Professional Documents
Culture Documents
Direct Research
Cheap Alternate Medical Diagnostic
System & Device
Faculty Advisor
Dr. Mohammad Monirujjaman Khan
Associate Professor,
Department of Electrical & Computer Engineering (ECE).
Fall, 2020.
DECLARATION
1
This is to certify that this Project is our original work. No part of this work has been submitted
elsewhere partially or fully for the award of any other degree or diploma. Any material
2. Ahmed Zisan
APPROVAL
2
The review paper entitled “Cheap Alternate Medical Diagnostic System & Device” by
partial fulfillment of the requirement of the Degree of Bachelor of Science in Computer Science
Supervisor’s Signature
Department Chair’sSignature
ACKNOWLEDGMENT
3
First of all, we wish to express our gratitude to the Almighty for giving us the strength to
The capstone project program is very helpful to bridge the gap between the theoretical
knowledge and real life experience as part of Bachelor of Science (BSc) program. This report has
been designed to have a practical experience through the theoretical understanding. We also
acknowledge our profound sense of gratitude to all the teachers who have been instrumental for
providing us the technical knowledge and moral support to complete the project with full
understanding.
It is imperative to show our appreciation for our honorable faculty member Dr. Mohammad
Monirujjaman Khan for his undivided attention and help to achieve this milestone. Also, our
gratefulness is divine to the North South University, ECE department for providing us a course
such as CSE 498R in which we could really work on this project and materialize it the way we
4
ABSTRACT
Millions of people are dying or living in severe critical situations due to their suffering from
various lethal and life-threatening diseases. Even though the disease is not severe, regular health
check-up and screening for pre-determined or hereditary disorders often become a hassle because
of the complex treatment and diagnosis techniques. Expensiveness make the patients distressed
along with their family, especially in resource-poor middle and lower-income countries.
Alternative and cost-effective ways to diagnose and detect multiple diseases by analyzing their
specific clinical manifestations are being developed in today’s world of science and technology.
Widespread introduction of such systems has been a blessing for low-income countries with low
resource mechanisms. These devices play a vital role in diagnosing critical medical conditions by
replacing the conventional systems with cheaper alternate ways. In this study, widely used and
well recognized reasonable alternate diagnostic systems and devices have been listed with
5
Table of Content
Page
Chapter 1: Introduction…………………………………………………………….. 01
1.1 Introduction ……………………………………………………………………………….
1.2 Background Study...………………………………………………………………………..
1.3 Problem statement…………………………………………………………………………
1.4 Literature Review………………………………………………………………………….
1.5 Organization of the paper………………………………………………………………….
Chapter 2: Method and Materials………………………………………………..
2.1 Test Kits…………………………………………………………………………………...
2.1.1 Early result pregnancy test………………………………………………………………
2.1.2 HCG pregnancy test strips………………………………………………………………
2.1.3 Rapid Detection Pregnancy test…………………………………………………………
2.1.4 LH Ovulation test……………………………………………………………………….
2.1.5 blood sugar Testing Kit…………………………………………………………………
2.2 Biosensors and Nano-particles……………………………………………………………
2.2.1 Diagnosis of Infectious Diseases……………………………………………………….
2.2.2 Diagnosis of HIV……………………………………………………………………….
2.2.3 Diagnosis of Hepatitis…………………………………………………………………..
2.2.4 Diagnosis of Influenza………………………………………………………………….
2.2.5 Diagnosis of Ebola……………………………………………………………….……..
2.3 Applications of Biosensor in Animal Health……………………………………………..
2.4 Biosensors in Cardiac Diseases…………………………………………………………..
2.5 Microfluidic Technology…………………………………………………………………
2.5.1 Lab on chip (LOC) Technology………………………………………………….……..
2.5.2 Point of Care Test (POCT) ……………………………………………………….……
2.5.3 CD-based Devices………………………………………………………………...……
2.5.4 Paper-based Devices……………………………………………………………………
2.6 Test by a Single Drop of Blood………………………………………………………….
6
2.6.1 Diagnosis of Sickle Cell Anemia………………………………………………………
2.6.2 Detection of Dengue, Ebola virus or yellow virus ……………………………………
2.6.3 Detection of Tuberculosis……………………………………………….……..………
2.6.4 Detection of HIV and Hepatitis C………………………………………….……..……
2.7 IoT based healthcare and patient monitoring and diagnosis system……………………..
2.7.1 IoT based monitoring system…………………………………………………………..
2.7.1.1 ESP32 Processor……………………………………………………………………..
2.7.1.2 Heartbeat Sensor……………………………………………………………………..
2.7.1.3 Body Temperature Sensor (LM35) ………………………………………………….
2.7.1.4 Room temperature Sensor (DHT11) …………………………………………….…..
2.7.1.5 CO Sensor (MQ-9) …………………………………………………………………..
2.7.1.6 CO2 Sensor (MQ-135) ………………………………………………………………
2.7.1.7 ECG Sensor ………………………………………………………………………….
2.7.2 IoT based diagnosis of cardiovascular diseases………………………………………..
2.7.3 Diagnosis of diabetes and blood level monitoring……………………………………..
2.8 Smart Heath Monitoring System…………………………………………………………
2.8.1 Sensor-based health monitoring system………………………………………………..
2.8.2 Smartphone based heath monitoring system…………………………………………...
2.8.3 Web-based Imaging technology………………………………………………………..
2.8.4 Microcontroller based health monitoring system………………………………………
2.9 latest Technology in Diagnostic System…………………………………………………
2.9.1 Application of Artificial Intelligence…………………………………………………..
2.9.2 Detection of hypoglycemia from raw ECG signal……………………………………..
2.9.3 Blood sugar monitoring without finger pricks…………………………………………
Chapter 3: Result and Analysis……………………………………………...
