MAKING ORTHOTICS SMARTER TO OPTI by fjwuxn

VIEWS: 4 PAGES: 33

									1        MAKING ORTHOTICS SMARTER TO OPTIMIZE FUNCTIONAL
2           AMBULATION FOR PERSONS WITH DISABILITIES
3                      BY BRADFORD C. BENNETT
4
5                THE MODERATOR:   GOOD MORNING.   WELCOME TO THE
6    LAST OF THE SERIES OF TUTORIAL SESSIONS FOR THE
7    NATIONAL ACADEMIES KECK FUTURES INITIATIVE ON SMART
8    PROSTHETICS.    MY NAME IS KHAL SALEH.   I'M YOUR
9    MODERATOR FOR THE MORNING.     I'M FROM THE UNIVERSITY OF
10   VIRGINIA.    I HAVE THE PLEASURE OF MODERATING THIS
11   SESSION WITH SEVERAL GOOD OUTSTANDING SPEAKERS WE HAVE
12   THIS MORNING.
13               THIS IS THE FOURTH TUTORIAL WEBCAST THAT WE
14   ARE EXCITED TO PRESENT TO YOU THIS MORNING.      THESE
15   WEBCASTS ARE MEANT TO OVERCOME THE DIFFERENCES IN
16   TERMINOLOGY AND HURDLES BETWEEN DIFFERENT
17   INTERDISCIPLINARY AND DISCIPLINARY FIELDS.      EACH EXPERT
18   WILL PROVIDE AN OVERVIEW OF HIS OR HER FIELD, LANGUAGE,
19   AND AREA OF INTEREST.    WHETHER IT'S RESEARCHERS,
20   SCIENTISTS, ENGINEER, OR MEDICAL RESEARCHERS, THE POINT
21   IS TO COLLATE THESE PEOPLE INTO THE SAME PLAYING FIELD.
22               IF YOU HAVE QUESTIONS, WE URGE YOU TO SEND
23   THEM TO MARTY PERREAULT, WHICH YOU ALREADY HAVE THE
24   CONTACT TO, DURING THE WEBCAST.     THIS WILL HELP US IN
25   MINIMIZING THE TURNAROUND TIME BETWEEN THE QUESTIONS,



                                   1
1    AND WE KINDLY ASK YOU TO GET THOSE BEFORE THE SPEAKER
2    IS FINISHED.
3                SO WITHOUT FURTHER DELAY, I WOULD LIKE TO
4    INTRODUCE OUR NEXT SPEAKER.         BUT I'M ALSO -- MARTY
5    ASKED ME TO REMIND EVERYBODY THAT THE COPY OF THE
6    PRESENTATIONS WILL BE -- ALREADY HAVE BEEN PUT ONLINE.
7    SO OUR FIRST SPEAKER.      IT'S A PLEASURE TO PRESENT A
8    COLLEAGUE AND A FRIEND FROM THE DEPARTMENT OF
9    ORTHOPEDIC SURGERY AT THE UNIVERSITY OF VIRGINIA,
10   DR. BRAD BENNETT IS HERE WITH US.        HE'S GOING TO BE
11   TALKING ABOUT MAKING ORTHOTICS SMARTER TO OPTIMIZE
12   FUNCTIONAL AMBULATION FOR PERSONS WITH DISABILITIES.
13               HE IS THE DIRECTOR OF THE MOTION ANALYSIS AND
14   MOTOR PERFORMANCE LABORATORY AND ASSISTANT PROFESSOR AT
15   THE DEPARTMENT OF ORTHOPEDICS AT THE UNIVERSITY OF
16   VIRGINIA.    SO WITHOUT FURTHER ADO, DR. BENNETT.
17               DR. BENNETT:    THANK YOU, KHAL.    IT'S A
18   PLEASURE TO BE HERE TODAY TO SPEAK TO EVERYBODY OUT
19   THERE IN CYBERSPACE.       LET ME START BY THANKING THE
20   NATIONAL ACADEMIES AND THE KECK FUTURES INITIATIVE BOTH
21   FOR INVITING ME TO BE HERE TO SPEAK TODAY AND FOR
22   CHOOSING THIS VERY INTERESTING AND, I THINK, IMPORTANT
23   TOPIC OF SMART PROSTHETICS.
24               I'M GOING TO TALK TODAY ABOUT HOW WE MIGHT
25   IMPROVE ORTHOTICS TO HELP PEOPLE WALK.         AND I WAS



                                     2
1    STRUCK BY SOME EARLIER SPEAKERS WHO MENTIONED THE
2    CONCEPT OF BIOCOMPATIBILITY.       THESE DEVICES, WHICH ARE
3    MOSTLY EXTERNAL TO THE BODY, STILL NEED TO HAVE SOME
4    TYPE OF BIOCOMPATIBILITY OR BIOFIDELITY OR EVEN
5    BIOMECHANICAL FIDELITY.    THEY NEED TO HAVE PROPERTIES
6    THAT THE PERSON WITH THEIR LEVEL OF ABILITY CAN EXPLOIT
7    TO HELP THEM WALK BETTER.    SO THIS CONCEPT OF
8    MAN-MACHINE INTERFACE AND HAVING THIS BIOFIDELITY IS
9    VERY IMPORTANT FOR US.
10               WELL, WE'LL GET THIS TO ADVANCE HERE, I'M
11   SURE.   THERE WE GO.   SO I'M GOING TO START BY GIVING A
12   LITTLE ABC'S OF WALKING, GIVING YOU THE BASIS FOR THE
13   UNDERLYING FUNCTION THAT THESE PEOPLE HAVE TO DO.      THEN
14   I'M GOING TO DISCUSS CURRENT DEVICES WHICH ARE
15   AVAILABLE TODAY, AND THEY'RE MOSTLY PASSIVE DEVICES.
16   NEXT I'M GOING TO TALK ABOUT SOME INTERESTING NEW
17   RESEARCH PEOPLE ARE DOING, AND FINALLY END UP BY
18   DISCUSSING THE ISSUES I THINK WE NEED TO ADDRESS TO
19   REACH THE NEXT GENERATION OF DEVICES.
20               BIPEDAL WALKING IS VERY MUCH A HUMAN
21   ACTIVITY.    WE EVEN DEFINE OURSELVES IN THIS WAY OFTEN.
22   AND MANY PEOPLE THINK IT WAS THIS ABILITY TO WALK
23   UPRIGHT, HAVE OUR HANDS FREE TO GO OUT AND BRING FOOD
24   BACK TO THE FAMILY THAT LED US ON OUR EVOLUTIONARY PATH
25   THAT BROUGHT US HERE TODAY.       BUT IT IS NOT AN INNATE



                                   3
1    ABILITY; THAT IS, IT IS A LEARNED ABILITY.   MOST PEOPLE
2    TAKE THEIR FIRST STEPS BETWEEN THE AGES OF ONE AND TWO
3    AND THEN REACH A MATURE ADULT-LIKE GAIT AROUND THE AGES
4    OF SEVEN OR EIGHT.   WHAT'S GOING ON IN THE INTERIM IS
5    PRACTICE, THOUSANDS AND THOUSANDS OF HOURS OF PRACTICE.
6    SO WE HAVE TO REMEMBER THAT WALKING IS A VERY
7    SOPHISTICATED LEARNED SKILL THAT WE'VE JUST SPENT YEARS
8    PERFECTING.
9               WALKING IS USUALLY DESCRIBED IN TERMS OF ITS
10   PHASES OF SUPPORT, SINGLE AND DOUBLE SUPPORT.    THAT IS,
11   ONE FOOT IS ALWAYS ON THE GROUND AT ALL TIMES DURING
12   WALKING.   HERE I SHOW THE GAIT CYCLE OR A STRIDE.    IT'S
13   USUALLY DEFINED AS WHEN ONE FOOT TOUCHES THE GROUND
14   UNTIL THAT FOOT ONCE AGAIN TOUCHES THE GROUND.      SO A
15   STRIDE IS MADE UP OF TWO STEPS.
16              HERE WE SEE A VIEW OF A PERSON WALKING
17   WITHOUT ARMS IN THE SAGITTAL PLANE.   AND MOST OF THE
18   ACTION IN WALKING OCCURS IN THIS SAGITTAL PLANE.      IF WE
19   LOOK AT THE TASKS LISTED BELOW, WEIGHT ACCEPTANCE,
20   SINGLE LIMB SUPPORT, LIMB ADVANCEMENT, WE CAN GET AN
21   IDEA OF THE FUNCTIONAL ABILITIES THAT ARE NECESSARY FOR
22   WALKING; THAT IS, THE LEG HAS TO SUPPORT THE WEIGHT
23   AND, FINALLY, SUPPORT THE ENTIRE BODY'S WEIGHT.      THEN
24   THE FOOT MUST PUSH OFF WITH SUFFICIENT FORCE TO PROPEL
25   US FORWARD.   AND FINALLY, THE LEG MUST SWING FORWARD.



