Beruflich Dokumente
Kultur Dokumente
microelectrode
Recording site
neurons
Interfacing with the
Nervous System
Interfacing with Neurons
- Microelectrodes
- Electrical, neurochemical
- Microdrive mechanisms
Micromachined array of
neurowells partially loaded with
neurons. Parylene neurocage
(inset)
http://www.its.caltech.edu/~pinelab/new_pinelab_page/current-
research/current_research.html
Dielectrophoretic trapping of
neurons on microelectrodes
Planar electrode for
Dielectrophoretic
trapping of neurons
Fromherz, P. (2003). Semiconductor chips with ion channels, nerve cells and brain
slices. Neural Engineering, 2003. Conference Proceedings. First International IEEE
EMBS Conference on.
Hydrogel scaffold for guided
3D cell growth and migration
Passivation of SiO2
Patterning
P. A. Passeraub, A. C. Almeida, N. V. Thakor, "Design, microfabrication and characterization of a microfluidic chamber for the perfusion of brain tissue slices," Biomedical Microdevices, vol. 5,
pp. 147-155, June, 2003.
Neurochemical Sensor:
Microelectrode
Figure 1. (a) Design of a carbon fiber electrode for neurotransmitter, NO, measurements. The
membranes Nafion and the polymers mPD and Resorcinol provide selectivity against ascorbate,
nitrites and other confounding chemicals. b) Calibration (upper right) and sensitivity/selectivity
test of the NO electrode. (bottom right)
J. K. Park, A. Bandyopadhyay
Screen-printed C electrode arrays
George et al. used the screen printing of carbon ink to develop a grid of microsensors
for the sensing of neurotransmitters.
Neurochemical Sensor:
Photolithographic Fabrication
200 µm
Fig. 2. Left:
Schematic of the
16-electrode sensor
array (2-x-2
(mm)2. Right:
Close-up of a
single site. The
sensor site was
200-x-200 (µm)2.
The connecting
wire was 50-µm
wide.
A. 200µm
B. C.
200 µm
Courtesy K. Moxon
200 µm
High density
electrode array for
recording.
(http://www.eecs.umich.edu/~wise/R
esearch/Overview/Wise_Research.pd
f)
Neural Probe for Drug
Delivery
Micromachined drug
delivery probe with
channels for drug
delivery and electrodes
for stimulating
J. Chen, K.D. Wise, JF Hetke, Jr SCB (1997a), “A multichannel neural probe for selective chemical delivery at the cellular level”, Biomedical Engineering IEEE Transactions, pp 760-769.
MEMS Probe
Microelectrode Actuator
A
VLSI Potentiostat
Hippocampal
slice chamber
K. Murari, M. Stanacevic, G. Cauwenberghs, N. Thakor, “Wide-Range, Picoampere-sensitivity Multichannel VLSI potentiostat for
Neurotransmitter sensing” IEEE/Eng Med Biol Soc, San Fransisco, 2004, in press.
Neurochemical Sensing and
VLSI Potentiostat Interface
Christian Sauer, Milutin Stanacevic, Gert Cauwenberghs and Nitish Thakor,“Power harvesting and telemetry in CMOS Implanted devices”
Neural
Microdrive
Mechanisms
Micro-actuators in
Neurophysiology
Coronal section of adult rat
Rt—Reticular brain showing 4 microelectrodes
thalamus positioned (5-7 mm deep) in the
reticular thalamic nuclei at
different depths. The size of the
microdrive chip is expected to be
approximately 4 mmx 7mm.
microelectrodes
Upper
layer VI
pyramid
The schematic of two microelectrodes RT
superimposed over a biocytin tracing of
intrathalamic axonal arborizations. (Biocytin VPM
tracing from Bourassa et al., 1995).
200µm
Overview of the design
Photograph of completed microdrive before
implant
Motor Drive
• Excellent design by Cham[13]
– Design similar to Fee[12]
– Piezoelectric actuators used instead of
DC motors
– Chamber (hollow plastic cylinder) exist
for protection of the electrode assembly
– One end of electrodes attached to brass
brushings by set screws, other end to
the motor assembly
– Guide tubes are 500um apart
– Vertical positioner adjusts the position of
motor
– Pt-Ir microwires are sharperned and
glass coated at the tip
– Actuators are operated with a saw tooth
voltage signal.