Result and Analysis………………………………………………………………………….
Chapter 4: Conclusion and Future Work…………………………………
Conclusion and future Work…………………………………………………………………
Bibliography………………………………………………………………………
7
List of Figures
Heartbeat Sensor
ECG Sensor
Taxonomy of the reviewed smart health monitoring systems
Prototype of heart disease monitoring system
Block diagram of Arduino based heath monitoring system
A device from “Abbott” that continuously monitors diabetics’ blood
sugar levels without requiring backup finger prick tests
8
List of Tables
9
CHAPTER 1
INTRODUCTION
10
1.1 Introduction
In the modern era of advanced science and technology, medical field is greatly benefitted with
massive expansion of effective healthcare system, medication, diagnostic and treatment tools.
The sole purpose of treatment system lies in the betterment of patients in the shortest, most cost-
effective and caring ways possible. However, when the advancement of medical tools and
devices are developing in full swing, rate of patients getting affected and killed from different
diseases is also increasing every day. Even in this 21st century where we have already entered a
new decade, millions of people are dying for various infectious disorders and the scenario is
more crucial in underdeveloped as well as developing countries [1]. The cost of providing
effective healthcare services is now considered unsustainable and hard to afford in most high-
income developed countries. With a view to contributing more in the medical sector, researchers
in this field have been working relentlessly to introduce more affordable and sustainable ways of
Alternative and cheaper ways of diagnostic systems and devices are being introduced in medical
science without compromising any quality, effectiveness and outcome of the provided care [2].
Alternate ways of widely used medical diagnostics include low-cost pregnancy kit, blood sugar
ELISA-LOC system, point of care tests (POCT), rapid detection tests (RDT), nucleic acid
11
amplification tests (NAAT), acupuncture and many more significantly used diagnostic methods
and devices.
In this modern era, we are getting a huge number of facilities in medical sector. The structure of
medical diagnosis system has improved a lot in the last decade. Then again, there are some issues
which are bothering medical diagnosis system still now.
Many human beings are demise or residing in extreme imperative conditions due to their
struggling from quite a number deadly and life-threatening diseases. Even though the sickness is
now not severe, ordinary fitness check-up and screening for pre-determined or hereditary
problems regularly end up a problem due to the fact of the complicated therapy and analysis
techniques. Expensiveness make the sufferers distressed alongside with their family, particularly
approaches to diagnose and realize more than one ailment with the aid of examining their
particular medical manifestations are being developed in today’s world of science and
technology. Widespread introduction of such structures has been a blessing for low-income
12
international locations with low aid mechanisms. These units play a fundamental position in
diagnosing fundamental scientific prerequisites via changing the traditional structures with more
cost-effective alternate ways. In this study, broadly used and properly identified sensible
alternate diagnostic structures and gadgets have been listed with classifications and their perfect
Test kits
Microfluidic Technology
13
CHAPTER 2
METHOD AND
MATERIALS
14
2.1 Test Kits
2.1.1 Early Result Pregnancy Test
This accurately conducted test can detect all four of different forms of the human chorionic
gonadotropin (hCG) hormones present in urine. The test sticks are prepared to show the result
within three to four minutes and if it is done in the day expected menstruation date, the result is
Pregnancy or gestation period begins when a woman develops a fetus (offspring) inside her
womb after the fusion of an egg and sperm cell. Diagnosis for pregnancy can be carried out in a
variety of ways. All the pregnancy tests work by measuring the level of human chorionic
gonadotropin hormone which is produced by the placenta. It is usually found in a woman’s urine
and blood immediately after the implantation of an embryo in her uterus which can take 6 to 12
days after fertilization. The amount increases every day after that and almost takes 19 days to
rise the optimal level needed to be detected by any urine pregnancy test [5].
for a result. The visibility of two different color bands indicates positive test result. It is one of
the most reasonable ways to test for pregnancy [6]. The figure below represents two different
bands for positive result and only one band for negative result.
15
Figure 1: HCG pregnancy test strips [6]
Fig. 1 shows the HCG pregnancy test strips. This represents two different bands for positive result and
only one band for negative result
of a wide, color changing tip that turns white from pink makes the test rapid while the “flood-
to learn about. Although it is assumed that women ovulate in a certain timeframe according to
their menstruation cycle, the time might shift every once in a while. For this reason, it is better to
be assured through the luteinizing hormone test for ovulation. Like a regular pregnancy test, the
woman needs to pass urine through the strip and wait for three to four minutes. Positive result
16
2.1.5 Blood Sugar Testing Kit
Glucose biosensors are vastly used in diagnosing diabetes mellitus measuring the precise blood
glucose level consistently. Biosensors of blood glucose level detection are being used as a home-
devices used to transform any biological response into an electrical signal and these devices are
with digital output and a signal. When the analyte is captured, biological response is converted to
the signal to be received. In the vast area of medicine and various diagnostic and treatment
techniques, the clinical applications of biosensors and nanoparticles are growing rapidly.
17
Figure 2: Schematic representation of a biosensor [7]
antimicrobial susceptibility and antibiotic resistance, viral disorders are being studied using the
biosensor technology which is more cost-effective than other conventional methods. Biosensors
are recognized for their great impact to turn the current analytical methods into diagnostic
strategies by restructuring their sensing module for the detection of biomolecules, especially
18
2.2.2 Diagnosis of HIV
Human immunodeficiency virus (HIV) severely infects the CD4+ helper T lymphocyte cells and
rapidly enter into the bloodstream while replicating in an unregulated manner. The final stage of
the most dreadful disease as well as a major issue in public health. Rate of mortality for this viral
infection is increasing worldwide. To ensure specificity and sensitivity in diagnosing this viral
well recognized. Paper based electrochemical biosensors have also been used to identify the
DNA of hepatitis viruses, especially hepatitis B virus which is the most lethal among all virus
types under this class. In the detection and monitoring of hepatitis A, a thermosensitive surface-
Fig. 3a shows the process of the diagnosis of influenza [10]. Conventional techniques of
diagnosing this virus are not sufficient and time consuming. Alternative methods are developed
in this regard. In the detection and isolation of H9N2 sub-type of influenza A virus, attachment
19
of anti-matrix protein 2 antibodies to some iron magnetic nanoparticles was conducted to isolate
the virus followed by the addition of fetuin A and gold nanoparticles to detect the virus tacking
advantage from fetuin-hemagglutinin interaction. Afterwards the extracted complex was treated
with an acid solution to collect the gold nanoparticles on a carbon electrode. This particular
biosensor was reported to provide the rapid detection of H9N2 subtype of influenza A virus.
