Professional Documents
Culture Documents
Belgaum-590014
Project Report
On
DIAGNOSTIC STETHOSCOPE
Bachelor of Engineering
IN
ELECTRONICS AND COMMUNICATION ENGINEERING
For the Academic Year 2013-2014
BY
BHANU PRATAP REDDY
BHARATH KUMAR V
CHETAN D
SHABANA BANU S
(1PE10EC018)
(1PE10EC019)
(1PE10EC023)
(1PE11EC420)
CERTIFICATE
This is to certify that the project work entitled DIAGNOSTIC STETHOSCOPE
carried out by Bhanu Pratap Reddy, Bharath Kumar V, Chetan D, Shabana Banu S,
bearing USNs 1PE10EC018, 1PE10EC019, 1PE10EC023, 1PE11EC420, respectively in
partial fulfillment for the award of Degree of Bachelors (Bachelors of Engineering) in
Electronics and communication Engineering of Visvesvaraya Technological University,
Belgaum during the year 2013-2014.
It is certified that all corrections/suggestions indicated for internal assessment
have been incorporated in the Report. The project report has been approved as it satisfies
the academic requirements in respect of project work prescribed for said degree.
______________
______________
Signature of guide
Mr. Kiran Kumar K V
Assistant Professor
Dept. of ECE
Signature of HOD
Dr. Subhash Kulkarni
HOD
Dept. of ECE
______________
Signature of the Principal
Dr. J Surya Prasad
Principal/Director
PESIT(BSC)
External Viva
Name of the Examiners
__________________
2.
__________________
ACKNOWLEDGEMENT
On the very outset of this report, we would like to extend our sincere and heartfelt thanks to our
guide Prof. Kiran Kumar K V and we are ineffably indebted to him for his conscientious
guidance and encouragement to the betterment of our final year project.
We are also grateful to our college PES Institute of Technology BSC for providing us the
opportunity and would also like to express a sense of gratitude to Prof. J Surya Prasad for the
continued effort in creating a competitive environment in our college.
We would also like to convey our heartfelt thanks to our H.O.D. Prof. Subhash S Kulkarni, for
giving us the opportunity to work on a project such as this and his encouragement throughout its
Course.
We also wish to thank all the staff members of the department of Electronics & Communication
for helping directly or indirectly in completing this work successfully.
Any omission in this brief acknowledgement does not mean lack of gratitude.
DIAGNOSTIC STETHOSCOPE
ABSTRACT
In medicine, diagnosis of a patient is in itself solves more than half of the problem of a
patient since knowing the problem gives us a reason to begin appropriate treatment. But it
is impossible for a diagnostician to be available conveniently. As we know that
sometimes if a problem goes undiagnosed then it may end up proving to be fatal.
Most of the abnormalities in a human being can be linked directly or indirectly to the
working of the heart. Hence medical diagnosis of heart problems should become
increasingly efficient and accurate. But sometimes or most of the times because of
inexperience or inability doctors prefer to consult Phonocardiogram, ECG and EKG
Physician thereby increasing the cost of medical care to the patient. So we look at means
of eliminating the human element by analysing various findings and applying diagnosis
algorithms for heart related problems.
Murmurs are a result of the presence of S3 and S4 symbols present along with S1 and S2
symbols in a heartbeat. Hence a partially portable device allowing us to diagnose heart
problems and consult a physician accordingly.
Hence the purpose of this project is to prototype a digital stethoscope to serve as a
platform for potential computer aided diagnosis applications i.e. heart rate calculation and
for the detection of cardiac murmurs.
DIAGNOSTIC STETHOSCOPE
TABLE OF CONTENTS
Abstract
Acknowledgement
CHAPTER
DESCRIPTION
PREAMBLE
1.1
1.2
1.3
1.4
Introduction
Overview
Basic Schematic Diagram
Circuit Diagram
2.1
2.2
Introduction
Literature review
PAGE No.
1
3
4
6
LITERATURE SURVEY
9
9
PROJECT PLANNING
3.1
3.2
4
14
16
18
20
21
22
Heart Rate
Heart Abnormalities and murmur
Wavelet transform
Noise Suppression
Hilbert Transform
Envelope Detection Using Hilbert Transform
24
27
30
31
34
35
MATLAB
Flow Chart for Heart Rate Calculation
Flow Chart for Heart murmur detection
39
40
41
DIAGNOSTIC STETHOSCOPE
CHAPTER
DESCRIPTION
PAGE No.
43
44
45
47
48
CONCLUSION
8.1
8.2
8.3
Conclusion
Project Outcomes
Future Enhancements
References
Appendix
List Figures
List of Flow Charts
52
53
54
55
56
DIAGNOSTIC STETHOSCOPE
LIST OF FIGURES
FIGURE DESCRIPTION
FIGURE
PAGE
NO.