                                 4
1    SO IF WE LOSE ANY OF THESE ABILITIES, WE CANNOT WALK.
2               SO LET ME GO THROUGH SOME OF THE TERMINOLOGY
3    FOR YOU FOR THESE ANGLES OF THE LOWER LIMB IN CASE
4    YOU'RE NOT FAMILIAR.   AT THE HIP WE HAVE EXTENSION AND
5    FLEXION, HIP FLEXION BEING LIFTING THE THIGH UP TOWARDS
6    THE FORETOE.   HIP EXTENSION, THE LEG MOVING BEHIND THE
7    BODY.   KNEE FLEXION, BENDING THE KNEE; EXTENSION,
8    STRAIGHTENING THE KNEE IN THE OPPOSITE DIRECTION.    AND
9    FINALLY, AT THE ANKLE, WE HAVE DORSI AND PLANTAR
10   FLEXION.   DORSIFLEXION BEING THE TOP OF THE FOOT COMING
11   CLOSER TO THE LOWER LEG.   PLANTAR FLEXION, THE OPPOSITE
12   DIRECTION, POINTING THE TOES.
13              SO NOW WE HAVE THIS UNDER OUR BELT, LET'S
14   LOOK AT WHAT GOES ON DURING A WALKING GAIT CYCLE.    ON
15   THE LEFT-HAND COLUMN WE HAVE THESE JOINT ANGLES.     THE
16   TOP CURVE IS FOR THE HIP, THE KNEE, AND AT THE BOTTOM
17   THE ANKLE.   NEXT TO THIS ARE MOMENTS ASSOCIATED WITH
18   THOSE JOINTS IN NEWTON METERS PER KILOGRAM, SO THEY
19   HAVE BEEN NORMALIZED BY A PERSON'S MASS.   TOP, AGAIN,
20   THE HIP, KNEE, AND FINALLY THE ANKLE.
21              NOW, IF WE BEGIN OUR GAIT CYCLE, LET'S SAY,
22   WITH RIGHT FOOT CONTACT TO THE GROUND, WE SEE THAT THE
23   FOOT IS IN MORE OR LESS A NEUTRAL POSITION.   THE KNEE
24   IS ALMOST COMPLETELY STRAIGHT, AND WE'RE FLEXED
25   SLIGHTLY AT THE HIP.   WE THEN GO THROUGH THE LEFT FOOT



                                 5
1    PUSHES OFF, AND NOW WE ENTER A PHASE OF SINGLE SUPPORT.
2    NOW, IF WE LOOK OVER HERE AT THE MOMENTS OF THE HIP AND
3    KNEE, WE SEE THEY REACH THEIR LARGEST VALUES HERE IN
4    SINGLE SUPPORT, SUGGESTING THAT, INDEED, THEY MOSTLY
5    SERVE A SUPPORT FUNCTION.   WE SWING THROUGH AND SINGLE
6    SUPPORT ENDS.   WE SEE, DURING THIS PERIOD OF TIME, THE
7    ANKLE WAS DORSIFLECTING, AND THEN, FINALLY, WE PUSH OFF
8    WITH THE RIGHT FOOT.   AND NOW NOTICE THAT THE ANKLE HAS
9    ITS PEAK MOMENT RIGHT HERE BEFORE PUSH-OFF, A VERY
10   IMPORTANT POINT.
11             THE IMPORTANCE OF THE ANKLE HERE IS
12   HIGHLIGHTED IN THIS PLOT, I THINK, OF POWERS.   AGAIN,
13   THE POWER HAS BEEN NORMALIZED TO THE PERSON'S MASS.    AT
14   THE TOP WE HAVE THE HIP AND THEN WE HAVE THE KNEE AND
15   FINALLY THE ANKLE.   THE SOLID LINES ARE THE MEANS,
16   DOTTED LINES ARE THE STANDARD DEVIATIONS OF THE MEANS.
17   AND I DON'T THINK IT WOULD BE HARD TO CONVINCE YOU THAT
18   MOST OF THE POWER DURING WALKING IS GENERATED AT THE
19   ANKLE JOINT.
20             THERE'S ANOTHER VERY INTERESTING FEATURE OF
21   THIS GRAPH, AND THAT IS THESE NEGATIVE OR DISSIPATIVE
22   POWERS.   AND SOME RESEARCHERS HAVE TAKEN THE APPROACH
23   TO TRY AND CAPTURE THIS DISSIPATED ENERGY, THIS ENERGY
24   WHICH NORMALLY WOULD BE DISSIPATED, AND USING THAT TO
25   POWER ORTHOTIC DEVICES.



                                 6
1               NOW, WALKING IS OUR MOST ECONOMICAL FORM OF
2    LOCOMOTION.    HERE WE HAVE A PLOT OF THE COST OF
3    TRANSPORT IN JOULES PER KILOGRAM PER METER FOR WALKING
4    AND RUNNING.   NOW, SOME OF YOU MAY KNOW THAT IF YOU
5    WALK FAST ENOUGH, YOU CAN ACTUALLY WALK AT A SPEED
6    WHERE IT'S MORE METABOLICALLY COSTLY THAN RUNNING, BUT
7    MOSTLY IT IS MORE ECONOMICAL.       IN FACT, MOST OF US WALK
8    AT A SPEED AT AROUND 1.3 TO 1.5 METERS PER SECOND,
9    WHICH IS, INDEED, OUR METABOLIC MINIMUM.       SO OVER TIME
10   THROUGH EVOLUTION, WE'VE BEEN ABLE TO SOMEHOW MINIMIZE
11   OUR ENERGY CONSUMPTION DURING WALKING AND CAN IDENTIFY
12   THE SPEED AT WHICH THIS OCCURS.
13              INTERESTINGLY ENOUGH, PEOPLE WITH
14   DISABILITIES SEEM TO MAINTAIN THIS ABILITY.       SO A CHILD
15   WITH CEREBRAL PALSY MAY WALK SLOWER AND USE TWICE AS
16   MUCH ENERGY TO WALK, BUT THEIR FREELY CHOSEN SPEED WILL
17   OCCUR AT THEIR METABOLIC MINIMAL COST.
18              SO WE USE ABOUT 200 WATTS OF METABOLIC POWER
19   TO WALK, AND THIS RESULTS IN ABOUT 55 WATTS OF
20   MECHANICAL POWER DELIVERED.       UNFORTUNATELY THIS POWER
21   ISN'T DELIVERED CONTINUOUSLY, BUT AS I'VE SHOWN YOU IN
22   SPURTS.   HERE ON THE RIGHT I SHOW A PLOT OF ANKLE
23   POWER.    THIS TIME ANKLE POWER FOR SLOW, MEDIUM, AND
24   FAST SPEEDS, AND I'VE PLOTTED THE PEAK POWER DEMANDS AT
25   THESE VARIOUS SPEEDS, AND WE SEE THAT IT REQUIRES FOUR



                                   7
1    TIMES -- THE PEAK POWER DEMANDS ABOUT FOUR TIMES THE
2    CONTINUOUS POWER LEVELS.
3                NOW, I SAID THAT WE'RE EFFICIENT, AND ONE OF
4    THE THINGS THAT WE DO TO BE EFFICIENT IS THIS CLEVER
5    WAY TO TRANSFER BETWEEN KINETIC AND POTENTIAL ENERGY.
6    WHEN WE WALK, OUR CENTER OF MASS GOES UP AND DOWN, AND
7    WE SPEED UP AND SLOW DOWN AROUND OUR MEAN SPEED.    OKAY.
8    HERE YOU SEE THAT WE HAVE THE POTENTIAL ENERGY, A SOLID
9    LINE, AND THE KINETIC ENERGY IN A DOTTED LINE.    NOW, IF
10   THIS WAS AN IDEAL SITUATION WHERE WE COULD RECOVER A
11   HUNDRED PERCENT OF THE ENERGY, THE MAGNITUDES OF THESE
12   TWO CURVES WOULD BE EQUAL AND THEY'D BE 180 DEGREES OUT
13   OF PHASE.    THAT IS TO SAY, THE PEAK OF ONE WOULD LINE
14   UP WITH THE MINIMUM OF ANOTHER.
15               BUT WE'RE PRETTY GOOD, AND WE CAN RECOVER
16   ABOUT 70 PERCENT OF THE ENERGY DURING WALKING AT OUR
17   PREFERRED WALKING SPEED.    UNFORTUNATELY, THIS MECHANISM
18   TENDS TO BREAK DOWN WITH PEOPLE WITH WALKING
19   DISABILITIES.    AND HERE'S A PLOT, SAME TWO CURVES,
20   POTENTIAL AND KINETIC ENERGY, FOR A CHILD WITH CEREBRAL
21   PALSY, AND YOU SEE THAT THE AMPLITUDES ARE NO LONGER
22   NEAR EQUAL, AND THE PHASING IS ALSO OFF.    AND SO THE
23   RECOVERY IS MUCH, MUCH LESS, ONE OF THE FACTORS THAT
24   CONTRIBUTES TO INCREASED METABOLIC COST OF WALKING WITH
25   PEOPLE WITH DISABILITIES.



