PATIENT SIMULATOR APPLIED TO AUDITORY EVOKED

POTENTIALS, ELECTROCARDIOGRAPHY AND

ELECTRONYSTAGMOGRAPHY

M. Tavares, C. Richter

Biomedical Engineering Laboratory, Catholic University of Pelotas, Rua Felix da Cunha 412, Pelotas, Brazil

R. Moraes

Electric Eng. Department, Federal University of Santa Catarina, Campus Universitário Trindade, Florianópolis, Brazil

T. Oliveira

PDI, Contronic Sistemas Automaticos Ltda.,Rua Rudi Bonow, 275, Pelotas, Brazil

Keywords: Simulator, AEP, ECG, ENG, electro-medical equipments maintenance.

Abstract: This paper describes an electronic device, named SimPac I, developed to simulate auditory evoked

potentials of short, middle and long latencies, ECG and electronystagmography signals. It uses sampled

waveforms in order to better reproduce real physiologic AEPs. The simulator is based on the ADuC841

microconverter, a device with an 8052-like core, FLASH memory and two 12-bit DACs. SimPac I is

portable and easy to operate, and it is very useful for calibration of AEP, ECG, ENG and VENG systems

during manufacture and maintenance. The simulator can also be used to support development and testing of

DSP algorithms intended to filter and/or average the above mentioned signals. As a result, examples of

several waveforms generated by SimPac I are shown.

1 INTRODUCTION

The auditory evoked potentials (AEP) play a

fundamental role in the audiology practice. The

capacity of capturing electric potentials generated in

response to acoustic stimulations resulted in many

relevant applications to the oto-neurologist (Katz,

1999). AEPs are classified in agreement with its

latency. Potentials of short latency occur in up to 10

ms after the auditory stimulation. Middle latency

potentials occur between 10 ms and 100 ms after the

stimulus, and long latency potentials are registered

after 100 ms from the stimulus. Short latency AEPs

are known by the acronyms BAEP – Brainstem

Auditory Evoked Potentials, BERA – Brainstem

Electric Response Audiometry or ABR - Auditory

Brainstem Response (Chiappa, 1997). The

electrocochleography (EcochG), used for cochlear

evaluation, is also considered a short latency AEP.

ABR is used for evaluation of the brainstem

integrity and also for objective audiometry. Middle

latency evoked potentials are identified with the

acronyms MLR or MLAEP (Middle Latency

Auditory Evoked Potential). MLAEP is indicated to

evaluate dysfunctions that could commit the hearing

pathways located between the brainstem and the

primary cortex.

P300 and MMN (Mismatch Negativity) are the

most used long latency AEPs in the clinical practice.

P300 presents a positive peal around the 300 ms

latency and using it makes it possible to obtain, in

only one test, information about the activity of the

transition thalamus-auditory cortex, the own

auditory cortex, about the hippocampus, hearing

attention and cognition. P300 is elicited through a

"rare paradigm", in which a few "rare" stimuli

happen randomly in a series of "frequent" stimuli.

The difference between both can be the intensity or

the frequency. The MMN test uses also rare and

frequent stimulus, but the result reflects the central

processing capacity (Caovilla, 2000).

130

Tavares M., Richter C., Moraes R. and Oliveira T. (2008).

PATIENT SIMULATOR APPLIED TO AUDITORY EVOKED POTENTIALS, ELECTROCARDIOGRAPHY AND ELECTRONYSTAGMOGRAPHY.

In Proceedings of the First International Conference on Biomedical Electronics and Devices, pages 130-134

DOI: 10.5220/0001052701300134

 SciTePress

ECG is certainly the most known bioelectric signal

generated by the human body. That signal is

captured and analyzed for use in clinical diagnosis,

surgical accompaniment and rehabilitation. Several

types of ECG simulators exist, able to simulate

changes in amplitude, heart beat frequency and many

types of arrhythmias (Prutchi and Norris, 2005).

They find application in the project of new monitors

and in the preventive and corrective maintenance.

The perfect corporal balance is very important

for the living organism orientation in the

environment. That balance widely depends on the

vestibular system, which acts in cooperation with the

visual system to maintain the vision focus during

head movements. The cerebral system that makes it

possible is known as vestibulo-ocular reflex (VOR).

