DESIGN AND DEVELOPMENT OF AUTOMATED SYSTEM FOR

LOCALISED ELF MAGNETIC FIELD STIMULATION OF THE

HUMAN BRAIN

Dean Cvetkovic and Irena Cosic

Australian Centre for Radiofrequency Bioeffects Research (ACRBR)

RMIT University, School of Electrical and Computer Engineering

GPO Box 2476V Melbourne VIC 3001 Australia

Keywords: ELF, EMF, Bioeffects, Brain, EEG, Robotic Arm, Stepper Motors, Bluetooth.

Abstract: The automated system was designed and developed for accurate and fast localisation of extremely low

frequency (ELF) electromagnetic field (EMF) exposure to any particular brain region and therefore record

for any changes in the EEG activity before and after stimulation. The automated system consisted of a

general user interface (GUI) where the users had the ability to precisely control and move an EMF source

(coil) via robotic arm to any EEG electrode position or region. The 3-D movements of the robotic arm were

controlled via a serial linked motor driver board that controlled two motors. The software was able to

initially store the estimated 3-D EEG electrode positions and therefore identify the areas where ELF EMF

exposure from the coil could be applied. The testing and final measurements of this system revealed the

robotic arm precision of 0.1mm and maximum speed of 0.211 cm/sec (x-axis) and 0.827 cm/sec (y-axis).

1 INTRODUCTION

A particular system needed to be designed with the

purpose to modernise bioelectromagnetic apparatus

and aid researchers in unravelling many of the

unknown changes in EEG activity due to ELF/RF

EMF exposures. Accurate and fast automated system

needed to be designed and developed to provide an

Extremely Low Frequency (ELF) Electromagnetic

Field (EMF) exposure at any specific 3-D positions

or region around the human head. Once the EMF

source (coil) could be positioned and exposure

applied to subject’s head, the resultant electrical

brain activity (EEG) could be recorded. The EEG

recording and ELF EMF exposure were not to be

applied simultaneously due to EMF interference and

induction of electrical fields within the EEG leads

and electrodes, resulting in inaccurate recorded EEG

signals.

1.1 Background

It has been known that the brain cells send electrical

signals along the fibres that make up their

communication networks. Considering that changing

(sinusoidal) magnetic fields can induce current in

electrical conductors, researchers thought that a

magnetic pulse might stimulate currents in the brain

and therefore alter the brain’s electrical activity

(EEG) (Nature, 2002). Researchers have used a

Transcranial Magnetic Stimulation (TMS) technique

in the past and have reported that depending on the

intensity, frequency and localisation of the magnetic

field applied to brain, the simple cognitive tasks

could be explained (Walsh V., 1999)(Ueno S.,

1999).

One study investigated whether an alternating

3Hz magnetic field of 0.1mT applied to the head

over a period of 20 minutes caused any changes in

EEG parameters. This study was able to justify that

alternating weak magnetic fields do affect the

specific EEG activity (Hausser K, 1997). It has also

been reported that magnetic field stimulation

localised in the region of the “Hess” area of cat’s

brain can cause an impulse and a proliferation of

reactions from neural cells, which as a result triggers

the sleep centre and causes it to resonate in

316

Cvetkovic D. and Cosic I. (2005).

DESIGN AND DEVELOPMENT OF AUTOMATED SYSTEM FOR LOCALISED ELF MAGNETIC FIELD STIMULATION OF THE HUMAN BRAIN.

In Proceedings of the Second International Conference on Informatics in Control, Automation and Robotics - Robotics and Automation, pages 316-321

DOI: 10.5220/0001186703160321

 SciTePress

synchronisation (Hu Y., 1998). A simulated brain

wave pattern of slow wave sleep was coupled from

the coil to cat’s brain, in order to induce the

encephalic electrical activity to synchronise

gradually with the external low-frequency magnetic

fields and thus induce sleep in the cat.

An interesting study, led by Cook and Persinger,

conducted an investigation to find any interaction

with and to control the EM correlates of

consciousness generated by the brain through

applied complex magnetic fields whose

characteristics simulated natural, transcerebral EMFs

(Cook C. 1999). They designed a device to produce

rotational, circumcerebral transient magnetic fields

between 5 and 10µT, using 8 solenoids, which were

arranged circumcerebrally (every 45°) at the level of

the temporal-orbital plane. There were two

conditions utilised for this experiment: clockwise or

counter-clockwise arrangement of the 8 solenoids.