NO.
1.1
1.2
Circuit of prototype 1
1.3
Circuit of prototype 2
1.4
Circuit of prototype 3
2.1
Plots Showing Heart Sound before and after removal of ambient noise.
10
3.1
A list of tasks planned and carried out during the course of the project
14
3.2
15
3.3
Resources Chart
15
3.4
Milestones Chart
16
4.1
18
4.2
19
4.3
19
4.4
20
4.5
Microphone sensor
20
4.6
21
4.7
22
5.1
36
6.1
39
7.1
43
7.2
43
DIAGNOSTIC STETHOSCOPE
FIGURE
FIGURE DESCRIPTION
NO.
PAGE
NO.
7.3
43
7.4
Prototype 1
44
7.5
Prototype 2
45
7.6
45
7.7
Prototype 3
47
7.8
48
7.9
49
7.10
49
sound
7.11
50
7.12
50
DIAGNOSTIC STETHOSCOPE
PAGE NO.
11
40
41
DIAGNOSTIC STETHOSCOPE
CHAPTER 1
PREAMBLE
DIAGNOSTIC STETHOSCOPE
PREAMBLE
INTRODUCTION
The average human life span as of today stands at around 80 years compared to that of an
average of 30 years when humans first appeared around 10,000 years ago. We owe this
significant improvement to the advances in medicine and healthcare. As of today medicine
and healthcare is major aspect in human lifestyle. Heart related ailments and treatments are
a major part of it, where PCG or Phonocardiogram is one of many methods utilised.
A Phonocardiogram or PCG is a plot of high fidelity recording of the sounds and
murmurs made by the heart with the help of the machine called phonocardiograph, or
"Recording of the sounds made by the heart during a cardiac cycle." The sounds are thought
to result from vibrations created by closure of the heart valves. There are at least two: the
first when the atrio ventricular valves close at the beginning of systole and the second when
the aortic valve and pulmonary valve close at the end of systole. It allows the detection of
sub audible sounds and murmurs, and makes a permanent record of these events. In
contrast, the ordinary stethoscope cannot detect such sounds or murmurs, and provides no
record of their occurrence. The ability to quantitate the sounds made by the heart provides
information not readily available from more sophisticated tests, and provides vital
information about the effects of certain cardiac drugs upon the heart. It is also an effective
method for tracking the progress of the patient's disease.
1|Page
DIAGNOSTIC STETHOSCOPE
PREAMBLE
In the case of our project, PCG is the basis of the algorithms we designed for heart rate
calculation and detection of murmurs, we intend to classify the input heart sounds as
normal or abnormal based on the features we extract. Although various features and
classification approaches have been successfully developed and tested, the performance of
the algorithms depends highly on the specific training and evaluation data sets. Pattern
recognition in medical diagnostics often suffers from a lack of data, particularly in
comparison to the problems being solved in non-medical fields such as voice recognition
in audio recordings or face detection in images. The lack of data is partly due to patient
confidentiality but is also caused by the limited number of available ground truth datasets.
Trained physicians and technicians are often needed to generate accurate ground truth
annotations. In comparison, almost any individual is capable of labelling faces in image or
the words that are being spoken in a recording.
2|Page
DIAGNOSTIC STETHOSCOPE
PREAMBLE
OVERVIEW
Cardiac murmurs are pathologic sounds that are produced by turbulent blood ow in the
heart. Detailed diagnoses of pathologic murmurs often require echocardiogram procedures.
Although the procedure is effective, it requires special equipment and trained technicians
to capture the necessary images and measurements. On the other hand, heart murmurs can
sometimes be detected by a physician using a standard stethoscope during auscultation.
This procedure is commonly performed during routine check-ups. However, depending on
the grade or severity of the murmur, the quality of the stethoscope, and the training and
skill of the physician, it can be difcult for a physician to distinguish a murmur from a
normal heartbeat. This design project aims to assist physicians in detecting heart murmurs
by analysing cardiac signals in real time during auscultation and reporting any detected
abnormalities. The task of designing a cardiac murmur detection algorithm has been
previously explored by several researchers in various academic groups. In general, the task
can be described as a pattern recognition problem using 1-dimensional medical data.
Pattern analysis typically involves two key steps:
Feature extraction
Classication
In feature extraction, one or more discriminative metrics are calculated using the input
data.
These metrics are then used in classication to assign a specic class label to the input data.
For the problem of detecting cardiac murmurs, the classication is binary assigning either
a normal or murmur label to the analysed data.
