Articulatory-vocal fold
coordination
CB: Chemical / Biological SEIRD: Susceptible/Exposed/Infected/Recovered/Dead RIF: Reduction in Force
Actions informed by early warning reduce the severity of outbreak.
Host response vital signs enable early warning of viral hemorrhagic fever infection 1-3 days in advance of fever.
Reduce transmission from 2 to 1
and
Reduce time to recovery by 50%
1) Quicker isolation of infected reduces
transmission frequency
2) Early stage medical countermeasure
efficacy reduces recovery time
Model Example:
Ten Soldiers from unit of 110 exposed to smallpox at infectious doses
Max RIF
Max RIF
Statistic Baseline Proactive Reduction
Max RIF 67% 26% 60%
Dead 30% 21% 30%
Susceptible Exposed Recovered2
Infected
Dead1
SEIRD Epidemiological Model
• Meets field health/perf. challenges
• Actionable data and effective response
• Tactically “acceptable”
• Low probability of detect/intercept
Sensor(s)
Power
Communications
Architecture/Interface
Signal processing
Applications
Interface with body
Key Wearable Sensor Subsystems Critical Field Issues
• Miniaturization and SWaP
• Integrating multiple sensors
• Embedded processor
• Possibly power management
• Radios (at the component level)
Likely Solved by Industry
• Fidelity/strength of signal
• Comfort/acceptance
• Minimum 72 hour mission duration
• No re-charging preferable
• Optimized suite for the application
• Accuracy
• Compensate for motion artifacts
• Power efficient algorithms
• Common, govt.-owned interfaces
• Cyber-secure throughout
• Enables integration of new technology
Field Sensor System
Stated Needs
COTS Consumer Grade COTS Medical Grade
Microsoft, Fitbit, Mio, Apple… Zephyr, Hidalgo, VitalSense
CommunicationsNo Bluetooth,
complete user controlBluetooth, Wi-fi Bluetooth, Wi-fi
Mission profile 72+ hours 24-48 hours 12-24 hours
Integration &
extensibilityOpen architecture Proprietary Proprietary
Accuracy High under field conditions VariousFDA 510k certified,
military field tested
DisplayReal-time actionable
information for
squad of 9-13
Basic readouts for
single device
Basic readouts for
team on laptop
Cost Minimal Under $0.5k > $1k
Commercially available systems do not completely meet the stated needs
BasisPeak
MioLink
MicrosoftBand
ScoscheRhythm
LifeBEAMCap
SpreeHeadband
PolarStrap
“Evaluation of Commercial Heart Rate Monitors”, MIT LL Project Report, 20 November, 2015
-80 -60 -40 -20 20 40 60 800
Heart Rate Error (bpm)
Sitting
Ambulation
Calisthenics
Overall
Heart Rate Accuracy (95% Bounds)
Wrist/Forearm
Head
Chest
Real-Time Physiological Status MonitoringLow SWaP
Sensors
Lincoln Contributions to Major Gaps
Goal: Develop, integrate, and transition low SWaP sensors, apps, and tactically-acceptable
communications for an open system enabling health monitoring and performance prediction
Open Systems Architecture
Cognitive• Voice / stress / emotion
Physiological• Heart rate/temp• Fluid intake
Load Carriage• Force sensor• Accelerometer
Environmental• Noise• Temperature / humidity
Tactically
Acceptable
Communications
Network
EmbeddedAnalytics
On-body Processing and
Mobile Applications
• Thermal strain
• Energy expenditure
• Injury prediction
• Mission planning
• Altitude readiness
• Hydration status
• Infection warning
Provide short-range (3-5m) data link from squad members to leader for 72-hour mission
Leader or
Medic/CorpsmanSmartphone with OBAN-PSM
Radio Dongle
Squad MemberCOTS Chest Strap
with OBAN-PSM
Sensor HubLow power
narrowband wireless link
1–1000 KHz 1–1000 MHz 1–10 GHz
Tunable Wideband
• 3.1–10.6 GHz
Industrial, Scientific, and Medical (ISM) Bands
• [1] 902.0–928.0 MHz
• [2] 2.400–2.500 GHz
• [3] 5.725–5.875 GHz
• ZigBee (IEEE 802.15.4) 1, 2
• Bluetooth (IEEE 802.15.1) 2
• ANT (proprietary) 2
• Wi-Fi (IEEE 802.11) 2, 3
Tunable Narrowband
• 200–960 MHz
Near-field Magnetic Induction
• [1] < 135 KHz
• [2] 13.56 MHz
• RFID 1,2
• NFC 2
• Polar 1
RF Spectrum
Individual Variation
Genome
Traits/Biomarkers
Interactome
Sleep FoodRx/
Exposures
Exposome
Pathogens
Combat-Related Stress
Powerful New Solutions to Biomedical Problems
Genotypes (SNP)
Exomes
Whole Genomes
Integrated individualized biomedical data architectures
Infectious Disease Resilience
Performance Optimization Injury Prevention
Mental/Physical Rehabilitation
Heart Rate, perspiration, EEG
Cognition, motor
Vocal, facial, eye movement
Pathway function
RNA/miRNA
Clinical
Activity
Open Data Architecture DecisionSupport
DataExploitation
Direction, Injury Avoidance, & Improved Health
• Individualized recommendations
• Interventions
• Optimized readiness
• Medical cost savings
Individuals
Lifespace
Monitoring Tools
Data Upload
& Feedback
Feed
back
• Data mining
• New correlations
• Risk / Outcomes
• Research
• System assessment metrics
• Consent
• Privacy
• Security
• Raw data standardization
Global
Information
Aggregated
Information
Individualized
Information
Tiered
Access
Data-Driven Optimization of Health,
Performance, and Readiness
Medical
Individuals
Leadership
ASSIST:Advanced Self-Powered Systems of
Integrated Sensors and Technologies
An NSF Engineering Research Center
• America’s first-ever flexible hybrid electronics manufacturing institute
• Brings together companies, universities and non-profits to improve national economic competitiveness and enable new capability
• Development centered around Technology Platform Demonstrations
– Wearable Medical/Human Monitoring
– Asset Monitoring
– Integrated Array Antennas
– Soft Robotics
• Heart and respiratory rate
• Core temperature
• Wireless readout
Ingestible Implantable/Bioengineered
• Sense neurotransmitters in vivo
• Additional small molecules
• Wireless readout
Bioengineered
Technology trends enable increased insight into the inner workings of the human system
Sensing
Motility
Catalysis
Replication
PersistenceFeedback
• Engineer human/microbial cells
• Incorporate sense and response with computational logic
2.5 cm
1 cm