Content uploaded by Omar Alsaif
Author content
All content in this area was uploaded by Omar Alsaif on Mar 24, 2022
Content may be subject to copyright.
Journal of Engineering Science and Technology
Vol. 17, No. 2 (2022) 1068 - 1077
© School of Engineering, Taylor’s University
1068
DESIGN AND IMPLEMENTATION OF
LOW-COST MEDICAL AUDITORY SYSTEM OF DISTORTION
OTOACOUSTIC USING MICROCONTROLLER
ABDULRAFA H. MARAY,OMAR IBRAHIM ALSAIF*, KIFAA H.TANOON
Mosul Technical Institute, Northern Technical University, Mosul, Iraq
*Corresponding Author: omar.alsaif@ntu.edu.iq
Abstract
New-borns hearing screening and diagnosis are very important for the early
discover of any problem that might affect their hearing and consequently their
communication. Babies may hear or respond to some sounds, but this is not
enough indication that they can hear all the sounds properly. The early diagnosis
of hearing efficiency and the functionality of the internal cochlea using
Otoacoustic Emission is necessary. In this research, a new device of Distortion
Otoacoustic system was designed and developed to be used as prototype
instrument for technical college student training. The low-cost microcontroller
developed device is efficient and sensitive as it is capable of generating and
capturing signals from the external ear canal, as well as analysing signals and
determining the efficiency of the inner ear. The proposed algorithm for the
microcontroller based on generating two sinusoidal waves with different
frequencies that transferred by MP3 shield cable. Analog to Digital Convertor
(ADC) will be responsible of the control operation for all the system parts. The
input signals are generated by Pulse Width Modulator technology (PWM), and
the number of samples are n =28 or n = 256, with frequencies range value 0.5-8
kHz which is the human auditory range that a person can hear.
Keywords: Distortion otoacoustic, Hearing screening, Low cost; Medical auditory
system, Microcontroller.
Design and Implementation of Low-Cost Medical Auditory System of . . . . 1069
Journal of Engineering Science and Technology April 2022, Vol. 17(2)
1. Introduction
Hospitals in many under-developing countries are lacking in equipment and funds.
Some devices and equipment are given less priority such as the new-borns hearing
screening due to a large number of patient and limited resources. An efficient low-
cost device will make them available in every hospital, medical centres, or even
cheap enough to be available for families at home [1]. This will save the
government and people money, effort and time instead of doing the test at certain
hospitals only. The early diagnosis of the hearing problem will lead to early
treatment that may save the children haring and communication [2].
Hearing loss is one of the most common diseases among new-borns. Detection
in time is very important operation, to prevent the nervous system and avoid serious
problems like difficulty speaking, language and cognitive development. The
common and recommended procedures for early detection are based on acoustic
emissions (OAE) and / or an auditory brainstem response (ABR).
This work aims to design a low-cost system using embedded system, low-cost
open sources programs such as microcontroller unit and design graphic users’
interface and present or save the signals in personal computer [3]. The signal then
will be processed using FFT to analyse the different signal frequencies.
2. Materials and Methods
The system saves the signals in the data base. The signals will be analysed and
detect the specified spectrums with noise cancelation to enhance the SNR (Signal
to Noise Ratio). This will give fast results with high accuracy and low cost [4].
This study consists of three main scenarios:
• Storing of signals in the database and the analysis step will be later. The
advantages of the design are; allowing for collecting data from a large number
of samples; it is mobile and flexible that can be used in remote areas where
computers are not available [5].
• Real time results present on the LCD screen. This can be done by analysing
the data by (FFT) inside the (MCU) and writing algorithms to filter and give
the results directly on the LCD screen or can be printed.
• Data collection only for post analysis. In this method, the Microcontroller will
be used as data acquisition (Data Acquisition). In this case, it is only
transferring the data via USB port in the (MCU) to the personal computer, or
the mobile phone, and designing a graphic user interface (GUI) to Store,
analyse, detect, process noise, and plot graphs [6].
3. How it works
The system generates different sound wave frequencies (sine waves) that can be
tuned based on international standards by the Microcontroller. The signal will be
sent into the human ear to stimulate the cochlea. The signals generated inside the
ear will be captured by the sound sensor and transferred to the (MCU) to be
analysed, filtered and processed [7]. LCD screen, a personal computer screen or the
mobile phone can be used via the user interfaces to display the results that help to
instantaneously diagnose any abnormal conditions in the ear. This can give the
result for all age groups at a very low cost [8].
1070 A. H. Maray et al.
Journal of Engineering Science and Technology April 2022, Vol. 17(2)
4. Medical Examination of the Cochlea
There are two methods for examining the human ear using acoustic devices;
Transient Evoked Otoacoustic Emissions (TEOAE) and Distortion Product
Otoacoustic Emissions (DPOAE) method, which are explained as following:
• TEOAE: Sound signals are generated at different frequencies, and pulses
according to international standards, to stimulate the cochlea which will
generate a sound signal which can be received by sound waves sensor [4, 9].