3|Page
DIAGNOSTIC STETHOSCOPE
PREAMBLE
MATLAB
DENOISE
ENVELOPE
DETECTION
THROUGH HILBERT
TRANSFORM
HEART RATE
CALCULATION
MURMUR
DETECT/DIAGNOSIS
4|Page
DIAGNOSTIC STETHOSCOPE
PREAMBLE
5|Page
DIAGNOSTIC STETHOSCOPE
PREAMBLE
CIRCUIT DIAGRAM
The Circuit Diagram of the three prototype circuit is given as follows:
Piezo Sensor Circuit 1
This circuit mainly acquired and amplified the Heart beat signal it did nothing to
filter the noise. We discuss in detail about this prototype in the later chapters.
Piezo Sensor Circuit 2
6|Page
DIAGNOSTIC STETHOSCOPE
PREAMBLE
The Only difference between Prototype 1 & 2 is the addition of TL072 in 2nd
prototype which allows some noise cancelling capability. We discuss in detail
about this prototype in the later chapters.
Microphone Sensor Circuit
7|Page
DIAGNOSTIC STETHOSCOPE
CHAPTER 2
LITERATURE SURVEY
8|Page
DIAGNOSTIC STETHOSCOPE
LITERATURE SURVEY
INTRODUCTION
With the advent of the 21st century medical diagnosis of heart problems has become
increasingly efficient and accurate. But sometimes or most of the times because of
inexperience or inability doctors prefer to consult ECG and EKG specialists thereby
increasing the cost of medical care to the patient. So we look at means of eliminating the
human element by analysing various findings and diagnosis algorithms for heart related
problems as would be diagnosed utilizing an electronic stethoscope.
Thus with the new methods being developed, which give us a different perspective
on the cardiac system using heart sounds as a potential parameter for diagnosing
Heart problems.
LITERATURE REVEW
9|Page
DIAGNOSTIC STETHOSCOPE
LITERATURE SURVEY
Fig2.1: Plots Showing Heart Sound before and after removal of ambient noise.
Whereas in the second paper Samuel E. Schmidt in Noise and the detection
of coronary artery disease with an electronic stethoscope(2010), focuses mainly on
the importance of a large data set and the types of noise encountered when acquiring heart
beat sounds through and electronic stethoscope and classified them as:
Ambient noise.
Recording noise.
10 | P a g e
DIAGNOSTIC STETHOSCOPE
Respiration noise.
Abdominal sounds.
LITERATURE SURVEY
In the third paper Haibin Wang is the one who really gets down to the
implementation in his paper Heart Sound Measurement and Analysis System with
Digital Stethoscope(2009) where utilizing a traditionally built electronic stethoscope he
describes the extraction of heart sound variables S1, S2 from normal patients and S1,S2,S3
11 | P a g e
DIAGNOSTIC STETHOSCOPE
LITERATURE SURVEY
and S4 from heart sounds of around 40 abnormal patients. He then summarizes his entire
diagnostic system in the following figure.
To conclude, the above are the substantial works of the respective authors which
adds weight to our project.
12 | P a g e
DIAGNOSTIC STETHOSCOPE
CHAPTER 3
PROJECT PLANNING
13 | P a g e
DIAGNOSTIC STETHOSCOPE
PROJECT PLANNING
Fig3.1:A list of tasks planned and carried out during the course of the project
14 | P a g e
DIAGNOSTIC STETHOSCOPE
PROJECT PLANNING
The above planned data was interpreted with the help of a gantt chart as follows:
15 | P a g e
DIAGNOSTIC STETHOSCOPE
PROJECT PLANNING
MILESTONES
With the project being planned into particular phases and the plan being carried out, it was
necessary for the success of each phase that we placed milestones. Ones that would mark
the end of a phase of the project and the beginning of another, every milestone is significant
to a particular block in the block diagram.
The projected milestones are:
WORKING PROTOTYPE
HARDWARE AND
ALGORITHM INTEGRATION
REALIZATION OF SENSOR
CIRCUIT
16 | P a g e
DIAGNOSTIC STETHOSCOPE
CHAPTER 4
HEART SOUND
RECORDING SYSTEM
17 | P a g e
DIAGNOSTIC STETHOSCOPE
18 | P a g e
DIAGNOSTIC STETHOSCOPE
For use as a sensor, the flat region of the frequency response plot is typically used, between
the high-pass cut-off and the resonant peak. The load and leakage resistance need to be
large enough that low frequencies of interest are not lost.
19 | P a g e
DIAGNOSTIC STETHOSCOPE
A simplified equivalent circuit model can be used in this region, in which Cs represents the
capacitance of the sensor surface itself, determined by the standard formula for capacitance
of parallel plates. It can also be modelled as a charge source in parallel with the source
capacitance, with the charge directly proportional to the applied force.