Another procedure involves the immobilization of DNA by using carbon nanotubes to detect
influenza virus. A DNA probe is attached to the biosensor and characterization is achieved by the
analyses of Fourier transform infrared interaction and Raman spectrometry. Observance of the
occurred changes on the surface of the biosensor helped in the detection of output signal of the
system. A fluorescent biosensor is also developed to detect H5N1 influenza virus in human
serum.
20
2.2.5 Diagnosis of Ebola
Fig. 3b elaborately exhibits the method of diagnosis of Ebola. Infection caused by Ebola viruses
are life threatening and often symptoms like hemorrhage, certain organ failure or shock-like
syndromes are observed. To diagnose the viral infection rapidly with point of care determination,
biotinylated hybrid with streptavidin and alkaline phosphatase conjugate [12]. The experiment is
thoroughly optimized with an electrochemical spectroscopy and the viral presence is detected by
Fig. 3c shows the technique is shown where the developed biosensor functions by using specific
antibodies and depends on the light transmission impact through plasmonic nanoholes. Direct
detection of the whole viral strains of Ebola is conducted by using optofluidic biosensor.. The
21
immobilization of the selected antibodies against Ebola glycoproteins on the biosensors and the
Figure 3c: Direct detection of Ebola virus using optofluidic biosensor [13]
Even though enzyme linked immune-sorbent assay (ELISA) is widely used in the
detection of dengue virus, researchers have been considering the advantages of biosensors in this
concern. The cost effectiveness, enhanced specificity and sensitivity, on-site monitoring with
quantitative analysis, simpler optimization along with rapid outcome are the major reasons
behind the popular use of this alternative method [14]. A silicon nanowire biosensor is broadly
used to detect the dengue serotype 2. Covalent attachment of a peptide nucleic acid on a silicon
silicon nanowire biosensor before and after the binding of the selected dengue serotype 2 with
22
A surface plasmon resonance biosensor is also developed to detect anti-dengue virus in
human serum sample [15]. Four dengue virus serotypes are used as ligands on the biosensor and
the obtained results show higher sensitivity and specificity in determining the ratio of each
dengue serotype used where a minimal amount of patient’s serum sample is needed.
management because of their timely diagnosis capacity with minimized economic loss. It is
particularly useful in dairy cattle and poultry farms. Determination of body temperatures and pH,
detection of sweat and saliva constituents, stress, behavior and movement, presence of any
pathogens and prevention of diseases, antibiotic detection to minimize antibiotic resistance all
fluorometric, immunoaffinity column assay and so on which are expensive, time-consuming and
require trained personnel to handle the procedure. Biosensors function by electric measurement
and are able to determine biochemical molecular condition with the expected selectivity and
specificity. Specific biomarker of interest is used here. In order to detect heart failure or other
cardiovascular difficulties in an early stage, hafnium oxide (HfO 2) based biosensor is developed
that can detect early-stage level of the human interleukin 10 (IL-10). Attachment and interaction
of recombinant IL-4 with selected monoclonal antibodies aid the detection of cytokine levels
23
after the administration of device. The interaction is characterized and analyzed by
Biosensors can also ensure the quantitative measurement of cardiac markers present in undiluted
serum. Extended applications are widely observed in immunosensor array for the clinical
resonance energy transfer, biochip for a quick and accurate detection of multiple cancer markers
Fluorescent biosensors function as imaging agents in cancer and drugs discovery and detection.
GFP (green fluorescent protein) based and FRET (fluorescence resonance energy transfer)
biosensors significantly work in the regulation of enzymes at a cellular level in this field of
medical science. The biosensor is a combination of small scaffolds where various fluorescent
probes are attached through receptor. The receptor then identifies specific analyte or target
molecule and transduce a fluorescent signal which can be readily detected and analyzed. Using
these biosensors, detection and diagnosis of arthritis, cardiovascular and inflammatory diseases,
viral infection, neurodegenerative diseases, cancer and metastasis are conducted. Potentiality and
specificity of specific drugs are analyzed by these biosensors for preclinical evaluation and
24
2.5 Microfluidic Technology
Microfluidics is referred to a specialized field of biomedical engineering extensively based on
the optimization and analysis of biological fluids such as blood, urine, saliva etc. by using
microchannels. This process is faster and more convenient compared conventional ways of
biological fluid analysis. Microfluidics based diagnosis and used devices are broadly classified
into two categories where first one is the development and use of a device that deals with the
flow of biological fluids through microchannels and the other one is based on technology that
deals with the detection or sensing of biomarkers of biological fluids for analysis [16].