                                  8
1               SO WHAT DO WE DO TO STUDY MOVEMENT THESE DAYS
2    OR WALKING IN PARTICULAR?    WELL, ONE WE CAN FILM PEOPLE
3    WALKING.    WE PUT REFLECTIVE MARKERS ON LANDMARKS TO THE
4    BODY.   WE HAVE THEM WALK OVER FORCE PLATES, WE FILM
5    THEM, WE APPLY A MODEL, AND WE CAN DETERMINE THESE
6    JOINT ANGLES IN JOINT MOMENTS.    WE CAN MEASURE THE
7    ELECTRICAL ACTIVITY OF THE MUSCLES, AND WE CAN CAPTURE
8    THE AMOUNT OF OXYGEN THEY USE AND DETERMINE THE
9    METABOLIC COST OF WALKING.    WE CAN EVEN GO A BIT
10   FURTHER AND, WITH SOME ASSUMPTIONS, EVEN INFER,
11   APPLYING MODELS, OF COURSE, WHAT THE FORCES ARE ON
12   INDIVIDUAL MUSCLES.    SO WE CAN GET A VERY GOOD IDEA OF
13   THE GROSS MOVEMENT DURING WALKING.
14              SO LET'S MOVE ON TO THE DEVICES THAT ARE
15   AVAILABLE TODAY.    I THOUGHT IT WOULD BE NICE TO GO ON
16   THE INTERNET AND FIND A PICTURE OF AN OLD STEEL AND
17   LEATHER DEVICE THAT PEOPLE USED IN THE '30S OR '40S
18   OR '50S, BUT I COULDN'T FIND ONE, BUT I DID FIND THIS
19   DEVICE.    YOU CAN BUY THIS TODAY, A STEEL AND LEATHER
20   DEVICE BUILT RIGHT INTO A SHOE.    AND, IN FACT, THIS
21   LEADS US TO ONE OF TWO CONCLUSIONS.    ONE, EITHER THESE
22   DEVICES ARE SO EFFECTIVE THAT THEY DON'T NEED TO BE
23   IMPROVED, OR TECHNOLOGY HAS REALLY NOT PENETRATED THIS
24   FIELD VERY MUCH YET.
25              NOW, THE PEOPLE THAT USE THESE DEVICES HAVE



                                  9
1    BOTH DEFICITS IN STRENGTH AS WELL AS MOTOR CONTROL AND
2    ALSO CAN HAVE CONDITIONS SUCH AS SPASTIC MUSCLES, A
3    VELOCITY-DEPENDENT CONDITION WHICH CAN LIMIT THE SPEED
4    AT WHICH THEIR JOINTS CAN MOVE, AND THEY CAN ALSO HAVE
5    REDUCED RANGE OF MOTION.    THIS POPULATION INCLUDES
6    PEOPLE WHO HAVE HAD HEAD INJURY, STROKE, CEREBRAL
7    PALSY, SPINAL CORD INJURY, AND MULTIPLE SCLEROSIS, TO
8    NAME SOME OF THEM.
9                SO HOW CAN A PASSIVE DEVICE HELP US IMPROVE
10   FUNCTION?    WELL, FIRST IT CAN HELP BY CONTROLLING THE
11   POSITION OF THE JOINT.    SUPPOSE SOMEBODY HAS A
12   CONDITION CALLED A DROP-FOOT WHICH IS A COMMON SIDE
13   EFFECT OF STROKES; THAT IS, THAT THEY CANNOT DORSIFLEX
14   THEIR FOOT.    THEY CANNOT LIFT THE TOES UP, SO WHEN THEY
15   GET TO THE SWING PHASE OF WALKING, THIS FOOT DRAGS AND
16   IS A PROBLEM.    HOWEVER, WE CAN PUT A BRACE ON THIS
17   ANKLE, HOLD IT AT A 90-DEGREE ANGLE, AND THIS WILL HELP
18   THE FOOT SWING THROUGH.    THIS WAY CONTROLLING POSITION
19   CAN HELP WALKING.    AND HOLDING IT IN AN ANATOMICALLY
20   CORRECT POSITION SUCH AS THIS, WE CAN ALSO HELP PREVENT
21   THE FORMATION OF DEFORMITIES OVER TIME.    FINALLY, THEY
22   CAN COMPENSATE FOR WEAKNESS.
23               SO SUPPOSE SOMEBODY DOES NOT HAVE SUFFICIENT
24   STRENGTH TO SUPPORT THEMSELVES DURING THAT SINGLE
25   SUPPORT PHASE OF WALKING.    WELL, THE DEVICE CAN LOCK



                                 10
1    THE JOINTS AND IT TAKE CAN THE WEIGHT AND ALLOW
2    WALKING.
3                SO WHAT WE HAVE TODAY AVAILABLE ARE ANKLE
4    FOOT ORTHOSES, KNEE ANKLE ORTHOSES, AND HIP KNEE ANKLE
5    FOOT ORTHOSES.    SO LET'S START WITH ANKLE FOOT ORTHOSES
6    OR AFO'S.    NOW, MOST OF THESE TODAY ARE MADE OUT OF
7    POLYPROPYLENE PLASTIC.    IT'S A VERY NICE MATERIAL
8    BECAUSE IT'S VERY EASY TO FORM.    TYPICALLY THESE
9    DEVICES ARE CUSTOMIZED TO THE INDIVIDUAL.    MOLDS ARE
10   MADE OF THEIR LOWER LEGS AND FOOT, AND THEN THIS
11   MATERIAL CAN BE HEATED UP AND FORMED TO THIS MOLD TO
12   CUSTOM FIT THE INDIVIDUAL.
13               NOW, THE MAIN DIFFERENCES HERE IN THESE
14   DEVICES ARE THE AMOUNT OF PLASTIC AND WHERE IT IS.
15   HERE FOR THE UPPER LEFT-HAND CORNER WE SEE THIS
16   COLORFUL AFO, WHICH HAS A LOT OF PLASTIC AROUND THE
17   FOOT AND ANKLE.    THIS PROVIDES -- SO THIS WOULD BE FOR
18   A PERSON WHO NEEDS SUPPORT FOR THEIR FOOT AND ANKLE.
19   THE CENTER AFO HERE WOULD BE CALLED A POSTERIOR LEAF
20   SPRING AFO.
21               THE CONCEPT BEHIND THIS IS THAT DURING THAT
22   STANCE PHASE OF WALKING WHERE THE ANKLE IS
23   DORSIFLEXION, THE LOWER LEG MOVING OVER THE TOES, THAT
24   THIS UPPER PART WOULD STORE THE ENERGY AND THEN RETURN
25   IT DURING PUSH-OFF.    UNFORTUNATELY, THERE'S A PROBLEM,



                                  11
1    AND THIS KIND OF SHOWS THE IMPORTANCE OF HAVING
2    BIOFIDELITY.   IN ORDER TO GET SUFFICIENT ENERGY RETURN,
3    YOU HAVE TO MAKE THIS UPRIGHT SO STIFF THAT IT IMPEDES
4    THE LOWER LEG MOVING FORWARD OVER THE TOES.    AND THAT
5    INTERRUPTS THE GAIT CYCLE.    SO, IN FACT, THESE DEVICES
6    HAVE NOT DONE A GOOD JOB AT HELPING WALKING.
7             WE CAN ALSO MAKE AFO'S OUT OF CARBON FIBER
8    MATERIALS.   CARBON FIBER IS NOT AS EASILY FORMED, SO
9    THEY TEND TO BE USED IN APPLICATIONS WHERE THEY DO NOT
10   NEED TO BE CUSTOM FITTED TO AN INDIVIDUAL.
11            AND, FINALLY, WE CAN PUT AN ANKLE JOINT IN
12   HERE, A HINGE, AND THEN ALLOW SOME SET RANGE OF MOTION,
13   ADD IN OUR STEEL AND LEATHER DEVICE, AND THIS IS WHAT'S
14   AVAILABLE AT THE ANKLE JOINT THESE DAYS.
15            IF WE GO UP ONE JOINT TO KNEE ANKLE FOOT
16   ORTHOSES, WE HAVE THESE DEVICES.    NOW, THESE DEVICES
17   ARE FOR PEOPLE WHO DON'T HAVE SUFFICIENT STRENGTH
18   AROUND THE KNEE TO SUPPORT THEIR WEIGHT DURING WALKING,
19   SO THESE DEVICES HAVE HINGES WHICH WILL LOCK DURING
20   SINGLE SUPPORT AND THEN ALLOW A PERSON TO WALK.    ON THE
21   FAR LEFT I HAVE, INDEED, A LEATHER AND METAL DEVICE
22   BUILT INTO A SHOE.   THEN WE HAVE NEXT TO THAT A METAL
23   AND PLASTIC DEVICE, AND FINALLY TO THE RIGHT OF THAT WE
24   SEE A POLYPROPYLENE VERSION.
25            NOW, ON THE POLYPROPYLENE VERSION YOU CAN SEE