MICROPHONE
A microphone is an acoustic-to-electric transducer or sensor that converts sound in air
into an electrical signal. Microphones
are
used in
as telephones, tape recorders, live and recorded audio engineering. Most microphones
today
use electromagnetic
induction (dynamic
microphone),
capacitance
change
20 | P a g e
DIAGNOSTIC STETHOSCOPE
21 | P a g e
DIAGNOSTIC STETHOSCOPE
22 | P a g e
DIAGNOSTIC STETHOSCOPE
CHAPTER 5
HEART SOUND
ANALYTICAL SYSTEM
23 | P a g e
DIAGNOSTIC STETHOSCOPE
HEART RATE
Heart rate, also commonly known as pulse rate, is the number of times your heart beats
per minute. A normal heart rate depends on the individual, with age, body size, fitness level,
heart conditions, whether youre sitting or standing, medication and even air temperature.
Emotions can also have an impact on heart rate, as heart rate goes up when danger is detected or
other stress factors are experienced.
Some of the common factors that affect heart rate are:
Air temperature: When temperatures (and the humidity) soar, the heart pumps a little
more blood, so your pulse rate may increase, but usually no more than five to 10 beats a
minute.
Body position: Resting, sitting or standing, your pulse is usually the same. Sometimes as
you stand for the first 15 to 20 seconds, your pulse may go up a little bit, but after a
couple of minutes it should settle down. Emotions: If youre stressed, anxious or
extraordinarily happy or sad your emotions can raise your pulse.
Body size: Body size usually doesnt usually change pulse. If youre very obese, you
might see a higher resting pulse than normal, but usually not more than 100.
Medication use: Meds that block your adrenaline (beta blockers) tend to slow your
pulse, while too much thyroid medication or too high of a dosage will raise it.
Experts suggest that you should sit quietly for at least 10 minutes before taking your resting heart
rate.
Resting heart rate
For adults 18 and older, a normal resting heart rate is between 60 and 100 beats per minute
(bpm), depending on the persons physical condition. For children ages 6 to 15, the normal
resting heart rate is between 70 and 100 bpm.
Athletes and those in excellent physical condition can have resting heat rate of 40 bpm.
24 | P a g e
DIAGNOSTIC STETHOSCOPE
206.9 - (0.67 x Age). For a 50-year-old: 0.67 x 50 = 33.5, and 206.9 - 33.5 = 173.4.
The second is slightly more precise that the first, but the first is easier and more convenient for
most people to remember.
Target heart rate
You gain the most benefits and lessen the risks of cardiac disease when you exercise in your
target heart rate zone. According to the Centre for Disease Control and Prevention, for
moderate-intensity physical activity, a person's target heart rate should be 50 percent to 70
percent of his or her maximum heart rate. For example, using the results calculated above for a
50-year-old person, 50 percent and 70 percent levels would be:
For intense exercise, a 50-year-old person's target heart rate should be 70 percent to 85 percent of
his or her maximum heart rate:
It is not recommended to exercise above 85 percent of your maximum heart rate, as this doesnt
typically provide any further benefits and increases cardiovascular and orthopaedic risks.
DEPT. OF ECE 2013-14|PESIT BSC
25 | P a g e
DIAGNOSTIC STETHOSCOPE
26 | P a g e
DIAGNOSTIC STETHOSCOPE
Heart sounds are the noises generated by the beating heart and the resultant flow of blood
through it. Specifically, the sounds reflect the turbulence created when the heart valves snap
shut. In cardiac auscultation, an examiner may use a stethoscope to listen for these unique and
distinct sounds that provide important auditory data regarding the condition of the heart.
In healthy adults, there are two normal heart sounds often described as a lub and a dub (or dup),
that occur in sequence with each heartbeat. These are the first heart sound (S1) and second
heart sound (S2), produced by the closing of the AV valves and semilunar valves, respectively.
In addition to these normal sounds, a variety of other sounds may be present including heart
murmurs, adventitious sounds, and gallop rhythms S3 and S4
Disease of the cardiac valves and other cardiac structures frequently result in abnormal turbulent
blood flow within the heart causing murmurs. Careful auscultation of heart murmurs is an
extremely valuable tool in the diagnosis of many cardiac conditions. Heart murmurs will be
discussed below. Heart sounds are discussed elsewhere.
When normal laminar blood flow within the heart is disrupted, an audible sound is created by
turbulent blood flow. Outside of the heart audible turbulence is referred to as a bruit, while inside
the heart it is called a murmur. A pictorial representation of systolic and diastolic murmurs are
below:
27 | P a g e
DIAGNOSTIC STETHOSCOPE
achieved and the murmur may lessen or even disappear. Thus, the intensity of a murmur is not
used to indicate severity of disease.