The electric evaluation of VOR is done through the

electronystagmography exam (ENG) or by its

variant, the vector-electronystagmography (VENG)

(Castagno et al., 1994). The evaluation of

ENG/VENG is based on the registration and analysis

of the nystagmus, that is, the reflex ocular

movements which happen when the labyrinth

receive caloric or rotational stimuli.

The proposed Patient's Simulator was called

SimPac I and incorporates the technological

solutions used in the previously developed prototype

SPEA (Freitas et al., 2006). SimPac I has the

purpose of serving as a tool for development,

validation, adjustment and maintenance of AEP,

ECG and ENG/VENG equipments.

2 MATHERIALS AND METHODS

SimPac I is able to generate in two channels all of

the short, middle and long latency AEP waveforms

mentioned in the introduction. The morphology of

those signals was shown in a previous paper (Freitas

et al., 2006). Real signals of AEP were scanned to

obtain 500 samples for each one, and after they were

digitally processed in order to ensure the use of

whole the dynamic range of the 12-bit D/A

converters present in the simulator. Figure 1

contains the blocks diagram of SimPac I. The main

hardware component is the microconverter (µC)

ADuC841 (Analog Devices, 2003), whose CPU is

compatible with Intel’s 8052, modified to execute

instructions to 20 MIPS peak. The simulator does

not require external memories, since it uses only the

µC internal program FLASH and internal RAM.

ADuC841 is responsible for waveforms generation

using DDS technique (Digital Direct Synthesis),

starting from the sample tables stored in FLASH

memory (Grover and Deller, 1999). Energy is

provided by four NiMH 1.2V batteries, or by 5 V

external source. The microconverter requires 5 V

supply while the analog circuits are supplied by a

symmetrical voltage of ±5 V. Those voltages are

generated by a circuit which combines boosts with

switching capacitors and inverters. The complete

power circuit uses two chips LM2621 (National),

one ADM8660 (Analog Devices) and one ICL7662

(Maxim), including also a soft starter commanded

by µC. User's interface was remodeled regarding

SPEA. The rotary encoder and the push-buttons

were substituted by a membrane keyboard and a

graphic liquid crystal display (GLCD) with 8

kpixels. The GLCD driver is made by the µC itself,

through the ports P0 and P2.

The effective AEP simulated signal generation

depends on an external request, since the averaging

operation demands synchronism between auditory

stimuli and electric signal acquisition. That external

event consists of applying a voltage border in the

"external trigger" input. The direction of the border,

rise or fall, can be programmed in the simulator.

Optical

couplers

Manual trigger

External trigger

Schimitt

trigger

ADuC841BS62-5

INT0

INT1

Power supply

External 5 VDC

NiMH bateries

+5V

+5Vanalog

DVDD AVDD

FLASH

SRAM

+5V, -12V

DAC0

DAC1

Smoothing

filters

Programmable

attenuators

SPI

Output 1

Output 2

+5Vanalog, -5Vanalog

UART

USB

Interface

Keyboard

GLCD

128 x 64 pixels

Figure 1: SimPac I blocks diagram.

PATIENT SIMULATOR APPLIED TO AUDITORY EVOKED POTENTIALS, ELECTROCARDIOGRAPHY AND

ELECTRONYSTAGMOGRAPHY

131

For test and continuous type signals generation,

a button named "manual trigger" was included. In

any trigger type, manual or external, the pulse passes

through a Schmitt trigger circuit that gives to the µC

an interrupt signal free from bouncing.

Manual trigger causes the INT0 core interrupt,

while external trigger causes the INT1 core

interrupt.

The D/A conversion rate is set by an internal µC

timer, which is reprogrammed in agreement with the

selected signal to be generated. Each D/A output is

followed by a conditioning circuit that filter (low-

pass reconstruction) and reduce the signals to the

real voltage levels observed in the human body

signals. The amplitude adjustment is made by an

attenuator of 120 dB in range and 0.5 dB in

resolution, programmed through the µC SPI port.

2.1 Software

When the equipment is turned on, the software

exhibits a greeting message in GLCD, which is

followed by the main menu. Starting from that menu

the user can select the signal to be simulated. For

each signal, a specific sub-menu is exhibited

allowing adjustment of parameters as frequency,

amplitude, heart rate, angular velocity and others.