• DPOAE, is a more common method. DPOAE is a medical amplifier device
that works to enlarge, filter and convert the input signals into audio signals that
the human ear can hear. It contains a button to enlarge from 20 dB to 40 dB as
needed. This amplifier works on two 9-volt batteries as shown in Fig. 1.
This method was applied in this study by generating two audio signals (sine
wave) at different changeable frequencies ranging between 0.5-8 kHz. These two
signals go to the amplifier to be converted into sound waves and then go to the ear
to stimulate the cochlea. The difference between these two signals is [10].
F1 = 1.2 × F2.
When cochlear is stimulated, it generates audio signals that can be captured and
analysed into a spectrum of waves. The received signals are combination of many
signals, including the original signals transmitted F1, F2 and the signal to be
captured Fd that generated by cochlear stimulation. The rest of the signals are
considered as noise.
In this paper, Arduino Uno controller microcontroller was chosen to be
programmed due to many reasons:
• Low cost.
• Contains a USB port to transfer data to a computer, a mobile phone or tablets
with high flexibility through the serial port.
• Contains 6 ADC ports with high conversion accuracy (20-bit).
• The ability to convert digital signals into analog signal using Pulse Width
Modulation PWM technology
• Low power source of 5v
• Small size and lightweight [11, 12].
5. The System Building Process and Components
The system can be divided into hardware and software.
5.1. Hardware specifications
Figure 1 shows the block diagram of the physical parts (hardware) of this system
[12]. The microcontroller chip used is ATmega328. Microcontroller chip is
connected to the sound card which is compatible with the microcontroller that
amplifies the signals and converts them into analog audio signals that can be
recognized by the human ear. The MIC port is the microphone and is used to
capture the audio signals generated inside the cochlea [13]. The picture below
shows the contents of the device that radiates inside the ear. It consists of two
Speicher’s (Piston Source) and one microphone (see Fig. 2)
Design and Implementation of Low-Cost Medical Auditory System of . . . . 1071
Journal of Engineering Science and Technology April 2022, Vol. 17(2)
Fig. 1. Block diagram of the physical parts of this system.
Fig. 2. Microphone and piston sensor from preamplifier DPOAE.
To convert the generated sine wave to audible signals using MP3 Shield with
Arduino card as shown in Fig. 3. which is equipped with data and voltages by
Arduino. The SPK stereo port is used as an output for the sinusoidal waves and the
MIC port is mono input port from feedback data [14].
Fig. 3. Amplifier sound card - shield with Arduino (VS1053- MP3).
1072 A. H. Maray et al.
Journal of Engineering Science and Technology April 2022, Vol. 17(2)
The proposed algorithm for the microcontroller generates two sinusoidal waves
with different frequencies transferred by MP3 shield cable controlled via the ADC
port. This port is connected to a variable resistance to calibrate the device to
generate the desired frequency [7, 9] as shown in Fig. 4 [15].
Fig. 4. Arduino Uno and pot resistance connection.
These signals are generated by Pulse Width Modulator technology. The number
of samples n =28 or n = 256, with frequencies between 0.5-8 kHz which is the
human auditory range that a person can hear [12].
The difference between these two signals is (F1 = 1.2 × F2). Table 1 shows the
practical signals that were used in this research.
Table 1. Frequency range of sine waves generation.
8
7.5
7
6.5
6
5.5
5
4.5
4
3.5
3
2.5
2
1.5
1
0.5
F2=kHz
9.6
9
8.4
7.8
7.2
6.6
6
5.5
4.8
4.2
3.6
3
2.4
1.8
1.2
0.6
F1=kHz
The feedback data from the cochlea is received using the microphone. These
signals go to MCU and saved in matrices form according to their frequencies. Due
to the small MCU memory, an external memory board MSD type is connected to
expand the storage memory to store the received data. The data then will be analysis
and the signals are converted to have the spectra of frequencies values of F1= 65
kHz, F2=55 kHz and Fd=15 kHz. Subprogram is needed to remove the unwanted
noise that represented the band pass filter of the signal [16].
Signal power characteristics
The processed data can be displayed in three methods:
First method: It display the output on the LCD screen attached to the MCU
through (I2c) card that change and converts the screen connection from a parallel
link to a chain link as shown in Fig. 5 [10].
Design and Implementation of Low-Cost Medical Auditory System of . . . . 1073
Journal of Engineering Science and Technology April 2022, Vol. 17(2)
Fig. 5. The prototype of the device after
assembling AVR-microcontroller and liquid display.