A second cause of a murmur is valvular insufficiency in which blood abnormally travels
backward through an incompetent valve causing turbulence when it meets normal, forward blood
flow.
If blood is forced through a congenital anomaly from one chamber to another, as in an atrial
septal defect (ASD) or ventricular septal defect (VSD), a murmur is produced again due to
turbulence.
Yet another cause of cardiac murmurs is increased flow of blood through a normal valve. In high
output states such as anaemia, thyrotoxicosis, or sepsis, a large amount of volume is passing
through the cardiac valves and the normal laminar blood flow may be disturbed. Still's murmur is
a normal aortic flow murmur frequently heard in childhood. This frequently disappears over
time.
Murmurs are described by their timing in the cardiac cycle, intensity, shape, pitch, location,
radiation, and response to dynamic manoeuvres. Using the above, a clinician can accurately
characterize the nature of a murmur and communicate their findings in a precise manner.
Describing Heart Murmurs
Timing
The timing of a murmur is crucial to accurate diagnosis. A murmur is either systolic, diastolic, or
continuous throughout systole and diastole. Remember that systole occurs between the S1 and S2
heart sounds while diastole occurs between S2 and S1.
28 | P a g e
DIAGNOSTIC STETHOSCOPE
Once it is determined if the murmur is systolic or diastolic, the timing of the murmur within
systole or diastole also becomes important when characterizing murmur. Systolic murmurs can
be classified as either midsystolic (a.k.a. systolic ejection murmurs or SEM), holosystolic
(pansystolic), or late systolic. A midsystolic murmur begins just after the S1 heart sound and
terminates just before the P2 heart sound, so S1 and S2 will be distinctly audible. Conversely, a
holosystolic murmur begins with or immediately after the S1 heart sound and extends up to the
S2 making them difficult, if not impossible to hear. A mid-late systolic murmur begins
significantly after S1 and may or may not extend up to the S2.
29 | P a g e
DIAGNOSTIC STETHOSCOPE
Fourier transform based spectral analysis is the dominant analytical tool for frequency domain
analysis. However, Fourier transform cannot provide any information of the spectrum changes
with respect to time. Fourier transform assumes the signal is stationary, but PD signal is always
non-stationary. To overcome this deficiency, a modified method-short time Fourier transform
allows to represent the signal in both time and frequency domain through time windowing
function. The window length determines a constant time and frequency resolution. Thus, a shorter
time windowing is used in order to capture the transient behavior of a signal; we sacrifice the
frequency resolution. The nature of the real Partial discharge signals is nonperiodic and transient;
such signals cannot easily be analyzed by conventional transforms. So, an alternative mathematical
tool- wavelet transform must be selected to extract the relevant time-amplitude information from
a signal. In the meantime, we can improve the signal to noise ratio based on prior knowledge of
the signal characteristics.
In this work, we state only some keys equations and concepts of wavelet transform. A continuoustime wavelet transform of (t) is defined as:
1
f (t) * ( t b ) dt
(1)
Here a, b R, a 0 and they are dilating and translating coefficients, respectively. The asterisk
denotes a complex conjugate. This multiplication of a 2 is for energy normalization purposes so
that the transformed signal will have the same energy at every scale. The analysis function (t) ,
the so-called mother wavelet, is scaled by a, so a wavelet analysis is often called a time-scale
analysis rather than a time-frequency analysis. The wavelet transform decomposes the signal into
different scales with different levels of resolution by dilating a single prototype function, the
mother wavelet. Furthermore, a mother wavelet has to satisfy that it has a zero net area, which
suggest that the transformation kernel of the wavelet transform is a compactly support function
(localized in time), thereby offering the potential to capture the PD spikes which normally occur
in a short period of time.
30 | P a g e
DIAGNOSTIC STETHOSCOPE
Wavelet selection
To best characterize the PD spikes in a noisy signal, such as Fig 18, and Fig 20, we
should select our mother wavelet carefully to better approximate and capture the
transient spikes of the original signal. Mother wavelet will not only determine how well
we estimate the original signal in terms of the shape of the PD spikes, but also, it will
affect the frequency spectrum of the denoised signal. The choice of mother wavelet can
be based on eyeball inspection of the PD spikes, or it can be selected based on correlation
between the signal of interest and the wavelet-denoised signal, or based on the
cumulative energy over some interval where PD spikes occur.
31 | P a g e
DIAGNOSTIC STETHOSCOPE
We choose to select the mother wavelet based on the last two methods: correlation
Between two signals and cumulative energy over some interval of PD spike
occurrence.