When the generation of a signal associated with the

external trigger is requested, the first task is to

program the reference timer with the corresponding

value for the signal’s sample rate. In the interrupt

INT1, the timer is trigged and its service routine is

responsible for the DDS signal generation. EcochG,

ABR, MLR and LLR are reproduced in the same

way: at each INT1 interrupt, a finite loop transfers

the samples stored in FLASH memory to the D/A

converters. The software routine which generates

P300 and MMN signals alternate the reproduction of

frequent and rare signals, stored in different tables,

to simulate the physiologic response to stimulation.

From each INT1 interrupt, the software establishes

which signal should be generated. The signals

generation sequence is established through an

oddball table also stored in FLASH. For generation

of the ECG signals, the start is given by the manual

trigger button. The timer defines the D/A conversion

rate and the different heart beat frequencies are

simulated varying the pause among two groups of

complex P-QRS-T and the T duration itself. For

ENG signals the technique is similar to that of ECG,

however the variations of angular velocity are

simulated changing the sweeping step of the tables

recorded in FLASH memory. The parameters used

on the SimPac I software regarding signal

characteristics are presented in Table 1. The

different combinations of periods and D/A

conversion rates are obtained programming the

reference timers prior to starting the generation of

each signal.

The SimPac I embedded software was developed

in C language with aid of the µC/51 V1.20.04

programming tool (Wickenhaeuser, 2005). After the

code compilation, the software was downloaded to

the program FLASH memory using WSD 6.7

(Analog Devices), and tests were made, which are

showed in next section.

Table 1: Main characteristics of simulated signals.

Signal Amplitude Period/frequency D/A Rate

EcochG

0.5 μV typical

5 ms 100 ksps

ABR

1 μV typical

10 ms 50 ksps

MLR

3 μV typical

50 ms 10 ksps

LLR

5 μV typical

500 ms 1 ksps

P300

5 μV typical

500 ms 1 ksps

MMN

5 μV typical

500 ms 1 ksps

ECG 0.5-1-2 mV 30-60-120-240 BPM 500 sps

ENG 0.1-0,5-1-2 mV 1-125 degrees/s 500 sps

Sine 0.5-1-2-10 mV 0.05-0.1-10-50-60-100 Hz 500 sps

Square 0,5-1-2-10 mV 0,05-0,1-10-50-60-100 Hz 500 sps

3 RESULTS

The ADuC841 was welded on the LQFP-to-DIL

adapting board showed in Figure 2, developed to

facilitate the access to the microconverter pins.

Figure 2: Adapter board from LQFP to DIL used in the

prototype, evidencing the ADuC841 µC and the serial

download interface to internal FLASH.

The other prototype components were mounted on a

universal pre-drilled board, and the whole circuit

was conditioned in a plastic box (Phoenix

Mecano/BOPLA), as it can be observed in Figure 3.

SimPac I prototype was exhaustively tested with a

commercial AEP equipment (Contronic, 2007),

presenting all AEP signals with synchronism,

amplitude and timing as expected.

BIODEVICES 2008 - International Conference on Biomedical Electronics and Devices

132

Figure 3: SimPac I layout.

Some of these signals are presented through the

software ATC Plus version 2.1.59 (Contronic,

2007a), and they are shown in Figures 4 through 6.

It is possible to note the perfect repetition of the

AEP signals generated by SimPac I.

Figure 4: Simulated ABR.

Figure 5: Example of simulated rare, frequent and

resulting P300 signals.

Figure 6: Simulated MLR (with VEMP) signal.

In Figures 7 and 8 some ECG and ENG signals from

SimPac I can be seen, which were registered directly

from the DAC outputs with the aid of a Rigol

DS5102MA oscilloscope. These simulated ENG

signals are being used in another project for

validation of automatic algorithms intended to

calculate the slow component of angular velocity of

positional, caloric and rotary nystagmus.

Figure 7: Simulated ECG signal at 60 and 120 BPM.

Figure 8: Simulated ENG at several angular velocities,

clockwise and counter clockwise.