For the clockwise condition, solenoid 1 was placed

over the right prefrontal region such that solenoids

1-4 were equally spaced (45°) over the right

hemisphere and solenoids 5-8 were equally spaced

over the left hemisphere, along the orbito-temporal

plane. For the counter-clockwise condition, solenoid

1 was placed over the left prefrontal region such that

solenoids 1-4 were placed over the left hemisphere

and solenoids 5-8 were placed over the right

hemisphere. The counter-clockwise arrangement

generated a field sequence that moved in a rostral to

caudal direction along the left hemisphere but in a

caudal to rostral direction along the right

hemisphere. The frequency characteristics employed

burst-firing and frequency modulated “Thomas

pattern” (Thomas A.W, 1997).

They concluded that during the counter-clockwise

rotations, the conscious awareness, as inferred by

perception of time intervals, involved a process that

may be recreated approximately every 20-25ms and

may be affected by the right hemispheric activity.

The design of the rotational eight solenoid system

with pulsed magnetic fields has led to undertake

design and develop an automated system that would

localise a single solenoid to any head region (EEG

electrode position) and apply a sinusoidal magnetic

field to it in order to investigate any changes in EEG

activity.

2 DESIGN PROCESS

In the project, there has been several functional

system attributes mentioned that the system should

and may include:

• Self calibration and alignment properties.

• Scan the X and Y coordinates of the robotic arm

and record these points as origin reference.

• Ability to keep a record of all coordinates and

10-20 International EEG Electrode Placement

Scheme points for each test subject.

• Capability to analyse the information in

conjunction with measured EEG readings at

specific testing intervals.

• Utilise Bluetooth module to send and receive

data to and from PC’s microcontroller board.

There have been many different processes used in

order to complete the design and development of

this system. The design processes have therefore

been divided into three major parts: Mechanical,

Software, and Hardware processes.

2.1 Mechanical Design Process

For the mechanical design process of the robotic

arm, important structure design factors needed to be

taken into account, such as precise component

dimensions; minimisation of friction due to movable

plate movements; light-weightiness; durability; and

material of non-electromagnetic properties. In order

to minimise any induced currents onto EMF

exposure system, and maintain stable EMF level

during the experiment, a non-electromagnetic

material such as acrylic was selected.

2.2 Software Design Process

The Control Software design process was

undertaken throughout 9 versions to test and

improve acceptable software modules and reject

unsatisfactory modules. There have been six

software modules developed: Table, Move, Head,

Web, Auto-calibrate, Settings and Communication

Module.

At the initial stage of software design process,

only Table, Move and Head modules were

undertaken. The other modules were completed in

the final design process stage. The main goal of the

‘Table Module’ (TM) was to display reference

positions for sets of points located on the head that

the arm must move to. The main function of this

module was to save a current reference position of

the arm to one set of points in the table and also to

DESIGN AND DEVELOPMENT OF AUTOMATED SYSTEM FOR LOCALISED ELF MAGNETIC FIELD

STIMULATION OF THE HUMAN BRAIN

317

update the selected point in the ‘Move Module’

(MM) to provide quick access to saved points. The

initial version of the TM only contained the data

table. At this stage only the main modules had been

created to visualise the software design. The second

version of the TM has been linked with the other

modules so that points could be saved in the table.

There was little more development to the TM until

version 6 where a large number of modules had been

altered. As a result the TM inherited a button from a

now obsolete module. This button allowed the TM

to record the current arm reference for the selected

point or to clear the selected point. The final version

of the TM has not visually changed much since

version 6. There has been a global option

implemented to change the display of the reference

points from Stepper motor positions to Degree

positions, which changed the way the module

displayed the saved points. There was also a

function implemented to save the points to an output

file, which can then be reloaded.

2D ‘Head Module’ (HM) has been provided as a

visual reference to select a set of points. A point

could be selected by clicking on any of the red dots

displayed in the image, which will update the current

selected point in the TM, as shown in Figure 1. The

first version of the HM was simply a plain 2D head

with marked X’s in the estimated point positions.