We found that the two methods give us a very similar outcome.
Threshold limits
Many methods for setting the threshold have been proposed. The most timeconsuming way is to set the threshold limit on a case-by-case basis. The limit is
selected such that satisfactory noise removal is achieved. For a Gaussian noise; if
we apply orthogonal wavelet transform to the noise signal, the transformed signal
will preserve the Gaussian nature of the noise, which the histogram of the noise will
be a symmetrical bell-shaped curve about its mean value. From theory, four times
the standard deviation would cover
99.99% of the noise. Therefore, we could set the threshold be 4.5 times of the
standard deviation of the wavelet-transformed signal to remove the Gaussian noise
in the signal.
We have found that for the fibre optic signals, we could simply apply the standard
Deviation methods, since the signal is mostly white noises however for the PZT
signals, we should set the threshold case-by-case to best denoise the signals.
Two rules are generally used for thresholding the wavelet coefficients (soft/hard
thresholding). Hard thresholding sets zeroes for all wavelet coefficients whose
absolute value is less than the specified threshold limit. It has shown that hard
thresholding provides an improved signal to noise ratio. In this study, we adopt the
hard thresholding method.
32 | P a g e
DIAGNOSTIC STETHOSCOPE
Level of Decomposition
From the previous section, we have known that the wavelet transform is constituted
by different levels. The maximum level to apply the wavelet transform depends on
how many data points contain in a data set, since there is a down-sampling by 2
operation from one level to the next one. In our experience, one factor that affects
the number of 42 Level we can reach to achieve the satisfactory noise removal
results is the signal-to-noise ratio (SNR) in the original signal. Generally, the
measured signals from the PZT sensors have higher SNR than that of the measured
signals from fibre optic sensors. So to process the PZT data, we need more level of
wavelet transform to remove most of its noise. For the fibre optic sensor data, we
could only go up to 4 or 5 level otherwise we would remove much of the PD signal,
therefore the PD spikes wouldnt be captured.
33 | P a g e
DIAGNOSTIC STETHOSCOPE
In mathematics and in signal processing, the Hilbert transform is a linear operator which
takes a function, u(t), and produces a function, H(u)(t), with the same domain. The Hilbert
transform is named after David Hilbert, who first introduced the operator in order to solve
a special case of the RiemannHilbert problem for holomorphic functions. It is a basic tool
in Fourier analysis, and provides a concrete means for realizing the harmonic conjugate of
a given function or Fourier series. Furthermore, in harmonic analysis, it is an example of
a singular integral operator, and of a Fourier multiplier. The Hilbert transform is also
important in the field of signal processing where it is used to derive the analytic
representation of a signal u(t).
The Hilbert transform was originally defined for periodic functions, or equivalently for
functions on the circle, in which case it is given by convolution with the Hilbert kernel.
More commonly, however, the Hilbert transform refers to a convolution with the Cauchy
kernel, for functions defined on the real line R (the boundary of the upper half-plane). The
Hilbert transform is closely related to the PaleyWiener theorem, another result relating
holomorphic functions in the upper half-plane and Fourier transforms of functions on the
real line.
The Hilbert transform of u can be thought of as the convolution of u(t) with the
function h(t) = 1/(t). Because h(t) is not integrable the integrals defining the convolution
do not converge. Instead, the Hilbert transform is defined using the Cauchy principal
value (denoted here by p.v.) Explicitly, the Hilbert transform of a function (or signal) u(t)
is given by
provided this integral exists as a principal value. This is precisely the convolution
of u with the tempered distribution p.v. 1/t ; Alternatively, by changing variables, the
principal value integral can be written explicitly ) as
34 | P a g e
DIAGNOSTIC STETHOSCOPE
When the Hilbert transform is applied twice in succession to a function u, the result is
negative u:
provided the integrals defining both iterations converge in a suitable sense. In particular,
the inverse transform is H. This fact can most easily be seen by considering the effect of
the Hilbert transform on the Fourier transform of u(t).
where:
35 | P a g e
DIAGNOSTIC STETHOSCOPE
These functions are respectively called the amplitude envelope and instantaneous
phase of the signal
and
The amplitude function, and the instantaneous phase and frequency are in some
applications used to measure and detect local features of the signal. Another application
of the analytic representation of a signal relates to demodulation of modulated signals.
The polar coordinates conveniently separate the effects of amplitude modulation and
phase (or frequency) modulation, and effectively demodulates certain kinds of signals.