                                  12
1    THE HINGE VERY WELL, AND YOU ALSO SEE THE MECHANISM
2    THERE THAT ALLOWS A PERSON TO LOCK OR UNLOCK THAT KNEE
3    JOINT.
4                NOW, UNFORTUNATELY, RESEARCH HAS SHOWN THAT
5    60 TO 80 PERCENT OF THE PEOPLE THAT RECEIVE THESE
6    DEVICES REJECT THEM VERY QUICKLY.     SO I THINK JUST BY
7    THAT ITSELF, IT SHOWS THAT THESE DEVICES PERHAPS COULD
8    BE GREATLY IMPROVED.     THE REASON FOR THIS REJECTION IS
9    USUALLY INCREASED METABOLIC COST.     THE WEIGHT OF THE
10   DEVICE PLUS HAVING TO WALK WITH A LOCKED KNEE GREATLY
11   INCREASES THE METABOLIC COST OF WALKING, AND PEOPLE OPT
12   FOR A WHEELCHAIR INSTEAD OF THESE DEVICES.
13               FINALLY, WE HAVE A HIP KNEE ANKLE FOOT
14   ORTHOSES.    ON THE LEFT YOU SEE A DEVICE ON A PERSON'S
15   ONE LEG AND TO THE RIGHT A BILATERAL PAIR OF THESE
16   DEVICES.    NOW, IF YOU USE THESE DEVICES, YOU DO NOT
17   WALK VERY WELL AND PROBABLY YOU USE CRUTCHES AND/OR A
18   WALKER AS WELL.     IN THIS CASE DEVICES ARE PERHAPS MORE
19   IMPORTANT FOR THE HEALTH BENEFITS OF BEING UPRIGHT AND
20   VERTICAL BECAUSE THE PERSON REALLY IS NOT VERY MOBILE
21   IN SUCH A DEVICE.
22               OKAY.   WELL, I SAID MAYBE THESE DEVICES WERE
23   EFFECTIVE AS IN THE OLD DESIGN.     I DON'T KNOW IF ANYONE
24   BOUGHT THAT.    BUT LET'S LOOK AT AFO'S.   AFO'S ARE A
25   GOOD DEVICE TO LOOK AT THE EFFECTIVENESS BECAUSE PEOPLE



                                  13
1    WHO USE AFO'S TYPICALLY CAN WALK WITHOUT THE DEVICES,
2    AND SO WE CAN ACTUALLY MEASURE THE PERFORMANCE
3    IMPROVEMENTS WITH THE DEVICES.      AND WE'RE NOT SEEING
4    VERY GOOD RESULTS.      BASICALLY IN MOST METRICS, WHETHER
5    THEY BE METABOLIC COST, STRIDE LENGTH, WALKING SPEED,
6    THERE'S NO OR A SMALL IMPROVEMENT.         AND THIS IS TOO BAD
7    BECAUSE THESE PEOPLE NEED BIG IMPROVEMENTS.         SO THEY'RE
8    JUST NOT DOING A GOOD JOB HERE, I DON'T THINK.
9               BUT ARE THERE MANY OF THESE DEVICES?        I
10   CONTACTED TWO COMPANIES WHICH MAKE THESE DEVICES, AND
11   BETWEEN THEM THEY SELL MORE THAN 450,000 OF THESE A
12   YEAR.    SO THERE'S QUITE A NEED, I THINK, FOR THESE
13   TYPES OF DEVICES.
14              OKAY.    WELL, NOW, LET'S LOOK AT SOME
15   INTERESTING RESEARCH THAT PEOPLE ARE DOING DEVELOPING
16   NEW ACTIVE DEVICES.      WE'LL START OUT WITH THE AFO'S
17   AGAIN.   AND THIS IS AN AFO DEVELOPED FOR THIS CONDITION
18   OF DROP-FOOT.      SO THE ACTUATOR YOU SEE THERE WOULD PUSH
19   DOWN ON THE HEEL AND LIFT THE TOES UP DURING SWING
20   PHASE.   NOW, I'VE PUT IN RED HERE THE BAD NEWS.           THIS
21   DEVICE WEIGHS 2.6 KILOGRAMS.       OKAY.    I DON'T KNOW.
22   IT'S NOT MY FIRST THOUGHT TO HELP SOMEBODY WALK, YOU'D
23   PUT 2.6 KILOGRAMS ON THEIR FOOT AND LOWER LEG.
24              THE RULE OF THUMB IS THAT FOR EVERY KILOGRAM
25   THAT YOU ADD TO AN ANKLE, YOU INCREASE THE METABOLIC



                                   14
1    COST ABOUT 4 PERCENT.    SO THIS DEVICE WOULD INCREASE
2    METABOLIC COST ABOUT 10 PERCENT.    IF AN INDIVIDUAL USED
3    TWO OF THESE, IT WOULD BE 20 PERCENT.    SO YOU CAN SEE
4    THAT YOU'VE DUG YOURSELF QUITE A HOLE AND THIS IS
5    CERTAINLY ONE OF THE PROBLEMS, ONE OF THE REASONS WHY
6    IN THE PAST PEOPLE HAVE NOT DEVELOPED ACTIVE DEVICES,
7    BECAUSE THE ACTUATORS ARE VERY HEAVY.
8                A NICE PART ABOUT THIS RESEARCH WAS THE
9    ADAPTIVE CONTROLLER.    SO YOU SEE IT HAS AN ANKLE ANGLE
10   SENSOR, IT HAS FOUR SENSORS, AND SO IT DETERMINES WHEN
11   YOU'RE IN SWING PHASE, AND IT CONTROLS ITSELF.
12               HERE'S A SIMILAR DEVICE IN THAT IT USES A
13   LINEAR ACTUATOR AS WELL.    THIS DEVICE SHOWN HERE IS THE
14   ROBOKNEE.    THEY ALSO HAVE A DEVICE THEY CALL THE
15   ROBOANKLE.    AGAIN, IT WEIGHS A GOOD DEAL.   IN THIS CASE
16   THEY WALK AROUND WITH THE DEVICE, AND YOU SEE THEY USE
17   4 KILOGRAMS OF BATTERIES THAT LAST LESS THAN TWO HOURS
18   IN HEAVY USE.    SO IF YOU NEEDED TWO OF THESE, THEY'D
19   LAST LESS THAN AN HOUR.    SO THIS HIGHLIGHTS ANOTHER
20   PROBLEM; THAT IS, THE MASS OF THE POWER SUPPLIES.
21   AGAIN, IT HAS A VERY NICE CONTROL ALGORITHM WHICH,
22   AGAIN, CONTROLS THE DEVICE, SO THE PERSON JUST USES IT
23   QUITE INTUITIVELY.
24               SO ONE WAY TO GET AROUND THE WEIGHT OF THESE
25   ACTUATORS AND POWER SUPPLIES IS TO NOT HAVE THEM.



                                 15
1    HERE'S A DEVICE THAT'S DEVELOPED THE IDEA THAT IT COULD
2    RECOVER THIS DISSIPATIVE ENERGY THAT I SHOWED YOU
3    EARLIER, STORE THIS ENERGY, AND THEN RETURN IT DURING A
4    DIFFERENT POINT IN THE GAIT CYCLE.    THIS IS AN EARLY
5    PROOF OF CONCEPT PAIR OF DEVICES, AND NOW THEY'RE BEING
6    REFINED TO REDUCE THEIR WEIGHT AND MAKE THEM MORE
7    ERGONOMIC AND BIOFIDELIC.
8             YOU CAN ALSO COMBINE FUNCTIONAL ELECTRICAL
9    STIMULATION WITH AN ORTHOSIS.    YOU CAN HAVE THE
10   ADVANTAGES OF THE PASSIVE DEVICE; THAT IS, IT CAN
11   SUPPORT A JOINT, IT CAN CONTROL THE RANGE OF MOTION,
12   AND THEN HAVE THE INCREASED FORCE GENERATION THAT THE
13   FES PROVIDES.   THE ORTHOSIS ALSO PROVIDES A PLATFORM
14   THAT YOU CAN PUT THE ELECTRONICS ON.
15            THERE ARE EVEN SOME COMMERCIAL SYSTEMS, I
16   BELIEVE, AVAILABLE WITH FES TO HELP WALKING, NOT WIDELY
17   USED, AND THERE ARE PROBLEMS WITH BOTH CONTROLLING AND
18   THE ELECTRODE ISSUES, WHETHER THEY BE SURFACE
19   ELECTRODES OR IMPLANTED ELECTRODES.
20            ONE BIT OF RESEARCH THAT'S TRIED TO ADDRESS
21   SOME OF THESE ISSUES IS THE BION.    IT'S AN INJECTABLE
22   ELECTRODE THAT IS WIRELESS, AND THIS HAS BEEN USED, I
23   THINK, IN HELPING AT LEAST ONE PERSON WALK.    I'VE SEEN
24   A CASE STUDY.   AND SO RESEARCH IS ONGOING WITH THIS
25   DEVICE WHICH MIGHT HELP IN THE AREA OF FES.