Examination through this technique is reported to be a vital element during endemics and
epidemics where analysis of biological fluids can easily pave the way for effective diagnosis and
detects the causative agents or microorganisms that eventually help to predict the proper
treatment procedure. And thus, microfluidics can facilitate diagnosis, treatment options and
prevention in the wide arena of medical field. Different types of microfluidics tools and device
technology has reduced the time and pressure in the process of treatment-action in any medical
condition which often become fatal due to the intervals in treatment options [16]. LOC devices
comprise of three basic components such as fluid flow controlling devices like microchannels,
mixers, microvalves etc., sensors (optical or electrochemical attached to the chip) and decision-
making capability technology like micro-electro mechanical system (MEMS), micro total
25
analytical system etc. [17]. Flow cytometry, MEMS, micro total analytical system, PCR
field gel electrophoresis (PFGE), optical or electrochemical sensors technology are applied in
examination in order to improve health care procedure. The treatment measures are effectively
predicted by this test method [16]. POCT is broadly divided into two categories such as small
handled devices that provide both qualitative and quantitative results by detecting one or
multiple analytes (examples include glucose biosensor strips, lateral flow strips that detect
cardiac markers) and large devices that are used near the patient’s bedside. Example of large
Four steps are involved in devices of POCT including i) sample collection and pre-preparation,
ii) reagent mixing, iii) analytical assay running and iv) detection of result [16]. Some of the most
functions like mixing, valving, separation, bubble or droplet formation in the area of advanced
26
microfluidics technology [18]. Complicated biological assays like ELISA or nucleic acid
establish “zero cost diagnostics” [18]. Paper matrix is composed of huge number of cellulose
fibers allowing it to maintain the capillary force and control the fluid flow efficiently in
detection and analysis of certain analytes [18]. Among these, the colorimetric approach is
globally used for qualitative and quantitative detection and analysis. Use of papers with
plasmonic resonance that are impregnated in nanoparticles has enhanced the sensitivity of this
approach and this has been used for kidney biomarker detection. Diagnosis of hepatitis B virus
device that can aid the improved diagnosis and treatment rate for cancer [19]. The test functions
as a pregnancy kit and determines cancer within minutes. The technology is particularly based on
nanoparticles that interact with proteases (proteins that are known to be overexpressed in
numerous tumors having the ability to cleave peptide bonds), which can release hundreds of
samples or water sample and diagnose diabetes, malaria as well as detect water pollutants [4].
The functional potentiality can be extended to further diagnose HIV, Ebola, Dengue, influenza or
hepatitis. A single drop of blood sample is inserted into the device embedded on a paper strip and
diagnosis can be done within minutes. The device is connected to a cellular phone via cable and
mobile kit where results can be obtained within 10 minutes [4]. Sickle cell disease is a known
disease of red blood cells where the cells grow in an abnormal way and often get destroyed;
causing anemia. Obtained result is visible depending on the floatation ability of the collected
blood sample.
MIT have created a cost-effective device that can test for dengue, Ebola and yellow fever within
around 10 minutes [4]. Based on the techniques of molecular biology, screening with a single
28
Figure 4a: Low-cost device to detect Ebola, Dengue and Yellow fever [4].
resistance against the widely used antibiotic for tuberculosis named rifampicin, a device called
GeneXpert MTB/RIF has been created that can provide faster results with accuracy, yet a bit
29
2.6.4 Detection of HIV and Hepatitis C
Fig. 4c shows device for testing HIV and Hepatitis C. A diagnostic kit has been developed by a
Canadian company named MedMira [4] and reported to be an effective device to test for both
devices over the internet [20]. Since the accessibility of internet is extensively widespread all
over the world, especially the current scenario in Bangladesh is massive despite being a
developing country. One of the most impressive outcomes of IoT based technology has been
reported in the field of medicine and diagnostics. Applications of IoT are widespread throughout
medical science, among them an overall monitoring system of patients has been well recognized.
30
2.7.1 IoT based monitoring system
A system has been proposed by Islam, Rahaman and Islam where they majorly focused on
monitoring the fundamental signs of all types of patients and the patients’ room environment
[21]. This IoT based healthcare system is designed to monitor the pulse and body temperature of
patients along with room humidity, total level of carbon dioxide and carbon monoxide via
sensors while transmitting the received data through Wi-fi (wireless fidelity) that enable the
medical staffs and doctors to optimize the data from the server. The developed system also
provides a solution for the problem of maintaining a single database of patients in hospitals using
a web server, apart from the personalization of critical health-related criteria [21]. This system
body temperature sensor (LM35) and room temperature sensor (DHT11) [21].
31
Figure 5a: Hardware components for the healthcare monitoring system. a) ESP32 processor, b) heart beat
sensor, c) body temperature sensor, d) room temperature sensor, e) CO sensor, f) CO2 sensor [21].
Fig. 5a shows the hardware components of healthcare monitoring systems. Here (a) is ESP32 processor,
(b) is heart beat sensor, (c) is body temperature sensor, (d) is room temperature sensor, (e) is CO sensor
and, (f) is CO2 sensor [21].
ESP32. This provides a complete Linux system at a very low price on a small platform. Using
GPIO pins, ESP32 links system sensors and actuators. In the healthcare system, ESP32 and IoT
combine to be a modern technology for innovation. With integrated antenna switches, RF-balun,
control amplification, low-noise amplifier and filters as well as power management modules, the
ESP32 is heavily designed. It may operate as a full stand-alone device or as a slave to a host
MCU, minimizing overhead interaction within the main processor of the application. The EPS32
32
can communicate through its SPI/SDIO or I2C/UART interfaces with other Wi-Fi and Bluetooth
devices. [21].
created. It tests the change in the volume of blood in the organ of someone that causes the
intensity of light to pass through that organ. In systems where the heart pulse rate has to be
monitored, the timing of the pulses is more crucial. The heartbeat rate determines the blood
volume distribution, and when the light is absorbed by the blood, the signal pulses are equivalent
temperature circuit that is linearly proportional to the centigrade temperature. The LM35 has a
33
view point over the Kelvin's linear temperature sensors, as a practical centigrade scaling does not
allow the user to remove the enormous constant voltage from the display. [21].
frequently. The sensor comes equipped with a dedicated NTC temperature measurement device
and an 8-bit microcontroller to process temperature and humidity values in series. The sensor is
Because of its high sensitivity and fast reaction time, measurements can be taken easily. T he
benzene, smoke, and CO2, as well as for testing air quality control systems. The MQ-135 sensor
module comes with a digital pin that allows this sensor to operate even without a microcontroller
and is useful for precise gas detection. Analog pins are used to measure the gases in PPM. The
analog pin is operated by TTL and operates at 5 V and hence it can be used with most modern
microcontrollers. [21]..