This reduces the numbers of wires between the LCD screen and MCU to four
wires only as shown in Table 2. Table 3 shows the address (labels/naming) of the
I2C card.
Table 2. I2C Board pines.
GND
VCC
SCL
SDA
0 V
5V-Dc
Clock
Data
Table 1. Knowledge address in I2C.
Inputs
I2c SLAVE
ADDRESS
A2
A1
A0
L
L
L
0×20
L
L
H
0×21
L
H
L
0×22
L
H
H
0×23
H
L
L
0×24
H
L
H
0×25
H
H
L
0×26
H
H
H
0×27
H=Open Jumper
L=Close Jumper
After generating the frequencies F1 and F2 using the (PWM) technique, the
analogue port (A3) send data from the amplifier to be saved in the internal memory
of the (MCU) in matrix. These data will be processed and analysed using (FFT).
The results of the hearing level will be judged by as percentage based on the
received signal frequency and intensity values [17].
This method does not require a computer system or a tablet, inexpensive, hardly
has any weight is mobile, and it can generate and receive signals self with Give the
result immediately after performing the necessary analyses and filtering it.
However, the main drawbacks of this method are the limited number of taken
samples and unable to double the accuracy and increase (Sampling Ratio).
The second method: This method is similar to the first method except for
replacing the (I2C) card and the display screen (LCD) with the external memory
card type (Micro DS).
1074 A. H. Maray et al.
Journal of Engineering Science and Technology April 2022, Vol. 17(2)
The data can be saved in an external memory in the form of matrices in order to
be analysed later by a computer or a portable tablet. This method does not require
include the signal analysis programs inside the microprocessor. It can take larger
numbers of testing samples of more than one patient, and it is able to increase the
accuracy of the signal more than the previous method. On the other hand, a computer
and a program (GUI) are needed for the hearing level result to be finalised [12, 17].
The third method: It is similar as the previous methods, where the
microcontroller generates audio signals (F1, F2), sends them to the patient's ear
and receive the audio signals (Fd). The user interface program inside the computer
will analyse process the signals and generate a graph that shows the detail results
and saves them inside the computer. In this case there is no need to consume the
internal memory of the processor or use an external memory [8].
5.2. Software algorithm
Figure 6 illustrates the MCU programming flowchart for the first method which
needs to configure the Cyril port in order to deal with the (I2C card and the LCD
screen). The microcontroller analyses the signals by (FFT) using C++ programming
language. To remove unwanted noise, this chart gives the percentage of diagnosis
and review of results directly on the LCD screen. The flowchart in Fig. 6 shows the
LCD and MCU programming method [17, 18].
Fig. 6. Flow chart algorithm for MCU and I2c-LCD.
Design and Implementation of Low-Cost Medical Auditory System of . . . . 1075
Journal of Engineering Science and Technology April 2022, Vol. 17(2)
6. Results and Discussion
Figure 7 shows the Matlab2019a analysis outcome of the system when data is taken
from the SD-Card. The signals are converted to spectra and the first frequency is 3.9
kHz and the second frequency is 3.3 kHz. The total result frequency is Fd=2.69 kHz)
on the x axis, and the y axis represents the signal intensity measured in units dB.
Fig.7. M-File using data from SD-Card
(F1=3.9 kHz, F2=3.3 kHz, Fd=2.69 kHz)
Figure 8 explains how the GUI operated using Matlab2019a software. The
required frequency can be chosen by clicking on the buttons to assign to the value
of the first frequency F1 to the microcontroller. Then the second wave will be
generated based on the formula of F1 = 1.2F2. these two signals will be sent first
to the amplifier, then to the ear. The signals generated by the cochlea are captured,
analysed and their spectrum displayed.
Fig. 8. GUI by Matlab2019a (F1=3.25 kHz, F2=2.7 kHz, Fd=2.2 kHz).
Figure 9 is a chart presents the level of analysis in the of the results of the
frequencies start from 1 kHz to 6 kHz. This figure shows that the transmitted waves
(F1, F2) are higher than the received signal Fd. The maximum Fd signal intensity
1076 A. H. Maray et al.
Journal of Engineering Science and Technology April 2022, Vol. 17(2)
(0.3 dB) was found at (2 kHz). In sick cases, the results are different from the forms
of these signals depending on the severity of the patient's response, and these
devices remain an aid to the doctor.
Fig. 9. Results analyses of data F1, F2 and Fd.
The horizontal axis in Fig. 9 explains the amount of change in frequency 1-6
kHz. The vertical axis also represents the amount of change in the intensity of the
signal (dB). As discussed previously, F1 and F2 are sine wave generated by the
system, while Fd represents the final result.
7. Conclusions and Future Work
In this paper, a new device of Distortion Otoacoustic system was designed, in order
to meet the requirements of the medical device market in terms of providing an
inexpensive device that works on a public microcontroller (Arduino).The results
shows a good capability of low-cost microcontroller to operate as a good device
that used to diagnose ear problems for children.