36 | P a g e
DIAGNOSTIC STETHOSCOPE
where
is the signal's complex envelope. The complex envelope is not unique; on the contrary, it
is determined by an arbitrary
with passband signals. If
is usually assigned to be a
37 | P a g e
DIAGNOSTIC STETHOSCOPE
CHAPTER 6
ANALYTICAL ALGORITHM
IMPLEMENTATION
38 | P a g e
DIAGNOSTIC STETHOSCOPE
MATLAB
The major aspect of this project as was discussed was the implementation of algorithms for
heart rate detection and murmur detection. There are innumerable number of ways to go
about achieving it namely with the help of microcontrollers, microprocessors, DSP
processors. In such a case it would increase the portability of the device but we regard
this as a future aspect of our prototype, as of know our main aim was to realise the
algorithm effectively if no efficiently hence we utilised MATLAB for the purposes of
signal processing wherein the input from the sensor circuit was given directly to the
microphone
port
of
the
PC
through
which
MATLAB
derived
the
input.
39 | P a g e
DIAGNOSTIC STETHOSCOPE
IF ABOVE
THRESHOLD
40 | P a g e
DIAGNOSTIC STETHOSCOPE
IF SATISFIES BOTH
THRESHOLDS
41 | P a g e
DIAGNOSTIC STETHOSCOPE
CHAPTER 7
PROTOTYPE
&
TESTING
42 | P a g e
DIAGNOSTIC STETHOSCOPE
Then MATLAB retrieves the signal by the process of recording it for a given duration
and later signal process it.
43 | P a g e
DIAGNOSTIC STETHOSCOPE
PROTOTYPE 1
The first prototype sensor was just basically a low voltage amplifier circuit which
employed LM386 heavily and had a constant gain. It also had the option for volume
control (refer chapter 1 for circuit diagram).
Fig7.4: Prototype 1
The piezoelectric sensor in the acoustic sensor needs to be biased in order for proper
operation. In addition, the output of the piezoelectric sensor is on the order of mill volts,
which is relatively small in magnitude. This makes it challenging for the Arduino to detect
changes in sensor output. In order to address both these issues, a bias and amplifier circuit
was designed and implemented to interface the raw sensor output with USB PnP Sound
Card.
DEPT. OF ECE 2013-14|PESIT BSC
44 | P a g e
DIAGNOSTIC STETHOSCOPE
PROTOTYPE 2
The second prototype was the extension of the first prototype where gain control was
introduce by adding a potentiometer between pins 1 and 8. A major aspect added is noise
reduction due to the addition of TL072 amplifier, because of its high CMRR it can reject
noise easily. (Refer chapter 1 for circuit diagram)
Fig7.5: Prototype 2
45 | P a g e
DIAGNOSTIC STETHOSCOPE
Circuit Description:
As per the circuit diagram mentioned in chapter 1 for prototype 2: U1a operates as a low-noise microphone preamp. Its gain is only about 3.9 because
the high output impedance of the drain of the FET inside the electrets microphone
causes U1as effective input resistor to be about 12.2K. C2 has a fairly high value
in order to pass very low frequency (about 20 to 30Hz) heartbeat sounds.
U1b operates as a low-noise Sallen and Key, Butterworth low pass- filter with a
cutoff frequency of about 103Hz. R7 and R8 provide a gain of about 1.6 and allow
the use of equal values for C3 and C4 but still producing a sharp Butterworth
response. The roll off rate is 12dB/octave. C3 and C4 can be reduced to 4.7nF to
increase the cutoff frequency to 1 KHz to hear respiratory or mechanical
(automobile engine) sounds.
U5 is a 1/4W Audio power amplifier IC with built-in biasing and inputs that are
referred to ground. It has a gain of 20. It can drive any type of headphones including
low impedance (8 ohms) ones.
46 | P a g e
DIAGNOSTIC STETHOSCOPE
PROTOTYPE 3
The third prototype was designed with durability in mind rather than quality of the signal.
It utilises a microphone rather than piezo sensor to acquire input signals. (Refer chapter 1
for circuit diagram)
Fig7.7: Prototype 3
47 | P a g e
DIAGNOSTIC STETHOSCOPE
TESTING
The entire system being designed is for naught if untested, hence we tested for the
feasibility of the stethoscope on both the levels i.e the hardware level as the software
level. The results below are of test cases of both recorded normal heart sound (namely of
a certain tem member) and that of an abnormal heart sound (from a database set of a
medical research university website).
Fig 7.8 depicts the opening screen of the matlab GUI program designed and its features
(Refer User Manual in Appendix for more Details.).
48 | P a g e
DIAGNOSTIC STETHOSCOPE
Fig7.9 and Fig7.10 are graphical representations of data or heart sound recorded and
analysis of a patient (namely a project teammate) as depicted by the MATLAB GUI
program.