34
2.7.1.7 ECG Sensor
Fig. 5c shows the ECG Sensor. An ECG Sensor with disposal electrodes attaches directly to the
chest to detect every heartbeat. The electrodes of ECG sensor will convert heart beat to electric
signal. ECG sensor is very light weight, slim and accurately to measures continuous heart beat
and shows data rate of heart beat. The AD8232 is a little chip used to measure the electrical
activity of the heart. The electrical activity can be charted as an ECG or Electrocardiogram.
Fig. 5d shows taxonomy of the reviewed smart health monitoring systems listed by a review
paper of all the systems of IoT in health monitoring applications prioritize of hardware
components, that is, which components were used more than others. All systems have therefore
35
been classified into three distinct categories: (1) Sensor-based health monitoring systems, (2)
Mobile and IoT technologies have been serving continuously in solving the problem of heart
diseases and patients care. It is feasible to monitor the vital functions of human no matter where
they are and what they are doing with IoT based diagnostic techniques. In addition, the data
collected can be sent with low-cost to the remote physicians, which means that these specialists
are constantly and in real-time aware of the physical state of patients. [40]
36
Fig. 5e shows the prototype of heart disease monitoring system. An IoT-based mobile
monitoring device for heart disease patients has been proposed. This device is designed to
continuously track the physical signs of patients such as blood pressure, ECG, SpO2, as well as
important environmental indicators, and provide four different modes of data transmission that
balance the need for health care and the need for connectivity and computing resources [40].
Due to the popularity of smart phones and the openness of the Android platform, an Android
smartphone is used in this proposed system as the connector. In Java, an application is further
implemented on a smart phone. This application is responsible for receiving and storing, the
monitored data through Bluetooth from the sensing devices and for transmitting the required data
according to the various modes of operation. A web-based program for doctors to query tracked
data is introduced on the remote server side. Figure 5f represents some devices used in the
system and an example of monitoring GUI (graphics user interface) at the doctor side. Figure 5f
(a) depicts a sensing device for SpO2 and pulse rate. 5d (b) shows an Android smart phone as the
connector of this system and 5f f(c) shows the monitoring GUI of patients’ SpO2 and pulse rate
for remote practitioners [40]. This system is significant for a developing country like Bangladesh
37
where a large number of people are rural residents and often suffer from intermediate to severe
For monitoring diabetic patients, tracking shifts in blood glucose levels is important. Strict
regulation of glucose decreases the risk of hypoglycemia, which can be potentially fatal if not
The sensor-based health monitoring system collects information through an electronic data signal
about the patient's health status and informs the patient via an audio alarm. ECG, temperature
and pulse rate sensors are commonly used amongst the different types of sensors. Body
temperature sensors (Max 30205), pulse rate sensors, temperature sensors have been used by
most health monitoring systems (BME 680) [23]. Other sensors such as humidity sensors, RFID
sensors, biochemical detection sensors such as glucometers, body position sensors and CO2
sensors are also used in many health monitoring systems. It is also proposed to develop a wrist
wearable body area network (WBAN) architecture for the healthcare system using cloud
computing [24]. In both industrial and home settings, the system performs. The established
device, however, is wire-connected. Having the computer wireless would improve the system's
versatility. A smart intensive care unit (ICU) system based on radio-frequency identification
(RFID) that collects medical data in a real-time system was shown [25]. The device has 3 layers.
38
They are the physical layer, logical layer, and application layer. The physical layer includes the
sensors used for the purpose of collecting data. The data obtained by the sensors in the physical
layer is analyzed in the logical layer. The logical layer manages the inter-sensor interactions
and media access management tasks. Based on the processed data generated by the logical layer,
One of the world's most powerful tools is smartphones [22]. Normally, a smartphone holds 14
types of sensors [26] and even more sensors will be added in the future. The wireless sensor,
Proximity, GPS tracker, Camera, NFC-near field sensor are some of the sensors included in a
smartphone and they are also commonly used in the implementation of health monitoring
systems [27]. A smartphone's mass storage facility is another primary benefit of a smartphone.
Modern smartphones can store data of the patient in primary storage easily. For data streaming,
device information management, and easy interaction, smartphones based on Android can be
used. A mobile voice health monitoring system has been developed using a smartphone that uses
the accelerometer sensor [28]. A miniature accelerometer is used as a voice sensor in the
reviewed system, and the smartphone is used as a medium for data acquisition. The device is
positioned around the patient's neck. Although the frame-based vocal parameters are used by this
device, the raw accelerometer data could also be used for monitoring purposes. There is a further
proposal for a multi-lead ECG health monitoring device based on a smartphone. [29]. Seven lead
real-time ECG is used for the acquisition of the signal in this device. Up to 500 Hz is the
sampling rate. It is very difficult for doctors to detect anomalies because of the large amount of
39
ECG data. So, it uses an automated warning system. When the device detects irregular ECG
data, the alarm turns on. Another mobile phone-based wireless health service system is
developed [30]. The system is designed for family health treatment. The contact between the
server and the Android terminal is processed by the data communication system. A continuous
heart rate monitoring system has been introduced that uses wireless signals for transmitting the
heartbeat of an individual to a smartphone [31]. Patients can quickly get their real-time heart rate
information from the computer monitor as the device monitors the continuous heart rate.