In future work, more sophisticated types of microcontrollers can be used, as
well as the ability to program them by artificial intelligence algorithms in order to
be more efficient and accurate.
References
1. Mohamad-Ali, B.S.; and Maree, A.H. (2012). Low coast medical data
acquisition system using AVR-MCU and PC. Journal of Theoretical and
Applied Information Technology, 42(2), 217-220.
2. Samanta, A.G. (2014). Interfacing Of PIC 18F252 microcontroller with real
time clock via I2C protocol. International Journal of Engineering Research
and Applications, 4(7), 152-156.
3. Abdullah, E.A.; Saleh, I.A.; and AlSaif, O.I. (2018). Performance evaluation
of parallel particle swarm optimization for multicore environment. 2018
International Conference on Advanced Science and Engineering (ICOASE),
Duhok, Iraq, 81-86.
Design and Implementation of Low-Cost Medical Auditory System of . . . . 1077
Journal of Engineering Science and Technology April 2022, Vol. 17(2)
4. Kaur; N.; Mahajan, R.; and Bagai, D. (2016). Air quality monitoring system
based on Arduino microcontroller. International Journal of Innovative
Research in Science, Engineering, and Technology, 5(6), 9635-9646.
5. Mohammed, N.L; Aziz, M.S.; and AlSaif, O.I. (2020). Design and
implementation of robot control system for multistory buildings. Telkomnika
(Telecommunication Computing Electronics and Control), 18(5), 2682-2689.
6. Chohan, A.Z.; Motla, Y.H.; Bibi, S.; Bibi A. (2019). Requirement - based
feature modeling in software product line. Pakistan Journal of Engineering
and Applied Sciences, 24, 57-65.
7. Saha, T. ; Jewel, M.K.H.; Mostakim, M.N.; Bhuiyan, N.H.; Ali, M.S.; Rahman,
M.K.; Ghosh, H.K.; and Hossain, M.K. (2017). Construction and development
of an automated greenhouse system using Arduino Uno. International Journal
of Information Engineering and Electronic Business, 9(3), 1-8.
8. Ogidan, O.K.; Temikotan, K.O.; and Chike, K.C. (2017). Development of an
Arduino microcontroller-based automatic load shedding module for teaching
and research. Conference: 2017 IEEE 3rd International Conference on
Electro-Technology for National Development (NIGERCON), Owerri,
Nigeria, 1130-1134
9. Hilberer A.; Laurent, G.; Lorin, A.; Partier, A.; Bobroff, J; Bouquet, F.; Even,
C.; Fischbach, J.M.; Marrache-Kikuchi, C.A.; Monteverde, C.A. ; Pilette, B.;
and Quay, Q. (2018). Temperature-dependent transport measurements with
Arduino. Paper in Physics, 10, Art. 100007, 1-8.
10. Butcher, P. (2009). Debug it!: Find, repair, and prevent bugs in your code
(Pragmatic programmers). (1st ed.). Pragmatic Bookshelf.
11. Alsaif, K.I.; Saleh, I.A.; Alsaif, O.I. (2014). Design and analysis WLAN nodes
performance based on roaming technique. Journal of Education for Pure
Science, 4(1), 228-235.
12. Greenberg, I. (2007). Processing: Creative coding and computational art. (2nd
ed.). Apress, Berkeley, CA, USA,.
13. Yahya, MA..; Alsaif, O.I; Saleh, I.A.; Abdulkhaleq, A.M.; Parchin, N.O.; Al-
Yasir, Y.I.A.; and Abd-Alhameed, R.A. (2019). Noise cancellation for
HIPERLAN/2 with open loop transmit diversity technique. Inventions, 4(3), 46.
14. Pine, C. (2006). Learn to program (Pragmatic programmers). (1st ed.).
Pragmatic Bookshelf.
15. Thanoon, K.H.. (2017). Morphological properties for feature extraction of
geometrical shapes. International Journal of Engineering and Innovative
Technology (IJEIT), 7(6), 43-49.
16. Zeebaree, D.Q.; Haron, H.; Abdulazeez, A.M.; and Zebari, D.A. (2019).
Machine learning and Region Growing for breast cancer segmentation. 2019
International Conference on Advanced Science and Engineering (ICOASE),
Zakho - Duhok, Iraq, 88-93.
17. Stroustrup, B. (1997). The C++ Programming Language. (3rd ed.). Addison
Wesley (Verlag).
18. Saleh, I.A.; Alsaif, O.I.; and Thanoon, K.H. (2020). Deep coverage strategy
for private wireless network power using hybrid (Salp Optimization-Genetic)
algorithms. Technology Reports of Kansai University, 62(3), 45-54.