Fig7.10: Envelope detected data and its peaks of teammates recorded heart sound
DEPT. OF ECE 2013-14|PESIT BSC
49 | P a g e
DIAGNOSTIC STETHOSCOPE
50 | P a g e
DIAGNOSTIC STETHOSCOPE
CHAPTER 8
CONCLUSION
51 | P a g e
DIAGNOSTIC STETHOSCOPE
CONCLUSION
CONCLUSION
A MATLAB based platform was developed for computer-aided diagnosis of cardiac
murmurs. Acoustic heart signals are captured and analysed in real-time with visualization
on a PC monitor for reporting. A new algorithm for murmur detection was tested and
evaluated and implemented in MATLAB on a Macintosh PC (can be implemented on any
other PCs) the resulting system is capable of aiding diagnosis by detecting murmurs with
reasonable accuracy.
The murmur detection algorithm that performed the best during preliminary testing was a
relatively simple feature. Calculations of the envelope require complex operations like the
Hilbert FT operator hence, would require a lot of effort when and if implemented on a DSP
Processor. However, the results of this project indicate that software part aside the
hardware decides the reliability factor of the circuit.
In the end, the diagnostic stethoscope although a relatively simple tool to build has great
potential as an everyday healthcare product.
52 | P a g e
DIAGNOSTIC STETHOSCOPE
CONCLUSION
PROJECT OUTCOMES
On the path to realise a device that has the capability to make a lifesaving diagnosis we had
to first map whatever we had assimilated from the curriculum over the entire course and
learn what we could and could not apply. We learned more than anything that it was our
basics that counted most than that of all the complex theories we were aware of, hence we
stuck to the basics and avoided any complex or roundabout methods. In terms of technical
knowledge we had to go beyond what the curriculum gave us, we had to become familiar
with aspects of bio-medical engineering such as PCG, and all theories relevant to the
algorithm for the device.
At first everything appears to be anarchy, until a clear project goal is defined. But this
wasnt the case. It took some time well before we could have a concrete vision for the
project. Once we had that there were little or no deviations, time was managed better
even though we lacked it in the end.
Overall the project outcome was more than favourable, its greatest achievement
according to us would be exposure to an aspect of engineering which we would have
never experienced if we remained in the shackles of the curriculum defined to us.
53 | P a g e
DIAGNOSTIC STETHOSCOPE
CONCLUSION
FUTURE ENCHANCEMENTS
The diagnostic stethoscope has few major aspects that can be improved upon in the
future, namely:
More reliable and effective sensor circuit, although this would mean increase in its
overall cost.
Increasing the complexity of the algorithm enough so that it can differentiate
between different types of heart related ailments.
Increasing the portability of the device by shifting the algorithm altogether to a
portable battery powered device namely microcontroller, microprocessor, DSP
processor.
54 | P a g e
DIAGNOSTIC STETHOSCOPE
REFRENCES
[1] "Heart Murmur." Wikipedia. Wikimedia Foundation, 12 Sept. 2012. Web.
[2] Delgado-Trejos E et. Al. Digital auscultation analysis for heart murmur detection.
Annals of Biomedical Engineering 2009;37:337-53.
[3] Thinklabs ds32a+ Electronic Stethoscope. <http://www.thinklabsmedical.com/>.
[4] Bentley, P. and Nordehn, G. and Coimbra, M. and Mannor, S. The PASCAL
Classifying Heart Sounds Challenge 2011.
<http://www.peterjbentley.com/heartchallenge/index.html>.
[5] G. Tzanetakis and P. Cook "Musical Genre Classification of Audio Signals", IEEE
Transactions on Speech and Audio Processing, 10(5), July 2002.
[6] Heart sounds and heart murmurs sepataion by Amina Atbi, Sidi Mouhamed Debbal
and Fadia Meziani
[7] An Electronic Stethoscope with Diagnosis Capability(2001),by Wah W. Myint
[8] Noise and the de- tection of coronary artery disease with an electronic stethoscope(2010),by Samuel E. Schmidt
55 | P a g e
DIAGNOSTIC STETHOSCOPE
APPENDIX
56 | P a g e
DIAGNOSTIC STETHOSCOPE
USER MANUAL
WHAT IS IT?
FEATURES
Heart Rate Calculator
Murmur Detector
SYSTEM COMPONENTS
INPUT
9V/0.5mA Battery Dry Cell
Acoustic Amplifier
Sensor
OUTPUT
Speaker (Audio Output)
Audio Jack/Port (Electrical Signal for Analysis)
PRECAUTIONS
Be Sure to utilise the device in environments devoid of any ambient noise for great results.