The final 3D HM has a three-dimensional view of

the head from multiple angles, which could allow

the user to focus on any required area of the head.

3D HM in version 6 was created from an AutoCAD

program of a head geometry that had been scanned

and measured to produce an accurate representation.

From this program, multiple jpeg images were

created of the head at different angles. The 3 buttons

at the bottom controlled the rotation of the head and

each labelled point was selectable.

‘Move Module’ (MM) was responsible for

sending the control signals, which moved the arm.

The final version of the MM provided 3 different

methods for scrolling to a point. The first method

was the fastest by a point and click on the head

diagram. The next method was to move the slider

bars on the left and the right of the head diagram.

These were useful if only movement in one axis was

required. The final method was the most accurate

and was useful for fine-tuning the arm position. It

was performed by pressing the arrow keys, the left

and right arrow keys move along the ‘Theta’ axis

and the up and down arrow keys move along the

‘Phi’ axis. The MM has been the most complex to

design, since it was the module responsible for

sending the control signals. The first version of the

move module contained only basic commands.

There was a select point option to select a previously

referenced point and a move button to go to the

selected point and some labels to indicate the current

position of the arm while it’s moving. Since the

control signal framework has not been established at

this point, it was difficult to determine the module

requirements.

Figure 1: Move and Head Modules

By version 6 the control signal framework was ready

to be defined and by the final version, the additional

features such as Connect/Disconnect and the

Movement Control options were introduced. The

Connect/Disconnect was used to secure a COM port

for communication to the PIC controller. The

Movement Control buttons moved the arm for the

initial calibration. Rather than use buttons to move

on an axis, slider bars and the original head model

from version 1 were used to visually select a point.

In the final version a button has been added to show

the maximum step readings sent by the PIC after a

full calibration has been performed. This also

allowed the software to request a full or a quick

calibration, as shown in Figure 1.

2.3 Hardware Design Process

The scope of the system hardware was centred on a

PIC18F6720 microcontroller, which was responsible

for moving the robotic arm and communicating with

the GUI. Firmware prototyping was done on a

PIC18F8720 EVB. Microchip MPLAB 6.63 was

used with the MC18C compiler for fast firmware

development and advanced debugging with the use

of the Microchip In Circuit Debugger (ICD 2) tool.

The following firmware modules were prototyped

on the EVB:

• RS232 communication structures / modules

• Interrupt driven Shaft encoder module.

• Stepper Motor control module.

• Most system algorithms, including the arm

movement and RS232 packet constructor and de-

constructor algorithms.

ICINCO 2005 - ROBOTICS AND AUTOMATION

318

Schematic and PCB design was done in Portal DXP

where many advanced features of the software

package were used to minimize error in the final

design. Due to the small track size and complexity

of the final PCB artwork and advanced PCB etching

process was used. Placement and soldering of the

surface mount

3 FINAL DESIGN SYSTEM

The final design system was once again divided into

three major parts: Mechanical, Software and

Hardware Design.

3.1 Mechanical Design

From the 1

design to the final design, the robotic

arm had undergone significant changes. Figure 2

below shows the final design of the robotic arm. It

was made up of a main frame (an arc), which could

rotate on its middle axis and a plate attached to it, to

move along the arc. The most vital factor was the

ability to rotate the robotic arm on its middle axis as

well as moving the plate to its full extent.

Figure 2: Move and Head Modules

In the initial design, a normal 180

arc was

designed to move the plate around. The U-shaped

corners allowed the plate to move to its full extent,

while the side of the U-shape corners could support

the full weight of the robotic arm. To hold the EMF

source (coil), a plate with a stepper motor attached

to it was designed. The plate has a holder

constructed of lightweight and strong aluminium,

consisting of non-magnetic properties. The stepper

motor was attached to a cog, which rotated and

moved the whole plate up and down the mechanical

structure. A laser pointer was also attached to the

middle of the plate. Figure 3 shows the EMF coil

positioned at top and side head positions.

Figure 3: EMF Coil Positioned at Top and Side Head

Positions.