An alternative and recognized way for optical detection in the field of medical diagnostics in
large areas with low resource settings is charge coupled device (CCD) cameras which is simpler
and cost-sensitive [3]. The device is able to detect multi-wavelength fluorescence and can be
used for fabricating microscopes, sensitive plate readers, lens-less optical detectors for medical
diagnostics [3].
In health monitoring systems, microcontrollers are the most popular and used devices worldwide
[32]. For the fast processing of raw sensor data, MCU's are very useful. To interface between the
Fig. 5f shows the block diagram of Arduino based health monitoring system [34]. An Arduino-
All the data provided from sensors is in analog form. Data is sent to the board of Arduino Uno.
40
Via an integrated analog to digital converter, the analog values obtained are converted to digital.
By Bluetooth, the digital values are passed to a smartphone. A Bluetooth module used by the
Figure 5f: Block diagram of Arduino based health monitoring system [34].
An m-Health solution for diabetes patients is introduced which is a home environment system
[35]. This system allows a patient with diabetes to monitor his/her health status, physical
An IoT based smart health monitoring technique is developed where the system application is
divided into three layers [36]. They are the detecting layer, the application layer, and
the transportation layer. A DS18B20 sensor is being used for the detection of body temperature
in the detection section and a pulse sensor is being used for the detection of a pulse. In the
41
transportation layer, through the Wi-Fi module and Ethernet shield, the information is uploaded
from Arduino to the cloud. The application layer eventually gathers data from the server.
Another system is developed to detect the abnormalities in ECG and transmits the signal via
Bluetooth [37]. It is a low-cost device and the three ECG signals are sent to a circuit and finally
presented in an oscilloscope. Using the Wi-Fi or IR modules might overcome the range
limitation of Bluetooth module. An IoT based patient monitoring system is proposed where the
Raspberry Pi is the main logic unit of the system [38]. Sensor signals are sent via an amplifier
circuit and signal conditioning unit to the raspberry pi (SCU). The Raspberry Pi data can be
accessed from any area of the world through the internet. Even so, the system only works in
indoor environments. Another system of a smart home and heartbeat monitoring system using a
their metabolic regulation and quality of life, aimed at enhancing self-management skills.
Important progress has been made by artificial intelligence (AI) technologies in converting
accessible genetic data and clinical information into important knowledge. [43]
The long-term chronic condition is known as diabetes mellitus (DM). The number of people
worldwide suffering from diabetes has reached 451 million, according to recently released data
from the International Diabetes Federation. [44] Therefore, diabetes diagnosis and control is
highly important and knowledge about diabetes is an essential part. Increasing AI-based tools for
42
diabetes healthcare have been built over the past several years [45]. More versatile and scholarly
access to skills and information on different aspects of diabetes self-management is provided for
patients, including diabetes prevention, lifestyle and dietary advice, exercise, insulin injection,
implementation of AI is becoming extensive and has yielded tremendous benefits. The main
benefits involve broad coverage, low cost, high efficiency, portability, diversity, and
"productivity" at the service end. At the same time, AI improves the route towards full-course,
individualized, and intelligent management of chronic diseases. Some of the most representative
and active reports are from systems of medical experts. The expert systems will make inferences
and decisions based on broad bodies of clinical expertise and experience gained by medical
experts by the use of AI technologies, thus emulating the decision-making processes used by
AI technologies have been increasingly incorporated with medical technology, such as machine
learning and data mining, and are eventually becoming key elements of medical growth. In
disease detection and prediction, surgical robotics, image recognition, virtual medical assistance,
pharmaceutical research and development, and health management, AI offers specific benefits
[48-53]. AI technologies have been widely used in diabetes prediction, dietary and exercise
guidance, insulin injection guidance, complication control, and self-management in the area of
diabetes education.
43
2.9.2 Detection of Hypoglycemia from Raw ECG Signal
Glucose testing has historically been done by a needle prick or capillary blood glucose test.
Doctors can monitor blood sugar levels through advanced artificial intelligence (AI) applications,
using a few heartbeats of raw ECG signals captured through wearable sensors. [42] Using a non-
invasive wearable system, a new technology was developed to detect low glucose levels via an
electrocardiogram (ECG). The device is related to the new artificial intelligence that can detect
The electrophysiology of the heart is influenced by hypoglycemia, and since it has slightly
different effects on the heart of each person, an AI system makes it possible to control glucose
levels in a highly customized way. [55] Currently, continuous glucose monitors (CGM) are used
for hypoglycemia diagnosis in patients who are needed to undergo continuous monitoring of
blood sugar levels. However, when blood sugar levels become down, CGM uses an invasive
needle that sends alarms. With invasive finger-prick blood checks, the system needs calibration
two times a day. As well as CGMs, the novel artificial intelligence device works, offering data
on glucose levels without invasive needles. In two pilot studies with healthy participants, the
researchers tested the system and found that the average specificity and sensitivity was 82
percent in the detection of hypoglycemia, which works much the same as the current CGM
process. [42]
44
Both type I and type II diabetic patients need to check blood glucose level routinely. In order to
calibrate or change the display, the currently accepted models require users to measure a drop of
blood twice daily. Many people are discouraged by the discomfort of finger sticks and the cost of
checking supplies from keeping close tabs on their blood sugar, which is important to control
insulin usage and change what they eat. [56] The recently developed and approved Food and
Drug Administration (FDA) FreeStyle Libre Flash Glucose Monitoring System [57] is a
complete setup that uses a small sensor attached to the upper arm. To see the present amount of
blood sugar and shifts over the past eight hours, patients wave a reader device over it.
Figure 6: A device from “Abbott” that continuously monitors diabetics’ blood sugar levels without
Fig. 6 shows a device from “Abbott”. This device can continuously monitor diabetics’ blood
45
For up to 10 days, both day and night, the system provides real-time glucose readings. Glucose
levels can also be read by the sensor through clothing, making monitoring discreet and easy.
