Transcript No Slide Title - Electrical & Computer Engineering

Lecture 16
Dimitar Stefanov
Functional Neural Stimulation for
Movement Restoration (FNS)
FNS – activation of skeletal muscles in attempts to restore
useful movement in upper and lower extremities of people
with impairments of the central nervous system.
1./ Unimpaired pathways to the muscle fibers:
Neural cell
excitation an electrical
stimulus
Muscle fibers
2./ Impaired pathways to the muscle fibers:
Electric source
Muscle fibers
Examples of the FES:
•Cardiac pace makers
•Moe and Post (1962) – first application of the FES as a functional orthosis
FES – solution of the problem of restoring of the human locomotion
and manipulation.
Best results – among people with incomplete paraplegia and people,
moderately affected with hemiplegia.
•Best results to restoration of the grip and release functions.
•Successful FES control of two type grasping: lateral prehension (e.g. key
grip) and palmar grip (e.g. three-jaw pinch).
•Elbow extension control – people with C4 lesions;
•Incensement of the movement range – people with C5/C6 lesions.
Control
systems
Biomechanics
FNS
Electrical
safety
Muscle and
nerve
physiology
Electrical
engineering
Three techniques for FES:
1. Surface stimulators - electrodes, placed on the skin surface over the muscle
or nerve to be stimulated (transcutaneous electrodes);
2. Percutaneous stimulators – permanently implanted electrodes with wires
chronically penetrating the skin which connect to to an external pulse
generator (intramuscular electrodes or percutaneous electrodes);
3. Implantable stimulators –
1. both the electrodes and the biocompatible enclosure are permanently
implanted in the body, near the excitable tissue.
2. A transmitting antenna on the skin surface delivers power and
information to the multiple simulation sites.
3. Excitation with minimal energy; activation of deep located muscle and
nerve tissues.
Stimulation with transcutaneous electrodes:
•Widely used (no surgical intervention is required)
•Poor selectivelity and reachibility of deep located muscles
•Great values of the stimulation voltage is required.
Example for an implantable multichannel FNS system
Functions of the wearable processor:
•Processing of biomechanical parameters (joint angles, foot contact);
•Generation of control signals;
•Maintain joint force against the muscle fatigue.
Problems, which should be solved for efficient gait:
•Choice of simulation patterns and voltages;
•Development of stimulus for full knee extension during
certain phases of the gait cycle;
•Coordinated control and graded contraction of different
muscle groups.
Selective simulation of nerve fibers
•Special electrodes for selective stimulation
•Tension control through starting with the slowest to the
fastest twitch motor units;
•More physiologically based control;
•Very useful solution in case of upper-limb FES.
Electrode for selective stimulation. The tube is placed around a
motor nerve.
Slow twitch motor units provide less tension than fast twitch
motor units but they do not fatigue as rapidly.
Unimpaired motor control:
Slow twitch fibers are recruited in activities which require low forces;
fast twitch fibers are recruited for activities requiring high speed and/or
high force. Slow twitch fibers are recruited for frequently occurring
activities which require low forces.
FES strategy for efficient gait restoration with electrodes for
selective stimulation (example):
•Slow twitch fibers could be recruited to maintain the postural
stability during standing;
•Fast twitch fibers could be used for joint movements (to initiate
and generate steps).
FES strategy for efficient upper limb movement
with electrodes for selective stimulation
(example):
•Slow twitch fibers could be recruited to maintain
the postural stability of the upper limb;
•Fast twitch fibers could be used for object lifting.
Response of the locomotor system to the FES
Individual character (black box)
Methods for parameters identification
A./ gripping force/electrical stimulus
B./ Elbow flexion force/FES
C./ Experimental parameterization of lower
extremity for FES control
Development of the strategies for FES based on the
modeling of the anatomical structure and location
of muscles and tendons
Knowing the location of the stimulated muscles and the average force,
produced during their stimulation, an efficient FES strategy can be
developed.
(a) Anatomical location of the muscle to be stimulated;
(b) Simple model of the Rectus femoris muscle;
(c) mechanical model of the Rectus femoris muscle.
Some research institutions where significant FES
research results are achieved:
University and Medical Center in Ljubljana, Slovenia –
peritoneal nerve stimulators, applied to over 2500 people;
gait simulators (four channels), applied to over 100 people
with spinal cord injuries;
Case Western Reserve University, the Cleveland Metro
Health Center, and Cleveland Veterans Affairs Medical
Center – development of peritoneal and implanted systems
for functional grasp, applied to 50 people with quadriplegia;
Cleveland Metro Health Center and Cleveland Veterans
Affairs Medical Center – complex gait for about 30 people
with paraplegia.
Description of the FNS signal
Pulse repetition rate (frequency) pulse width (duration),
amplitude.
Stimulus strength is related to the charge density.
charge density = current • time/area
range of 10-300 mC cm2
The current of the FNS depends on the proximity of the
electrodes to the target muscle tissue.
Typically 1-50 mA.
•Pulsed waveform
•Monophasic or biphasic pulses
•Burst (carrier) signal – reduction of the potential pain
•Sufficient current in the target area for a length of time (100 – 600
mS)
•Pulse amplitude – depends on the size of the electrode and the degree
of current spread between the electrode and targets.
Current generator:
•Produce regulated current between 0 and 60 mA;
•Voltage – from 0 to 180 V
•Maximum 50 pulses per second to prevent the fast fatigue
•Low duty-cycle – to maintain the charge balance and to minimize the risk of
tissue damage
Optimal stimulation of the concrete muscle group – very important condition for
successful FNS.
Requires knowledge of the response of each muscle to stimulation and
knowledge of the muscle fatigue.
Systems for FNS:
1. Open loop control – time pattern stimulation. Sensitive
to external disturbances, the muscle fatigue cannot be
considered.
2. Closed loop control – (non-linear and adaptive
controllers; fuzzy-controllers).
3. Problems:
1. It is difficult to be obtained meaningful
physiological feedback from the stimulated musclejoint system; time delay between the stimulation
and force production; nonlinear characteristics
between the muscle length and the muscle force.
2. It is difficult to be found precise model of the
muscle recruitment for the concrete patient.
Increased burst time is applied in case of muscle fatigue.
Block diagram of FNS model
The time delay between muscle stimulation and muscle
activation is called the neural dynamics.
Non-linear relationship between input activation and generated joint torque
(depends on the joint angle, the joint velocity and acceleration).
Limb dynamics – depends on the mass and inertia characteristics of the limb.