– final implant weighted 40g and
implanted in rat.
Schematic of the design
Motor and loading of electrode
Electro-static microactuator
Principle
(1) d
Rotational motion
y
(2) V y-drive
x
(3) A
Link-
arms
B
Support x-drive
spring
SEM of gear assembly –
rotational to linear motion.
Left Right
(Sandia Labs, NM)
bank 1 bank 1
8
Non-conducting
Calibration with NO Epoxy
6
Current (nA)
Carbon fiber
4
D
2 A A B C C mPD/ R polymer
Nafion
0 Carbon fiber
150 250 350 450
Time (sec)
Screen Printed Sensor Array
(a)
(d)
(c)
60 µm (b)
30 µm
0 µm
1 mm
4 mm
Gold contacts
Bond wires
Chip with telemetry
circuitry
Telemetry link
for power/ data
www.medtronic.com
www.cyberonic.com
DEEP BRAIN STIMULATION
• Components
• Implantable
• Neurostimulators: A pacemaker-like device that
contains a battery and microelectronic circuitry for
controlled electrical pulse generation. Implanted
near the clavicle.
• DBSTM Electrodes: Four thin, insulated, coiled
wires bundled within polyurethane insulation. Each
wire ends in a 1.5 mm electrode, resulting in four
electrodes at the tip of the lead. DBS Leads are
implanted bilaterally in the brain
Details at http://www.spineuniverse.com/displayarticle.php/article2096.html
COCHLEAR IMPLANT
NUCLEUS® 3 COCHLEAR IMPLANT SYSTEM
1. Sound is picked up by a directional
microphone.
2. Sound is sent from the microphone to
the speech processor.
3. The speech processor analyzes and
digitizes the sound into coded signals.
4. Coded signals are sent to the
transmitter via radio frequency.
5. The transmitter sends the code across the skin to the internal implant.
6. The internal implant converts the code to electrical signals.
7. The signals are sent to the electrodes to stimulate the remaining nerve
fibers.
8. The signals are recognized as sounds by the brain, producing a
hearing sensation.
Details at http://www.cochlear.com/
RETINAL IMPLANT
A SILICON RETINA FOR
RESTORING VISION
Details at http://www.2-sight.com
IMPLANTED PILL: BION
• TELEMETRICALLY
CONTROLLED.
• CAPABLE OF SEVERAL
FUBCTIONS
• SELECTIVE FUNCTIONAL
STIMULATION
• DIAGNOSTIC READOUTS FROM WITHIN THE BODY
Details at http://www.advancedbionics.com/products/bion.html
Brain-Machine
Interface
BRAIN COMPUTER INTERFACE
• Brain talks to a computer !!!!
• a computer can learn what the brain is doing or going to do.
• brain can accept the command from computer.
• Goals
• To provide a way for people with damaged sensory/motor
functions to use their brain to control artificial devices and
restore lost capabilities.
• To train and control the animal to behave according to
human's command.
CASE STUDY: QUADRUPLEGIC
COMMUNICATOR
THE PROBLEM
A contradiction !!!
SOLUTION
Generate an on-screen keyboard and use inputs to navigate the on-screen
keyboard. Thus, a grid of characters was formed and the following navigational
controls were assigned.