[57].
The system provides three critical pieces of data with each scan:
The touch-screen reader often keeps data for up to 90 days, enabling individuals to control their
The FreeStyle Libre CGM eliminates finger sticks for calibration, which translates into less
manual effort [58]. People with diabetes can scan their CGM (a painless procedure) instead of
testing finger sticks several times per day to get real-time glucose readings to be used for insulin
dosing, meal preparation, etc. In addition to improved health control for daily diabetes, large
volumes of data and trend analyses can also help diabetic patients to minimize their risk of low
blood sugar and assist them in assessing the dosage of insulin and meal preparation. [58]
46
CHAPTER 3
RESULTS AND
ANALYSIS
47
Result and Analysis
1 This Blood sugar test and diabetes Test strips, IoT and AI Cost-
paper management, pregnancy test, tests based-devices/systems, effective
for cardiac diseases, viral infections microfluidics,
and critical conditions like cancer, biosensors.
organ transplantation.
48
2 Ref [47] Parkinson’s disease AI based device Cost-
effective
49
CHAPTER 3
CONCLUSION AND
FUTURE WORK
50
Conclusion and Future work
Since medical diagnostic procedure is considered to be a critical element for effective healthcare
and medical treatment, advanced tools and devices used in this concern need to be cost-effective,
affordable and sustainable for every patient, especially for people living in middle-income and
low-income countries. These poorly resourced countries require robust, low-cost, easy-to-use
and portable diagnostic devices that are compatible to meet every diverse or adverse medical
51
condition [3]. At the same time, it would be time-saving and the medical staffs would be able to
learn the proper use of the diagnostic methods or devices. Low-cost and alternatively used
medical diagnostics and devices function with proper specificity and sensitivity without dropping
the quality of treatment or diagnosis as well as maintaining the overall health condition of the
patients. Future study and advancement specially in lower income regions of the world, will
BIBLIOGRAPHY
52
[1] D. Mabey, R. Peeling, A. Ustianowski et al., “Diagnostics for the developing world,” Nat
[2] R. Jessup, “Identifying alternative models of healthcare service delivery to inform health
https://bmjopen.bmj.com/content/bmjopen/10/3/e036112.full.pdf.
[3] J. Balsam, “Low-cost technologies for medical diagnostics in low-resource settings,” Ebook,
2013.
[4] "7 Low-cost devices for diagnosing diseases in poor countries | Openmind," Openmind,
2015. https://www.bbvaopenmind.com/en/technology/innovation/low-cost-devices-for-
diagnosing-diseases-in-poor-countries/.
[5] T. Marigrace, "Cheap vs. expensive pregnancy tests: Differences, pros and cons," What To
cheap-expensive-pregnancy-tests-difference/.
53
[6] “The 8 best pregnancy tests of 2021,” Verywell Family, 2021.
https://www.verywellfamily.com/best- pregnancy-tests-4173458.
[7] G. Marrazza, “Biosensors and Nanotechnology,” 2018. [online] Google Books. Available at:
https://books.google.com.bd/books?
hl=en&lr=&id=XKZFDwAAQBAJ&oi=fnd&pg=PA95&dq=related:3oah8sdgLZaUgM:sch
olar.google.com/&ots=bdG35HLCdj&sig=uAf7ieV-
1y7be_u4_OzUpU8uydI&redir_esc=y#v=onepage&q&f=false
[8] P. Mehrotra, “Biosensors and their applications - A review,” Journal of Oral Biology and
https://doi.org/10.1016/j.jobcr.2015.12.002.
[9] S. Yeşeren, E. Özgecan, Ü. Serhat and D. Adil, "An Alternative Medical Diagnosis Method:
https://doi.org/10.3390/bios9020065
10.1016/j.bios.2018.02.018.
[11] P.D. Tam, N. Van Hieu, N.D. Chien, A.T. Le, M. A. Tuan, “DNA sensor development
based on multi-wall carbon nanotubes for label-free influenza virus (type A) detection,” J.
[12] H. Ilkhani, S. Farhad, “A novel electrochemical DNA biosensor for Ebola virus detection,”
54
[13] A. A. Yanik., M. Huang, O. Kamohara, A. Artar, T.W. Geisbert, J.H. Connor, H. Altug,
“An optofluidic nanoplasmonic biosensor for direct detection of live viruses from biological
[14] O. Parkash, R. Shueb, “Diagnosis of dengue infection using conventional and biosensor
based dengue diagnostic test using surface plasmon resonance biosensor,” Sci. Rep.
[16] R. Madhav and M. Raje, "Microfluidics in disease diagnosis: Past, present, and future-An
https://www.imedpub.com/abstract/microfluidics-in-disease-diagnosis-past-present-and-
futurean-overview-24543.html.
[17] Book: Lab on chip technology: Bimolecular separation and analysis, Editor-K. E. Herold,
https://link.springer.com/article/10.1007/s41403-016-0009-1#citeas
[19] S. D. Garimella, "Cheap and easy-to-use diagnostic tests to detect disease biomarkers,
tests-to-detect-disease-biomarkers-including-cancer.
[20] I. J. Hoque, M. S. Navid, R. B. Reza and M. I. Mahbub, “IoT based patient monitoring
system,” 2018.