                                 16
1                NOW WE MOVE ON TO SOME BIG DEVICES, I GUESS.
2    THIS IS THE BERKELEY LOWER EXTREMITY EXOSKELETON
3    DEVELOPED BY DR. KAZEROONI BERKELEY.    NOW, I SHOW A
4    PICTURE OF THE DEVICE IN ITS EARLIER FORM.    THERE'S NOW
5    WHAT THEY CALL A BLEEX 2, AND THAT'S THE PROPERTIES I
6    SHOW HERE.    NOW, THIS DEVICE CAN CARRY A PAYLOAD OF 150
7    POUNDS.   IT'S BEEN DEVELOPED FOR THE MILITARY.   OKAY.
8    IT'S HYDRAULICALLY DRIVEN, ELECTRICALLY POWERED; AND IN
9    ORDER TO INCREASE BATTERY TIME BETWEEN CHARGES, IT
10   SCAVENGES ENERGY.    SO WHENEVER THERE'S A DISSIPATIVE
11   MOVEMENT, IT PUMPS HYDRAULIC FLUID BACK AND GENERATES
12   ELECTRICITY.    HAS A UNIQUE CONTROL STRATEGY.   THE
13   SYSTEM IS DESIGNED TO BE UNSTABLE, AND THEN IT'S
14   SOMEHOW ATTRACTED TO THE MOVEMENT OF THE PERSON, AND IT
15   WORKS QUITE WELL.    I'VE SEEN VIDEOS, AND IT'S TRULY AN
16   IMPRESSIVE DEVICE.
17               HERE'S ANOTHER DEVICE DEVELOPED FOR THE
18   MILITARY.    THIS IS A FULL-BODY EXOSKELETON DEVELOPED BY
19   STEPHEN JACOBSEN AT UTAH AT THE SARCOS RESEARCH
20   CORPORATION.    THIS IS ALSO HYDRAULICALLY POWERED, BUT
21   ITS POWER SUPPLY IS AN INTERNAL COMBUSTION ENGINE.      AND
22   IT HAS A DIFFERENT CONTROL STRATEGY.    THEY CALL IT A
23   GET OUT OF THE WAY CONTROL STRATEGY.    THERE ARE FOUR
24   SENSORS PLACED BETWEEN THE INDIVIDUAL USER AND THE
25   DEVICE.   AND THEN IF A FORCE IS APPLIED ON ONE OF THESE



                                 17
1    SENSORS, THE DEVICE MOVES IN A WAY TO MINIMIZE THAT
2    FORCE.   SO THEY SAY IT GETS OUT OF THE WAY; AND, OF
3    COURSE, ITS JOB IS INDEED TO FOLLOW WHAT THE USER WANTS
4    TO DO.
5               AND ONE LAST FULL-BODY EXOSKELETON, THIS ONE
6    FROM JAPAN.   IT'S CALLED THE HAL-5.   IT HAS ELECTRIC
7    MOTORS, AND IT REALLY HAS TWO CONTROL SYSTEMS.     IT'S
8    CONTROLLED VIA EMG SIGNALS FOR A HEALTHY INDIVIDUAL;
9    AND THEY SAY IF SOMEBODY IS DISABLED, THEY CAN TRAIN IT
10   TO WORK.   THEY LIST A TWO-HOUR -- I MEAN A TWO-MONTH
11   TRAINING PERIOD, AND I DON'T KNOW IF ANY DISABLED
12   INDIVIDUALS HAVE GONE THROUGH THIS TO ACTUALLY USE IT.
13   I DON'T KNOW THE LIFETIME OF THE BATTERIES THERE.     BUT
14   AN IMPORTANT POINT IN ALL THESE EXOSKELETONS IS THEY
15   HAVE THE ADVANTAGE, WHEN YOU BUILD A DEVICE LIKE THIS,
16   THAT THEY CAN SUPPORT THEIR OWN WEIGHT.    SO IF I HAVE
17   THAT AFO, I HAVE TO SWING THAT MASS THROUGH THE AIR AND
18   SUPPLY MY OWN ENERGY TO DO THAT; BUT THESE DEVICES, IF
19   I LIFT UP ONE LEG, THE OTHER LEG IS STILL ON THE GROUND
20   AND IT CAN MOVE ITSELF.   SO THIS IS AN ADVANTAGE OF
21   THIS EXOSKELETAL FULL-BODY KIND OF APPROACH.
22              SO WHAT DO WE NEED TO DO TO HELP INDIVIDUALS
23   WITH WALKING DISABILITIES?    I THINK IF WE WANT TO
24   IMPROVE WALKING, WE HAVE TO DECIDE UPON A METRIC OF
25   WALKING.   THAT IS, IS IT INCREASED STEP LENGTH?    IS IT



                                  18
1    INCREASED WALKING SPEED?    WHAT IS OUR METRIC?   AND I
2    PUT FORWARD THAT A GOOD METRIC WOULD BE TO MINIMIZE THE
3    METABOLIC COST OF WALKING FOR TWO REASONS.    ONE, THE
4    METABOLIC COST IS A NICE COLLECTIVE VARIABLE.     WHATEVER
5    YOU DO WILL BE REFLECTED IN THIS.    AND SECONDLY, PEOPLE
6    WITH WALKING DISABILITIES TYPICALLY HAVE VERY HIGH
7    METABOLIC COST.
8                FOR EXAMPLE, CHILDREN WITH CEREBRAL PALSY
9    TYPICALLY USE TWO TO THREE TIMES AS MUCH ENERGY TO WALK
10   WHILE WALKING SLOWER THAN TYPICALLY DEVELOPING
11   CHILDREN.    INDIVIDUALS WHO HAVE HAD STROKE OFTEN USE
12   TWICE AS MUCH ENERGY AS THEIR HEALTHY COHORTS.     IN
13   COMPARISON WITH AMPUTEES, THEY TYPICALLY HAVE ONLY A
14   10- TO 30-PERCENT INCREASE IN METABOLIC COST.     SO THERE
15   SEEMS TO BE SOMETHING WITH LACK OF MOTOR CONTROL WHICH
16   INCREASES THE METABOLIC COST OF WALKING.
17               IN ADDITION, WE WANT TO DEVELOP DEVICES WHICH
18   MAXIMIZE THE MOVEMENT POSSIBILITIES.    WE WOULD LIKE
19   SOMEBODY TO BE ABLE TO WALK SLOW, FAST, HELP THEM WALK
20   FORWARDS, BACKWARDS, SIDEWAYS, AND WE'D LIKE TO DO THIS
21   IN THE MAXIMUM POSSIBLE ENVIRONMENT; THAT IS, UPHILL,
22   DOWNHILL, UPSTAIRS, DOWNSTAIRS, THESE TYPES OF THINGS.
23   I THINK ALL OF US HAVE BEEN THROUGH SIMILAR SITUATIONS
24   PERHAPS WHERE WE TRIED TO WALK DOWNSTAIRS IN A HEAVY,
25   LIKE, A SKI BOOT KIND OF THING.    YOU KNOW THE



                                 19
1    DIFFICULTY WHEN YOU CAN'T POINT YOUR TOES IN WALKING
2    DOWNSTAIRS.   NOW, IMAGINE YOU ALSO CAN'T BEND YOUR KNEE
3    WEARING A KAFO.   WELL, YOU CAN SEE THAT WALKING
4    DOWNSTAIRS CAN BE VERY UNCOMFORTABLE.    SO WE WANT TO
5    HAVE DEVICES WHICH CAN ENABLE PEOPLE TO WALK IN
6    ENVIRONMENTS THAT THEY WILL ENCOUNTER IN DAILY LIFE.
7              WE ALSO NEED TO DEVELOP IMPROVED MODELING
8    CAPABILITY.   WHAT WE NEED ARE PATIENT-SPECIFIC,
9    PHYSICS-BASED FORWARD DYNAMICS MODELS.    FOR THESE TYPES
10   OF MODELS, WE WOULD MEASURE THE GAIT PATTERN OF AN
11   INDIVIDUAL WALKING AND CREATE A MODEL OF THIS WITH THIS
12   INPUT.   IF THE MODEL WALKS LIKE THE PERSON, THEN WE'D
13   SAY WE HAVE A GOOD MODEL.    NOW WHAT WE'D LIKE TO DO IS
14   SAY, WELL, WHAT WOULD HAPPEN WITH THEIR GAIT IF WE
15   INCREASE THEIR PLANTAR FLEXION MOMENT AT PUSH-OFF BY 15
16   PERCENT OR CHANGE THE ANGLE OF FOOT CONTACT?
17             IN ORDER TO DO THIS, WE HAVE TO HAVE THESE
18   FORWARD-BASED MODELS.   CURRENTLY SUCH MODELS ARE VERY
19   COMPUTATIONALLY EXPENSIVE.    SO WE NEED TO EITHER
20   DEVELOP FASTER COMPUTERS, WHICH IS OBVIOUSLY GOING ON,
21   AND WE ALSO NEED TO DEVELOP MORE REFINED TECHNIQUES TO
22   MAKE THIS A MORE EFFICIENT PROCESS SO THAT WE CAN
23   DEVELOP THESE MODELS JUST FOR A SPECIFIC PERSON BECAUSE
24   IT'S VERY HARD TO DEVELOP A DEVICE TO HELP A SPECIFIC
25   PERSON IF YOU DON'T KNOW EXACTLY WHAT YOU NEED.      AND I



