Dr. Jeffrey Palmer

Group Leader

Bioengineering Systems & Technologies

MIT Lincoln Laboratory

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

• Wearable sensing and analytics will greatly improve health and performance

• Challenges

• Future

• 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

• Wearable sensing and analytics will greatly improve health and performance

• Challenges

• Future

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

Accuracy of Wearable Devices for Estimating Total Energy Expenditure: Comparison With Metabolic Chamber and

Doubly Labeled Water Method, Murakami et al, JAMA Internal Medicine, March 21, 2016

TEE: Total Energy Expenditure

19 Subjects

24 hours in metabolic chamber