To enable the user to position the robotic arm

directly on top of subject’s head, a base was

constructed to slide forwards and backwards and be

locked to specific position. To limit the movement

of the robotic arm to 180

on its X & Y-axis, micro-

switches were attached to the sides. Once the arc or

plate touched a switch, the software would register

that point as the limit. This was particularly useful

for auto-calibration and ensuring that the arm did not

exceed its maximum movement.

3.2 Software Design

The Control Software was composed of 7 main

modules. Each of these performed too many

functions to mention them all, but the main functions

they perform were: select a point; save a reference to

that point; alter arm reference point (Position to

move to); and move arm to reference point.

A functional block and directional arrows diagram

is shown in Figure 4 above, indicating flow of data

or an instruction.

Figure 4: Software Functional Block Diagram

The TM held arm reference locations for the 19

electrode placement points. It also held the currently

selected point and could be instructed to change the

selected point via the user interface or the 3D HM.

DESIGN AND DEVELOPMENT OF AUTOMATED SYSTEM FOR LOCALISED ELF MAGNETIC FIELD

STIMULATION OF THE HUMAN BRAIN

319

When the selected point has been changed, it sent

the arm reference locations to the MM for display. If

an empty point has been selected and the button

“Calibrate Point” was selected, the current reference

position of the arm was saved to the selected point

and updates the table. If by any chance a point was

selected that already has a reference position saved

to it, the button changes to “Reset Calibration” and

if selected, it will delete the reference position for

that point. The only function the 3D HM provided

was to update the selected point on the TM.

The primary responsibility of the MM was to hold

the selected arm reference position to be transmitted

to the PIC Controller. The arm reference position

could be altered by selecting a point with a saved

reference position or by manipulation of the controls

in the MM. Since the ‘Communication Module’

(CM) has no direct user interface, the MM has some

extra buttons to provide that interface. The

“Connect” and “Disconnect” captured a COM port

for the software to use for communication to the

PIC. When it was connected and the “Move” button

was selected, the selected reference position of the

arm was transmitted to the PIC. The axes used to

reference the arm positions were common circular

polar axes.

A small HTTP server has been integrated into the

software. This allowed a wireless enabled PDA with

a Web Browser to transmit move instructions,

allowing the operator to precisely conduct control

commands. Using ‘Auto-calibrate Module’ (AM)

and by calibrating the points listed, the remaining

reference positions could be estimated. This saves

the operator time by only having to save 6 reference

positions instead of the entire 19 points. The

reference positions were calculated by an

approximate estimate. The Settings Module served

to fine tune settings of the other modules.

The Communication Module (CM) was

responsible for interfacing between the MM and the

serial communications port. The Windows API

functions for the serial port were used to transmit

and receive data. A communications class for

Borland Builder C++ has been used as an interface

for the Windows API. When the “Calibration Info”

button on the MM was selected, it brought up a

window containing the current calibration data. If

“Yes” was selected then a full calibration request

was sent from the CM and the calibration data was

updated. If “No” was selected then a quick

calibration request was sent from the CM and arm

moved to a known position. If the “Move” button

was selected on the MM, a move request was sent

via the CM. The final testing the CM and parts of

the MM, required knowledge of the serial output.

This was accomplished by the creation of a PIC

Emulator program, which has been set up to test all

functions of the software program.

3.3 Hardware Design

The PCB was implemented with mostly surface

mount components and connectors. The hardware’s

PIC18F6720 microcontroller had an ability to move

the arm to any position, determine the current

position and direction of movement and

communicate with the GUI. Figure 5 shows the

functional block diagram of the hardware subsystem

interconnections to the micro controller.

Figure 5: Hardware Interconnections

The PIC18F6720 controller provided the

following functionality to:

• Handle both RX and TX communications with

the GUI.

• Drive the system stepper motors, via the

UCN5804 high power stepper motor drives.

• Determine current location from quadrate

output shaft encoders.

• Sense a domain limit changes on the limit

switch inputs.

• Move the arm to the required position as

instructed by the GUI.

The UCN5804 high power stepper drivers were

used to rotate the stepper motors at variables speeds

in both forward and reverse directions. There were

two UCN5804 drivers on the final design. These

drivers controlled the X and Y movements. The PCB

has prevision for a third driver to control automated

Z-axis movement if required in the future. Due to

the high power requirements of the stepper motors,

an average of 0.5A was passed though each of the

UCN5804’s, thus processor controlled cooling fans

were positioned over the drivers to avoid heating

beyond their specifications.