                                  20
1    THINK THE WAY TO FIND THIS IS THROUGH MODELING.
2               NOW, IF WE WANT TO BUILD A DEVICE, WE'RE
3    GOING TO NEED A STRUCTURE, A POWER SUPPLY, WE HAVE TO
4    KNOW WHAT'S GOING ON, HAVE TO CONTROL EVERYTHING, AND,
5    FINALLY, YOU WANT TO SUPPLY POWER TO THE JOINTS.
6               THE STRUCTURE HAS TO TRANSMIT THE FORCES AND
7    MOMENTS TO BETWEEN LIMB SEGMENTS AND BETWEEN THE USER
8    AND THE GROUND.   IF IT GOES UNDER THE FOOT, IT HAS TO
9    BE STRONG ENOUGH TO BE WALKED UPON.      I DON'T THINK THIS
10   IS A LIMITING FACTOR SO MUCH IN DEVELOPMENT OF THESE
11   DEVICES.   WE CAN HAVE MATERIALS WHICH ARE, INDEED,
12   STRONG ENOUGH AND DURABLE ENOUGH AND LIGHT ENOUGH, I
13   THINK, TO DO THE JOB.   SO I DON'T THINK THIS IS A
14   CRITICAL FUNCTION, BUT WE DO HAVE POWER SUPPLY ISSUES.
15   IF YOU'RE WILLING TO WEAR A MINIATURE INTERNAL
16   COMBUSTION ENGINE AROUND, THEN YOU'RE IN GOOD SHAPE.
17   THE POWER DENSITY OF HYDROCARBON FUELS IS VERY HIGH,
18   AND YOU CAN EASILY WALK AROUND ALL DAY.      IF YOU NEED
19   BATTERIES, WELL, YOU SAW THAT THEY'RE HEAVY AND DON'T
20   LAST VERY LONG, MOST CASES.      AGAIN, YOU'D WANT TO MAKE
21   A DEVICE AS EFFICIENT AS POSSIBLE AND PERHAPS HAVE SOME
22   ENERGY REGENERATION.    FUEL CELLS AREN'T THERE YET.
23   RECOVERED ENERGY IS A POSSIBILITY, BUT CERTAINLY WOULD
24   NOT BE SUFFICIENT FOR MANY APPLICATIONS.
25              WE ALSO NEED TO SENSE WHAT'S GOING ON.     WE



                                 21
1    HAVE FORCE/PRESSURE SENSORS.      WE CAN MEASURE ANGULAR
2    POSITION, THE MUSCLE ELECTRICAL ACTIVITY, IF THAT'S
3    IMPORTANT.   WE CAN PUT ACCELEROMETERS ON PEOPLE TO
4    GATHER INFORMATION ABOUT MOVEMENT.      AGAIN, I DON'T
5    THINK THIS IS A CRUCIAL FACTOR BECAUSE I THINK WE CAN
6    GET A GOOD ENOUGH PICTURE WITH THE PRESENT DAY DEVICES.
7    IN FACT, IN WORKING IN OUR LAB WITH PEOPLE USING A
8    WALKER, WE'RE ABLE TO DETERMINE GAIT CYCLE EVENTS JUST
9    FROM THE FORCES THEY PLACE ON THE HANDLES OF THE
10   WALKERS.   SO KNOWING WHAT'S GOING ON, I THINK WE CAN
11   GET A VERY COMPLETE PICTURE OF PEOPLE WALKING THESE
12   DAYS WITH SENSORS THAT THEY CARRY WITH THEM.
13              BUT CONTROL IS A VERY IMPORTANT ISSUE.     HOW
14   ARE WE GOING TO CONTROL THESE DEVICES?      I LIST TWO
15   POSSIBLE DIRECTIONS TO GO.    THE FIRST, I CALL IT SHARED
16   CONTROL WHERE THE ALGORITHMS OF THE CONTROL SYSTEM TAKE
17   IN ALL THIS SENSORY INFORMATION AND THEN INTERPRET WHAT
18   THE PERSON IS DOING AND THEN SAYS I KNOW WHAT YOU WANT
19   TO DO IN THE NEXT FRACTION OF A SECOND AND THEN ACTS TO
20   ASSIST THAT ABILITY.   SO THIS IS ONE DIRECTION TO GO.
21              NOW, IN THIS CASE THE ALGORITHM IS LEARNED,
22   SO WOULD BE THE TYPE OF SITUATION THAT THE HAL-5 IS
23   USING.   AND SO THE ALGORITHMS -- AND THEN HOPEFULLY
24   WHEN THE PERSON WEARS THIS DEVICE, IT'S JUST NATURALLY
25   INTUITIVE.



                                  22
1               THE OTHER WAY TO GO IS WITH DIRECT CONTROL,
2    CONTROLLING A DEVICE THROUGH MYOELECTRIC OR BRAIN
3    SIGNALS.    NOW, YOU CAN'T USE -- PERHAPS WITH MANY
4    PEOPLE WITH DISABILITY, YOU CAN'T USE JUST THE
5    MYOELECTRIC SIGNALS OF THE JOINTS BECAUSE THAT'S THE
6    REASON WHY THEY HAVE A DISABILITY.    THEY DO NOT HAVE
7    GOOD MYOELECTRIC SIGNALS.    SO YOU'D HAVE TO GO
8    SOMEPLACE ELSE IN THE BODY PERHAPS TO PICK THESE UP,
9    AND, OF COURSE, THE BRAIN IS ALWAYS THERE.    THE PROBLEM
10   IS THIS IS PROBABLY FURTHER OUT; AND, AGAIN, IN THIS
11   SITUATION, THE USER HAS TO LEARN HOW TO CONTROL THE
12   DEVICE.    NOW, I'M NOT SURE WHICH WILL TURN OUT TO BE
13   BETTER, BUT THERE IS OBVIOUSLY SOME CONTRAST.      EITHER
14   THE PERSON HAS TO LEARN OR THE ALGORITHMS HAVE TO
15   LEARN.    TIME WILL TELL WHICH MIGHT BE THE BEST.
16              AND FINALLY, WE MUST HAVE SUPPLY POWER.     I
17   LISTED IN BLUE HERE THE HYDRAULICS, ELECTRIC MOTOR,
18   LINEAR ACTUATORS, DEVICES THAT YOU SAW WERE USED ON
19   SOME OF THESE DEVICES THAT ARE CURRENTLY AVAILABLE,
20   AND, INDEED, THESE CAN SUPPLY SUFFICIENT FORCE.      THE
21   PROBLEM IS THEY TEND TO BE HEAVY, AND WE NEED TO WORK
22   TO REDUCE THE WEIGHT OF THESE DEVICES.    AND, IN FACT,
23   THERE MAY BE SOME ISSUE WITH ACTUALLY RUNNING INTO THE
24   PHYSICS THAT PREVENTS US FROM MAKING THESE DEVICES
25   LIGHT ENOUGH AS WE'D LIKE.



                                  23
1                DOWN HERE ARE SOME FUTURISTIC THINGS THAT
2    MIGHT BE AVAILABLE IN THE FUTURE, ARTIFICIAL MUSCLES,
3    MEMORY METALS, RHEOLOGICAL MATERIALS, MATERIALS THAT
4    PRESENTLY CANNOT SUPPLY SUFFICIENT FORCE, BUT WITH
5    RESEARCH PERHAPS THESE DEVICES CAN, INDEED, BE
6    DEVELOPED TO DO THAT, BUT THEY ALSO HAVE TO DEVELOP THE
7    ENERGY OR THE POWER QUITE QUICKLY.    AS I SAID, THE
8    POWER IS DELIVERED TO THE JOINTS IN LITTLE PULSES ABOUT
9    A TENTH OF A SECOND, BUT THERE'S EVEN A MORE IMPORTANT
10   SPEED REQUIREMENT, AND THAT'S SAFETY.    IF WE'RE OUT
11   WALKING, WE SLIP, WE STUMBLE, WE MISSTEP, AND THESE
12   PEOPLE WHO USE DEVICES WILL HAVE THESE SAME EFFECTS,
13   AND THEY HAVE TO BE ABLE TO RECOVER FROM THIS.     SO WE
14   NEED THE SPEED OF THE POWER SUPPLY TO BE SUCH THAT THEY
15   CAN REACT QUICKLY ENOUGH TO PREVENT A FALL FROM
16   SOMEONE.
17               ON THE OTHER HAND, THESE DEVICES MUST BE VERY
18   RELIABLE.    AGAIN, IF YOU'RE OUT WALKING WITH A
19   DISABILITY, YOU DON'T WANT YOUR DEVICE TO FAIL AND END
20   UP IN A FALL.    SO THESE DEVICES NEED TO BE 99.999
21   PERCENT RELIABLE.    AND IF THEY DO FAIL, THEN WE WANT TO
22   HAVE THEM FAIL IN A WAY WHICH WILL PROTECT THE USER.
23               WELL, TO SUM THINGS UP, I THOUGHT I WOULD
24   PRESENT MY CONCEPT FOR AN ADVANCED AFO, A FUTURISTIC
25   AFO HERE.    IN THIS DESIGN I'VE COMBINED THE STRUCTURE