55
http://dspace.bracu.ac.bd/xmlui/bitstream/handle/
10361/10106/12321021%2C12301040%2C14121084%2C12301048_CSE.pdf?
sequence=1&isAllowed=y.
https://doi.org/10.1007/s42979-020-00195-y
[23] M. Tsakalakis and N. G. Borbakis, “Health care sensor-based systems for point of care
Conference of the IEEE Engineering in Medicine and Biology Society, pp. 6266- 6269,
2014. http://dx.doi.org/10.1109/EMBC.2014.6945061
[24] F. Wu, T. Wu and M. R. Yuce, “Design and implementation of a wearable sensor network
system for IoT-connected safety and health applications,” 2019 IEEE 5th World Forum on
IoT.2019.8767280
[25] A. S. Ahouandjinou, K. Assogba and C. Motamed, “Smart and pervasive ICU based-IoT for
http://dx.doi.org/10.1109/BIOSMART.2016.7835599
56
[26] M. Grossi, “A sensor-centric survey on the development of smartphone measurement and
https://doi.org/10.1016/j.measurement.2018.12.014
[27] M. A. Kumar and Y. R. Sekhar, “Android based health care monitoring system,”
health monitoring using a wearable accelerometer sensor and a smartphone platform,” IEEE
http://dx.doi.org/10.1109/TBME.2012.2207896
[29] H. Gao, X. Duan, X. Guo, A. Huang and B. Jiao, “Design and tests of a smartphones-based
multi-lead ECG monitoring system,” 35th Annual International Conference of the IEEE
http://dx.doi.org/10.1109/EMBC.2013.6609989
[30] X. Kong, B. Fan, W. Nie and Y. Ding, “Design on mobile health service system based on
http://dx.doi.org/10.1109/IMCEC.2016.7867504
[31] J. Turner, C. Zellner, T. Khan and K. Yelamarthi, “Continuous heart rate monitoring using
57
[32] J. Sütő, S. Oniga and I. Orha, “Microcontroller based health monitoring system,” IEEE 19th
International Symposium for Design and Technology in Electronic Packaging (SIITME), pp.
[33] S. Trivedi and A. N. Cheeran, “Android based health parameter monitoring,” International
Conference on Intelligent Computing and Control Systems (ICICCS), pp. 1145-1149, 2017.
http://dx.doi.org/10.1109/ICCONS.2017.8250646
[34] “Smart health care monitoring system based on IoT,” Project Hub, 2019.
https://create.arduino.cc/projecthub/156471/smart-health-care-monitoring-system-based-on-
iot-f559b
“Prototyping Arduino and Android based m-health solution for diabetes mellitus patient,”
http://dx.doi.org/10.1109/ISMAC.2017.8058240
ECG via Bluetooth,” IEEE International Conference on Smart Cloud (SmartCloud), pp. 41-
[38] R. Kumar and M. P. Rajasekaran, “An IoT based patient monitoring system using raspberry
58
[39] M. R. Desai and S. Toravi, “A smart sensor interface for smart homes and heart beat
http://dx.doi.org/10.1109/CTCEEC.2017.8455124
[40] “The IoT-based mobile monitoring system for patients with heart diseases,”
DiscussionPapers/11.pdf
[41] I. J. Hoque, M. S. Navid, R. B. Reza and M. I. Mahbub, “IoT based patient monitoring
system,” 2018.
http://dspace.bracu.ac.bd/xmlui/bitstream/handle/
10361/10106/12321021%2C12301040%2C14121084%2C12301048_CSE.pdf?
sequence=1&isAllowed=y
[42] A. B. B. Laguipo, “AI replaces prick test to detect blood sugar levels,” [online] News-
prick-test-to-detect-blood-sugar-levels.aspx#:~:text=A%20team%20of%20researchers
%20at,evens%20from%20raw%20ECG%20signals.
[43] J. Li, J. Huang, L. Zheng and X. Li, “Application of artificial intelligence in diabetes
education and management: present status and promising prospect,” Frontiers in public
https://www.frontiersin.org/articles/10.3389/fpubh.2020.00173/full
diabetes atlas: global estimates of diabetes prevalence for 2017 and projections for 2045,”
59
[45] I. Contreras and J. Vehi, “Artificial intelligence for diabetes management and decision
support: literature review,” J Med Internet Res, 2018. 20:e10775. doi: 10.2196/10775
lung disease: current status and future potential,” Curr Opin Pulm Med, 24:117–23, 2018.
doi: 10.1097/MCP.0000000000000459
10.1016/j.amjsurg.2004.08.025
[50] R. Gargeya and T. Leng, “Automated identification of diabetic retinopathy using deep
[51] T. B. Talbot, K. Sagae, B. John and A. A. Rizzo, “Sorting out the virtual patient: how to
exploit artificial intelligence, game technology and sound educational practices to create
engaging role-playing simulations,” Int J Gaming Comp Med Simul, 2012. 4:1–19. doi:
10.4018/jgcms.2012070101
[52] W. Duch, K. Swaminathan and J. Meller, “Artificial intelligence approaches for rational
drug design and discovery,” Curr Pharm Design, 2007. 13:1497–508. doi:
10.2174/138161207780765954
60
[53] M. Furmankiewicz, A. Sołtysik-Piorunkiewicz and P. Ziuzianski, “Artificial intelligence
systems for knowledge management in e-health: the study of intelligent software agents,” In:
[54] M. Porumb, S. Stranges, A. Pescape and L. Pecchia, “Precision Medicine and Artificial
Intelligence: A Pilot Study on Deep Learning for Hypoglycemic Events Detection based on
[55] A. Pashler, “Could AI replace the finger prick blood sugar test?” Medical News Today.
the-finger-prick-glucose-test
[56] “FDA approves first blood sugar monitor without finger pricks” – STAT, 2017. [online]
without-finger-pricks/
[57] “Revolutionizing CGM with FreeStyle Libre,” Abbott. [online] Available at:
https://www.abbott.com/corpnewsroom/diabetes-care/revolutionizing-cgm-with-freestyle-
libre.html
[58] B. Cervoni, “FreeStyle Libre: A Glucose Meter Without a Finger Prick,”. [online] Verywell
meter-without-a-finger-prick-4154266#citation-5>
61