PATIENT SIMULATOR APPLIED TO AUDITORY EVOKED POTENTIALS, ELECTROCARDIOGRAPHY AND

ELECTRONYSTAGMOGRAPHY

133

4 DISCUSSION

SimPac I carried out all the expected basic functions.

Several improvements were made regarding the

previous prototype (SPEA): ECG and ENG signals,

inclusion of performance signals intended to be used

on amplifiers and filters verification (sine and

square), inclusion of optical isolation in the external

trigger input, new user interface using GLCD,

function keyboard and soft starter. A circuit for USB

communication was added to make easy the future

inclusion of new signals to simulate.

For next version we intend to add a white noise

generator to simulate the EEG signal in which the

auditory evoked potential is immersed. That

characteristic will allow extension of the tests of

AEP equipments to the averaging quality.

Some simulators in the market generate

waveforms through complex mathematical formulas,

demanding digital signal processors for its

implementation. SimPac I generates the waveforms

from samples tables by DDS, and the execution can

be made through a simple microconverter. The

SimPac I main advantages are: the generation of

signals that resemble those observed on biological

systems, however with known amplitudes and

latencies; the generation of signals with excellent

repeatability; the substitution of the patient or

volunteer during the development of medical

equipments; and also in development, the

elimination of undesired factors of difficult control,

such as the electrode-skin impedance, other

bioelectric signals like spontaneous EEG or EMG,

and electromagnetic interference.

5 CONCLUSIONS

In this work we demonstrated the viability of

creating an equipment intended to simulate AEPs of

several types, ECG, ENG and performance test

signals. SimPac I simulated signals were verified

through a commercial system for AEP acquisition

and a digital oscilloscope, showing reliability and

precision in the requirements of synchronism,

amplitude, timing and repetition. The use of this

simulator can facilitate the software development

and validation for processing AEP, ECG and ENG,

as well as the hardware adjustment in production,

and the preventive and corrective maintenance of

electro-medical equipments.

ACKNOWLEDGEMENTS

The authors thank to Brazilian Agency CNPq by the

financial support under the grants 507363/2004-3,

310611/2005-9, 120005/2007-7 and 310058/2007-4

(DTI and ITI scholarships) and also to the designer

Leonardo de Jesus Furtado by the SimPac I layout

project.

REFERENCES

Analog Devices, 2003. MicroConverter 12-bit ADCs and

DACs with Embedded High Speed 62-kB Flash MCU.

Analog Devices Inc. Norwood.

Caovilla, H. H., 2000. Audiologia Clínica. Atheneu. São

Paulo.

Castagno, L.A., Tavares, M.C., Richter, C.M. et al., 1994.

Sistema Computadorizado de Eletronistagmografia e

Vectonistagmografia “UCPel/Castagno” (Versão 3.0).

Anais do IV CBIS, pp 26-31.

Chiappa, K. H., 1997. Evoked Potentials in Clinical

Medicine. Third Ed.. Lippincott-Raven Publishers.

Philadelphia.

Contronic, 2007. Manual do Usuário - Módulo de

Aquisição de Sinais Bioelétricos – MASBE – Rev. 4.

Contronic. Pelotas.

Contronic, 2007a. Manual do Usuário - ATC Plus -

Software para Audiometria de Tronco Cerebral –

Build 2.1.X – Rev. 4. Contronic. Pelotas.

Freitas, G.M., Oliveira, T.R, Moraes, R. e Tavares, M.C.,

2006. Simulador de Potenciais Evocados Auditivos de

Curta, Média e Longa Latência Baseado em

Microconversor. Anais do CBEB 2006, v 1, pp 1224-

1227.

Grover, D.; Deller, J. R.; 1999. Digital Signal Processing

and the Microcontroller. Motorola University

Press/Prentice Hall PTR. New Jersey.

Katz, J., 1999. Tratado de Audiologia Clínica, Ed. 4.

Manole. São Paulo.

Prutchi, D., Norris, M., 2005. Design and Development of

Medical Electronic Instrumentation. Wiley. New

Jersey.

Wickenhäeuser, 2005. uC/51 V1.20.04 User´s Manual.

Wickenhäeuser Elektrotechnik. Karlsruche.

BIODEVICES 2008 - International Conference on Biomedical Electronics and Devices

134