1. Right Eye Closed = Move Right 2. Left Eye Closed = Move Left
3. Puff = Move Down 4. Sip = Delete 5. Both Eye Closed = Select
CASE STUDY: QUADRUPLEGIC
COMMUNICATOR
Right Eye
Left Eye
Blink Sensor Comparator
(IR sensor)
CPU
Sip
Sip and Puff Comparator
Pressure Sensor
Puff
CASE STUDY: QUADRUPLEGIC
COMMUNICATOR
CASE STUDY: QUADRUPLEGIC
COMMUNICATOR
Brain-Computer
Brain-Machine
Interface
J.R. Wolpaw et al. 2002
BCI (BMI) bypasses the brain’s normal pathways of peripheral nerves and muscles
Motivation for BCI/BMI Research
rebuilding instead of repairing
http://ida.first.fhg.de/projects/bci/bbci_official/
EEG based BCI (Brain Computer Interface)
• non-invasive
• promising for some therapies
• time-consuming
• not suitable for precise control
Mussa-Ivaldi & Miller, 2003
Neuron Spike based BMI (Brain Machine Interface)
Nicolelis, 2001
Trends in BCI/BMI research
• Brain to Action
• Sensation to Brain
• Brain2Brain Communication
Project
Toward two-way BCI/BMI
CCD camera
SCREEN
EEG Amplifier
Maze
Stimulus Command
Subsystems of the project
• Human EEG recording
– Electrodes (where, how many, holding cap)
– Amplifier (IBT or BioSemi, USB interface)
– Recording software (Labview)
• BCI software
– Visual feedback and stimulation (with rat motion on the screen)
– Command translation from EEG
– BCI paradigm for real time control (SSVEP or Motor imagining)
• Wireless micro-stimulator
– Transmitter and receiver pair (Bluetooth)
– Microprocessor to produce bipolar stimulation (TTL pulse train)
• Rat model
– Animal use/care protocol;
– Electrode implantation
– MFB reinforcement training
• Video motion tracking
– Rat motion tracking (to plot the route)
– Derivation of movement direction for proper MFB stimulation
– Real time target display for visual feedback of human
Wireless micro-stimulator
6
2
4
6
140 120 100 80 60 40 20 0
Rat model
• Medial Forebrain Bundle
3.0-4.8 mm posterior to bregma,
0.7-1.6 mm lateral, 8.3-8.4 mm
below the skull surface
• Wisker Barrel Field
3.0 mm posterior, 5.5mm lateral,
0.5mm below the cortex surface
PlasticsOne electrode
Implanting surgery
Shanbao Tong Ph.D and Homayoun Mozaffari, M.D.
Video Motion Tracking
Human EEG recording and BCI software
Key points of technique
• Speed compatibility between rat motion
and human mind control
• Position accuracy of micro-electrode
implantation
• Efficiency of wireless micro-stimulator
How to speed up EEG translation?
SSVEP solution
fL
fR
Motor imagery
Signal processing methods
(Time-Frequency domain)
• FFT
• Subband entropy
• Bandpass filtering + AR modeling
(use reflective coefficient ki instead of ai
because k vectors are orthogonal to each
other )
ECoG based gamma BCI project
140
127
120
100
80
60 SCI paper
48
40
20
2 4
0
1985- 1991- 1996- 2001-
1990 1995 2000 2004
What is Neurorobotics?
Neuronal
Population
A spinal injury is Recordings
The injury can damage and sever 3. Quantitatively measure the loss of
these axons. passive sensory input and motor
output
Cells in the brain will atrophy and 4. Use neurorobotics to address the
some will die. ‘functional’ capacity of the
surviving brain circuits
Outline
bonding pads
conducting
lines
recording
sites
Substrate
+ +
+
+
+ +
Mixed
+
+ +
Interface Zone
+
+
+ Coating
+
+ Interactive
+ Site +
+
+
Energetic Ions
Coating Atoms Substrate (s) Substrate Holder
Vacuum
Chamber
Energized
Ions
Evaporated
Alumina
Ion Source
Evaporator
Scanning
Electron
Micrographs
1.
A.
A.
B.
B.
C.
C.
2. 3.
Ceramic –based Multisite Electrode
In-vivo Implantation
A. 24 hours Post Surgery B. 3 weeks Post Surgery
site1
site2
site3
site4
Recordings from Single Cortical Barrel
A. B. C.
200µm
200 µm
200 µm
Neurorobotics: Part 1 Hardware Development:
VLSI Development
Input 1
OUT1
BIAS
Hybrid Circuit Neural Control Device
VDD
VSS
VSS
electrode subassembly PC Board
16 channel µ-probe
R1
variable
spacing
C1 VSS
C2 VDD
ground
OUT16
VDD
VSS
VDD
Input 16
ACG
Obeid, I. M., Morizio, J.C., Moxon, K.A., Nicolelis, M.A.L. Wolf, P.D. , IEEE Transaction on
Biomedical Engineering, 2002
Hybrid Circuit Neural Control Device
elevated platform
Future Directions