                                 24
1    AND ACTUATORS INTO A SINGLE MATERIAL.     SO THE FIBERS OF
2    MY SOCK WILL CHANGE THEIR PROPERTIES TO SUPPLY THE
3    REQUIRED MOMENT.    SO IF YOU NEED ADDITIONAL PUSH-OFF,
4    THEN IT WILL CREATE A PLANTAR FLEXOR MOMENT AT THIS
5    TIME.   IF YOU STEP IN A HOLE WHERE YOU MIGHT TWIST YOUR
6    ANKLE, IT WILL STIFFEN UP, REACT FAST ENOUGH TO STIFFEN
7    UP AND PROTECT YOUR ANKLE AND YOUR STABILITY.
8                IN THIS ADVANCED DEVICE, THE SENSORS, THE
9    POWER SUPPLY, THE CONTROLLERS ARE ALL EMBEDDED RIGHT IN
10   THE SOCK.    IT CAN GO EASILY MORE THAN A DAY WITHOUT
11   BEING RECHARGED.    AND IT HAS A NICE FUNCTION FOR
12   UPDATING.    EVERY MONTH OR COUPLE MONTHS, I HOOK IT UP
13   TO THE WIRELESS CONNECTION ON A COMPUTER, AND THEN GO
14   THROUGH A SERIES OF MOVEMENTS OF ACTIVITIES, AND THE
15   DEVICE WILL UPDATE ITS ALGORITHMS TO MY CURRENT LEVEL
16   OF NEED.    SO IF I'VE GAINED WEIGHT OR LOST WEIGHT,
17   PERHAPS I WENT TO THE GYM, I GOT STRONGER, OR MY
18   ILLNESS HAS PROGRESSED, AND I'VE GOTTEN WEAKER, THE
19   DEVICE UPDATES IT AND IT'S READY TO GO.
20               SO THAT'S MY FUTURISTIC DESIGN.   AND WITH
21   THAT, I'LL THANK YOU AND TAKE SOME QUESTIONS.
22               THE MODERATOR:   WELL, THANK YOU, DR. BENNETT.
23   THAT WAS A WONDERFUL PRESENTATION AND A WONDERFUL
24   OVERVIEW OF THIS VERY COMPLICATED AND RICH TOPIC.      SO
25   WE'RE GOING TO MOVE FORWARD NOW, AND WE HAVE ALREADY



                                  25
1    RECEIVED SEVERAL QUESTIONS FOR YOU.
2                I'D LIKE TO START WITH THE FIRST QUESTION.
3    YOU SPOKE ABOUT MINIMIZING THE ENERGY COST OF WALKING.
4    DOES ENERGY COST OF WALKING DEPEND ON ANY PARAMETERS
5    OTHER THAN WALKING SPEED?
6                DR. BENNETT:   YES.    WALKING SPEED ITSELF
7    DEPENDS UPON BOTH THE CADENCE AND THE STEP LENGTH.          AND
8    SO WHEN PEOPLE WALK, BOTH OF THESE PARAMETERS ARE
9    IMPORTANT.    IN FACT, IT TURNS OUT THAT PEOPLE DO HAVE,
10   INDEED, A TUNED FREQUENCY.        SO IF YOU HAVE PEOPLE WALK
11   AT THE PREFERRED WALKING SPEED, BUT IF YOU INTERRUPT
12   AND HAVE THEM WALK AT A DIFFERENT FREQUENCY, THE ENERGY
13   CONSUMPTION WILL GO UP.      SO IT'S TUNED NOT JUST TO
14   WALKING SPEED, BUT ACTUALLY TO THE FREQUENCY OF WALKING
15   AS WELL.
16               THE MODERATOR:   NEXT QUESTION.     WE HAVE ONE
17   OF THE RELATIVE BENEFITS OF EXTERNALLY POWERED
18   EXOSKELETONS VERSUS ELECTRICALLY ACTIVATING THE
19   PARALYZED MUSCLES TO DRIVE THE MOTION OF AN ADVANCED
20   ORTHOSES.
21               DR. BENNETT:   WELL, THAT'S AN INTERESTING
22   TRADE-OFF.    I MEAN THAT, AGAIN, THAT DEPENDS UPON THE
23   LEVEL, I WOULD SAY THE LEVEL OF TECHNOLOGY.        I MEAN
24   BOTH.   IF YOU CAN DO IT WITH IMPLANTED FUNCTIONAL
25   ELECTRICAL STIMULATION OR FUNCTIONAL ELECTRICAL



                                   26
1    STIMULATION ANYWAY, I THINK THAT THERE CERTAINLY CAN BE
2    A WONDERFUL SYSTEM.    THE IDEA IS ON BOTH OF THESE
3    SYSTEMS IS THE CONTROL ISSUE, I WOULD SAY, THAT WHETHER
4    IT'S BETTER TO CONTROL SIGNAL TO THE MUSCLES.     THE
5    PROBLEM WITH POWERING THE MUSCLES THEMSELVES IS YOU CAN
6    ONLY POWER A PART OF THE MUSCLE.
7              NOW, WE DON'T THINK YOU CAN PUT
8    STIMULATORS -- LIMITED NUMBER OF STIMULATORS IN A
9    MUSCLE, SO YOU'RE JUST GOING TO BE USING ONE PART OF A
10   MUSCLE.   AND SO THAT MIGHT BE PROHIBITIVE.
11             WITH THE EXOSKELETONS, AGAIN, YOU HAVE TO
12   CARRY AROUND A BULKY DEVICE.     SO THAT'S NOT A POSITIVE.
13   AND, OF COURSE, IT REQUIRES A LOT MORE POWER.     SO I
14   WOULD SAY THAT THE TRADE-OFFS ARE THERE.     BOTH OF THEM,
15   I DON'T KNOW THAT WE HAVE SUFFICIENT TECHNOLOGICAL
16   DEVELOPMENT TO SAY WHICH WOULD BE BETTER NOW.
17             THE OTHER ISSUE THAT COMES TO MIND WITH THE
18   FES, OF COURSE, IS THE LONG-TERM EFFECT OF STIMULATING
19   A MUSCLE WITH FES.    CAN WE NOT DO DAMAGE TO THE NERVE
20   ENDINGS, ETC.   SO I THINK WE JUST DON'T KNOW WHICH WILL
21   TURN OUT THE BEST.
22             THE MODERATOR:   WE HAVE ANOTHER QUESTION FROM
23   THE AUDIENCE.   HOW IMPORTANT IS SENSATION IN ACHIEVING
24   MORE NORMAL LOCOMOTOR FUNCTION?     AND DO WE KNOW THE KEY
25   PARAMETERS; I.E., JOINT ANGLE, PRESSURES, ETC., AND HOW



                                 27
1    TO FEED THEM BACK TO THE DEVICE USER?
2             DR. BENNETT:    WELL, SENSING IS VERY
3    IMPORTANT.   IF YOU CAN'T SENSE WHERE YOUR LIMB IS,
4    YOU'RE NOT GOING TO MOVE IT.    SO TWO THINGS.   ONE IS WE
5    CAN TODAY VERY EASILY GIVE SUFFICIENT INFORMATION TO
6    COMPUTER ALGORITHMS TO DESCRIBE, I THINK, THE WALKING.
7    NOW, WHETHER THIS INFORMATION, IF SOMEBODY HAS SENSORY
8    DEFICITS, WHAT INFORMATION CAN BE FED BACK TO THE
9    CENTRAL NERVOUS SYSTEM IS OBVIOUSLY THE KEY THERE.     AND
10   I HAVE TO TELL YOU THAT I JUST DON'T KNOW ENOUGH I CAN
11   SAY HOW THAT WOULD BE DONE OR IT WOULD BE WONDERFUL IF
12   WE COULD DO, BUT I DON'T KNOW HOW TO DO THAT.
13            THE MODERATOR:    NEXT QUESTION.   YOUR EXAMPLES
14   FOCUSED ON ANKLE, HIP, AND KNEE MECHANISMS.      HOW
15   IMPORTANT IS TRUNK STABILITY OR RATHER TRUNK MOBILITY
16   IN ACHIEVING STANDING AND WALKING FUNCTION?
17            DR. BENNETT:    WELL, CERTAINLY IN ORDER TO
18   WALK, YOU HAVE TO BE ABLE TO STABILIZE AND CONTROL YOUR
19   TRUNK FROM THE POINT OF -- PERSPECTIVE OF BALANCE.     SO
20   IF A PERSON LACKS THAT ABILITY, THE DEVICES SHOWN HERE,
21   EXCEPT FOR PERHAPS THE FULL-BODY EXOSKELETONS, ARE NOT
22   APPROPRIATE.   SO WE'RE ASSUMING HERE THAT A PERSON CAN,
23   INDEED, CONTROL THEIR UPPER BODY.   IT'S ESSENTIAL TO BE
24   ABLE TO DO THAT FOR BALANCE.
25            THE MODERATOR:    NEXT QUESTION.   WHAT SENSORS