The PCB contained a MAX232 serial driver for

signal conditioning and data transmission. The

ICINCO 2005 - ROBOTICS AND AUTOMATION

320

MAX232 interfaced with the PIC18F6720 USART

and serial crossover was done on the PCB to make

the hardware more compatible with standard PC’s.

Communication to and from the PCB was done via

packet transmission and reception (9600 bps, 8 data

bits, parity none, stop bits 1 and no hardware flow).

The packet structure was a fixed length with an

identifier and data field. Bluetooth communications

module was attached to the RS232 DB9 connector

which then interfaced the USART module in the

PIC18F8720, enabling wireless communications

between the hardware and GUI.

4 PERFORMANCE & RESULTS

To test the hardware performance, a sequential test

procedure was constructed in a proceeding order

where the following test procedure required the

previous test to be valid. Arrays of small firmware

testing applications were written to exercises

specific components of the subsystem under test.

Once the automated system was finalised and fully

operational, particular measurements were taken to

determine the accuracy and speed of our system. It

was observed that the system had an accuracy of

0.1mm for both X and Y-axis. The speed of the X-

axis was 0.211 cm/sec and the speed of the Y-axis

was 0.827 cm/sec, as shown in Table 1.

Table 1: The Accuracy and Speed of X and Y-Axis

X-axis Y-axis

Accuracy (mm) 0.1 0.1

Speed (cm/sec) 0.211 0.827

5 CONCLUSION

The automated system was designed and developed

for accurate and fast localisation of extremely low

frequency (ELF) electromagnetic field (EMF)

exposure to any particular brain region. The

automated system consisted of mechanical, software

and hardware designs which all passed through a

development process of refinement and

improvement. The final design of the robotic arm

was constructed of a main frame (an arc), which

could rotate on its middle axis and a plate attached

to it, to move along the arc. A general user interface

(GUI) was also developed to precisely control and

move an EMF coil via robotic arm to any EEG

electrode position or region. The software was able

to initially store the estimated 3-D EEG electrode

positions. The 3-D movements of the robotic arm

were controlled via a serial linked motor driver

board that controlled two motors.

ACKNOWLEDGEMENT

The authors gratefully acknowledge the NHMRC for

their support of the ACRBR which has assisted this

work.

REFERENCES

Magnetic Mind Games, 2002, Nature, 417, pp.114-116.

Ueno S., 1999, Biomagnetic Approaches to Studying the

Brain, IEEE Engineering in Medicine and Biology, pp.

108-120.

Walsh V. and Rushworth M., 1999, A Primer of Magnetic

Stimulation as a Tool for Neuropsychology,

Neurolopsychologia, 37, pp.125-135.

Hausser K., Tellschaft D. and Thoss F., 1997, Influence of

an Alternating 3Hz Magnetic Field with an Induction

of 0.1 mT on Chosen Parameters of the Human

Occipital EEG, Neuroscience Letters, 239, pp. 57-60.

Hu Y., Feng Y.M., Wang MS., Lu WW., 1998, The Effect

of Magnetic Stimulation on Potential Rhythm of

Cerebral Cortex, IEEE Engineering in Medicine and

Biology Society, 20(6), pp. 3288-3289.

Cook C. M., Koren S. A. and Persinger M. A., 1999,

Subjective Time Estimation by Humans is Increased

by Counter-clockwise but not Clockwise

Circumcerebral rotations of Phase-Shifting Magnetic

Pulses in the Horizontal Plane, Neuroscience Letters,

268, pp. 61-64.

Thomas A.W., Kavaliers M., Prato F.S. and Ossenkopp

K.P., 1997, Antinociceptive Effects of Pulsed

Magnetic Fields in the Land Snail: Cepaea Nemoralis,

Neuroscience Letters, 222, pp. 107-110.

DESIGN AND DEVELOPMENT OF AUTOMATED SYSTEM FOR LOCALISED ELF MAGNETIC FIELD

STIMULATION OF THE HUMAN BRAIN

321