                                28
1    ARE NEEDED FOR AN ADVANCED POWERED ORTHOSES TO PROVIDE
2    BOTH BALANCE AND CONTROLLED WALKING?
3              DR. BENNETT:    WELL, I THINK THE ISSUE THERE
4    IS IT DEPENDS.   YOU NEED THE SAME INFORMATION, I THINK,
5    THAT WE'VE TALKED ABOUT.     YOU NEED TO KNOW JOINT
6    ANGLES, YOU NEED TO KNOW FORCES, YOU NEED -- FOR
7    BALANCE, THE KEY THING, OF COURSE, IS TO KNOW WHERE THE
8    CENTER OF MASS IS RELATIVE TO THE FEET.     SO THIS IS
9    SOMETHING THAT, AGAIN, WE CAN MEASURE.     RIGHT NOW YOU
10   HAVE TO -- AGAIN, IF A PERSON NEEDS HELP WITH BALANCE,
11   YOU'VE ENTERED, I THINK, INTO A WHOLE OTHER REALM OF
12   DEGREE OF DIFFICULTY.     SO IT'S A VERY -- IT'S VERY
13   CHALLENGING, MUCH MORE SO THAN JUST EVEN HELPING A
14   PERSON WALK.   SO JOINT ANGLES, THE JOINT ANGLES, THE
15   FORCES INVOLVED, THE MOVEMENTS AND KNOWING WHERE THAT
16   BASE OF SUPPORT IS AND WHERE THE CENTER OF MASS IS IS A
17   CRUCIAL THING FOR BALANCE.
18             THE MODERATOR:    WE HAVE AN INTERESTING
19   QUESTION ON COSMESIS.     CAN YOU SPEAK A LITTLE BIT ABOUT
20   THAT?   IS IT TRUE THAT EXOSKELETAL SYSTEMS ARE RELIABLE
21   AND PROVIDE NEAR NORMAL FUNCTION?
22             DR. BENNETT:    NOW, WHAT WAS -- I MISSED THE
23   FIRST PART OF THE QUESTION.
24             THE MODERATOR:    IT'S A TWO-PART QUESTION FROM
25   ONE AUDIENCE MEMBER.     AND SAID THE FIRST QUESTION IS



                                  29
1    CAN YOU SPEAK A LITTLE BIT ABOUT COSMETICS?
2              DR. BENNETT:   OH, COSMETICS.   OKAY.    WELL,
3    COSMETICS, AS YOU KNOW, TRADITIONALLY IN THIS FIELD,
4    PEOPLE HAVE TRIED TO MAKE THINGS LOOK NORMAL.      AND
5    THAT'S BEEN THE SOCIALLY ACCEPTED THING.    NOW,
6    CURRENTLY SOME OF THE VETERANS RETURNING FROM THE WAR
7    LIKE TO SHOW THEIR PROSTHETIC LIMB, THAT THEY'RE
8    SHOWING WHY THEY'RE WALKING FUNNY.    SO I THINK THAT'S A
9    DEGREE OF SOMETHING TO DO WITH WHATEVER SOCIAL NORM IS,
10   WHAT PEOPLE LIKE AS FAR AS COSMETICS.
11             TRADITIONALLY, PEOPLE WITH DISABILITIES,
12   THEY'VE TRIED TO MAKE DEVICES THAT WILL GO UNDERNEATH
13   THEIR CLOTHING.   AND NOW, OF COURSE, THESE EXOSKELETONS
14   AND DEVICES HAVE TO BE EXTERIOR BECAUSE THEY'RE SO
15   BULKY.   SO, AGAIN, THAT'S A SOCIAL ISSUE, I THINK.
16             THE MODERATOR:   WE HAVE ANOTHER QUESTION NOW
17   ON MATERIALS.   WHAT ARE THE NEWEST ADVANCES IN
18   MATERIALS THAT YOU CAN SEE THAT WILL IMPACT ORTHOTICS?
19   ARE THE CURRENT MATERIALS SUFFICIENT TO THE DEPLOYER
20   SUCH AS OP PERSONNEL HAVE THE CAPABILITIES TO FABRICATE
21   WITH THESE MATERIALS?    ARE THESE MATERIALS, THEIR USE
22   IN FINAL DEVICE COST COMPETITIVE WITH ALTERNATIVES SUCH
23   AS THE CURRENTLY USED ONES?
24             DR. BENNETT:   THERE'S A LOT.   SO, WELL,
25   CURRENTLY THE PROSTHETIC ORTHOTIC PEOPLE, THEY MOSTLY



                                 30
1    WORK WITH THE PLASTICS.    AND THEY HAVE ABILITIES TO
2    WORK WITH OTHER MATERIALS AS WELL.     I THINK THAT IF NEW
3    MATERIALS CAME IN, THERE WOULD HAVE TO BE SOME
4    RECALIBRATION FOR THE PEOPLE WHO MAKE THESE DEVICES AND
5    HAVE TO TAKE IN THIS NEW TECHNOLOGY.
6              RIGHT NOW I THINK THAT THE POLYPROPYLENE IS
7    CLEARLY PROBABLY THE LEAST EXPENSIVE WAY TO GO.     IT'S
8    EASILY CUSTOMIZABLE FOR THE INDIVIDUAL.     I THINK THAT
9    THE MAIN ISSUE WILL BE -- IT'S A COST ISSUE.     IT WILL
10   BE -- I THINK OTHER MATERIALS WILL PROBABLY BE MORE
11   EXPENSIVE, BUT THE WHOLE DEVICE WILL OBVIOUSLY BE MORE
12   EXPENSIVE.   AND THE QUESTION IS HOW LONG DOES IT LAST?
13   I THINK THAT WE HAVE TO ADDRESS THIS AS WELL; THAT IS,
14   IF YOU WANT TO GIVE AN ORTHOTIC DEVICE TO A CHILD,
15   THEY'RE GOING TO OUTGROW IT IN A YEAR OR SO.     HOW CAN
16   WE RECYCLE ACTIVE COMPONENTS SO THAT WE CAN JUST
17   REPLACE CERTAIN PARTS THAT THEY'LL OUTGROW AND CONTINUE
18   ON?
19             THE MODERATOR:   NEXT QUESTION WE HAVE ABOUT
20   MODELING ON THE UPPER EXTREMITY.     CAN YOU COMMENT ABOUT
21   THAT?   CAN YOU ENVISION HOW THE REAL DYNAMIC LOADS THAT
22   ARE TRANSMITTED DURING AMBULATION ON UNEVEN TERRAIN
23   SUCH AS OUTDOORS WALKING CAN BE MODELED?
24             DR. BENNETT:    OH, YES.   I MEAN I DON'T KNOW
25   THAT MODELING IS TOO MUCH OF A PROBLEM.     SO I THINK



                                 31
1    THAT WE CAN -- INDEED, WE CAN PUT IN AN UNEVEN TERRAIN
2    AND CREATE A MODEL FOR THAT.     I DON'T SEE A PROBLEM.
3    TO BE QUITE HONEST, MOST PEOPLE TO DATE HAVE NOT DONE
4    THIS BECAUSE THEY'RE STILL WORKING OUT THE KINKS, AND
5    THE EASIEST PROBLEM IS OBVIOUSLY A FLAT WALKING,
6    WALKING ON FLAT GROUND, I SHOULD SAY.     AND SO WE CAN
7    MODEL UNEVEN GROUND.    IT IS A HARDER PROBLEM, BUT
8    CONCEPTUALLY NO HUGE CHALLENGES THERE.
9             THE MODERATOR:    AND WE APPRECIATE THE
10   OVERSIGHT YOU'VE GIVEN US, AND THIS IS THE LAST
11   QUESTION WE'LL HAVE FOR YOU.
12            COULD YOU COMMENT, BACK AGAIN TO THE TRUNK,
13   CAN YOU COMMENT ON THE POSSIBLE IMPORTANCE OF TRUNK
14   PELVIC ORTHOSES IN DEVELOPING OR ASSISTING LOCOMOTION
15   SUCH AS IN PARAPLEGIA?
16            DR. BENNETT:    WELL, I THINK THAT THE
17   POSSIBILITY -- SEE, THE INTERESTING THING IS FOR THAT
18   TYPE OF POPULATION, I'D SAY WE'RE BACK TO THESE
19   EXOSKELETONS.   IT'S QUITE INTERESTING.    THEY DEVELOP
20   THESE EXOSKELETONS THAT SEEM TO WORK QUITE WELL FOR
21   HEALTHY INDIVIDUALS.    TO DATE I DON'T KNOW THAT THEY'VE
22   TRIED TO SEE WHAT WOULD HAPPEN IF SOMEBODY WITH LESS
23   MOTOR CONTROL TRIED TO OPERATE SUCH DEVICES.      CLEARLY
24   WHEN YOU GET TO SOMEBODY WHO HAS LOST FUNCTION OF BOTH
25   LEGS, NOW THE CONTROL SYSTEM HAS TO TAKE OVER ALMOST



                                 32
1    COMPLETELY.    IT'S A VERY CHALLENGING PROBLEM BECAUSE,
2    AGAIN, YOU HAVE TO GET THIS FEEDBACK FROM THE SYSTEM AS
3    TO HOW THE PERSON CAN SHIFT THEIR BALANCE TO HELP
4    MAINTAIN THE TRUNK STABILITY SO THAT THEY DON'T FALL
5    OVER.    THEY NEED TO WORK ON THIS, AND I DON'T KNOW HOW
6    THAT WILL WORK OUT.    IT'S, INDEED, ONE OF THE
7    CHALLENGES.    WE'LL HAVE TO SEE WHAT HAPPENS WITH THAT.
8               THE MODERATOR:   DR. BENNETT, THANK YOU SO
9    MUCH FOR DOING SUCH A GREAT JOB ON A VERY IMPORTANT
10   TOPIC.
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25



                                 33

								
To top