review paper about intellectual property information security

1

1

Abstract—With the advent of the technological world, the

technology is getting more and more advanced day-by-

day. Artificial Intelligence (AI) can possibly affect pretty

much every part of medical care, from identification to

forecast and anticipation. The appropriation of new

advances in medical services, nonetheless, slacks far

behind the rise of new advances. An elementary

understanding of developing Artificial Intelligence

proceedings can be basic though wellbeing couldn't care

less experts. These advancements incorporate master

frameworks, mechanical cycle robotization, regular

language preparing, Artificial Intelligence, and deepest

understanding. In the research article, different

technologies have been derived for the detection of

different health diseases. First of all, background

knowledge has been taken under consideration. After

that, diseases like Diabetes, Alzheimer’s disease and

health disease have been discussed. It has been evaluated

that technologies are providing extremely efficient results

with higher level of accuracy which shows that the

discussed technologies are contributing at their best level.

The proposed methods for the discussed diseases in

different research articles have also been evaluated and

highlighted. Every technology has its own benefits. The

proposed article illustrate that how Artificial Intelligence

is contributing in healthcare department and in the

detection of different health diseases.

Index Terms— Expert System, Decision making

Support, Artificial Intelligence, Clinical Decision Support

System, Magnetic Resonance Imaging (MRI), Alzheimer’s

Disease

I. INTRODUCTION

A. Artificial Intelligence

Artificial intelligence is how different machines exhibit

intelligence compared to natural intelligence used by different

humans and animals. In simple words, the theory related to

the growth of computer systems to perform tasks usually

needs human intelligence, for instance, visual perceptions,

decision making, translation of languages, and speed

recognition (Fei Jang, 2017). It is known as a digital

computer's capability or called a computer-controlled robot to

execute tasks usually connected with intelligence. This term

AI is applied to those projects related to developing systems

bestowed with factors of human or intellectual processes, for

example, the ability to reason, generalizing, abstracting, learn

from past experiences, or to discover meaning. In the 1940s,

digital computers evolved and came into existence, so from

1940, since now, computers are designed to perform

complicated and complex tasks, for instance, working on

advanced proofs and theorems from mathematical portions as

well as playing chess. Despite continued advances in the

speed of computer processing and memory capacity still, there

is a gap in programming that they cannot be as flexible as

human beings. This system is not proficient in broader

Expert System (AI) for Decision making Support

Abdullah Alshathri 442105936

College of Computer and Information Sciences

King Saud University, Riyadh, Saudi Arabia

[email protected]

2

domains or tasks requiring a vast level of information and

knowledge (Kun-Hsing Yu, 2018).

On the contrary, some programs have gathered the program's

level of exports to the execution of specific tasks; hence,

artificial intelligence is found in applications playing multiple

medical diagnosis roles, search engines, or handwritten plus

voice identification.

However, it has been studied that all human behavior is

known as intelligence, but psychologists do not consider

human intelligence by only one trait but with multiple

abilities. Artificial Intelligence mainly focuses on learning,

reasoning, solving problems, anticipations, and utilizing

different languages.

In the learning process, the simple form is trial and error. It

merely means that memorizing several items and process are

called rote learning in artificial intelligence. This term is

comparatively easy to learn and understand. Moving further,

developing and executing more challenging tasks are known

as generalizations. It involves using previous past experiences

to new and analogous situations (Yaping Zang, 2015). It can

be explained through the example that the program used to

work and understand past tense would be difficult for that

program to understand English in present tenses. The

reasoning is also an essential characteristic of artificial

intelligence, which works based on inferences. Inferences are

always of two types, Deductive and Inductive. In the

deductive method, the reasoning is made based on the

hypothetical method and the base of illustrations and

examples.

On the other hand, in the inductive method, the reasoning is

based on facts and figures. Problem-solving is a factor that

works on the systemic approach through a range of specific

actions. Problem-solving is always done to achieve

predefined targets and objectives (Thomas M, 2019). In

perception, the environment is examined through different

means of certain sensory parts, real or artificial. The last

factor of artificial intelligence is language. It is a system that

depicts meaning by old conventions. It is also said that

language is required to be confined in terms of speaking

words. The traffic signal is a clear example of language under

artificial intelligence. A productive language can execute a

diverse variety of sentences and words.

However, artificial intelligence is contributing to the health

sector of different countries positively. Many countries are

availing benefits by using artificial intelligence to diagnose

diseases and even then, in the treatment of those diseases.

This study is conducted to find the impact of artificial

intelligence in the health sector (Sandeep Reddy, 2019).

B. Artificial Intelligence in Healthcare

The paper is about the artificial intelligence in health care

departments and with the diagnosing diabetes and Alzheimer.

AI in health department has enormously populated in last

decades and is peak potential to deliver the control and

establish proper systems to diagnose and propose treatments

for health care and severe diseases like diabetes and

Alzheimer (Trishan Panch, 2018). With Artificial Intelligence

and its contribution in health care field plots so much

information but yet there is a need to explore more of artificial

intelligence helping in health care as well as there is a need to

grow in artificial intelligence field. To study the health

professions, evolve in this field and the contribution of

artificial intelligence will show remarkable changes in field.

Artificial intelligence creates and stimulate the human-like

intelligent behavior in complex machines (Brian Wahl, 2018).

It is used in health cares and most importantly it used mostly

in human health care, like diagnosing diseases, in process of

treatment, prognosis and predictive health service.

Artificial Intelligence needed more exploration and work to

make use of its various fields related to human, like chemical

engineering, management and human health aspects in severe

diseases. It’s in early development and need more focus with

respect to clinical readiness. An artificial intelligence

followed in major experimental and conceptual health care

evolutions and domains. In future, almost every kind of

practitioners from specific analyst to coroner by applying

Artificial intelligence theories and constitute a great part in

teach (Shizhen Wang, 2019). The study aims to give brief

description about the artificial intelligence usage in health

care with severe diseases like diabetes, Alzheimer and heart

problems. The description is important as it improves the part

3

of literature and benefit in using most part in literature in

future. In previous research there is a vast information present

about radiology.

Which is used previously but today the concept of radiology

is transformed in to artificial intelligence and automation.

According to research the professions will be impacted

differently for different fields or disciplines like discipline

bounded quarterly by reliance on interconnection, task should

be repetitive and data reliance with creativity level should be

appropriate (Arash Shaban-Nejad, 2018). The dermatology

and radiology varying degrees more influence on profession

but varying less in case of profession as dentistry thus the use

of artificial intelligence being predicted to be useful for the

fields in which more complexity is being seen. Online

predicted the upcoming health hazards and pro-actively

perform functions to overcome disease.

The predictive record gets through social media or online

media electronic health care records. The control of disease is

possible by geocoding the information regarding the disease

spread and help epidemiologists to accurately monitor and

control the severe disease before it spread out rapidly.

Artificial intelligence used in various fields like robotics,

image and voice recognition and natural language to

processing expert system. With its growing capabilities and

dynamic approaches and broad rapidly growing techniques

with upgrading system it is widely used in medicine field

early in 1950 when physicians successfully diagnose the

improvement in treatment due to computer aided program. Al

categorize in to two types, the capability of representing

human mind and perform all intellectual task to perform

human task.

II. LITERATURE REVIEW

Expert system is known as knowledge-based system. The ES

system is said to be an expert system which is very competent

to resolve specific problems like an expert level. The creation

of expert system is also named as knowledge engineering.

Expert system comprises of two main prime components

(David Wiljer, 2019). Just like Knowledge base and reasoning

engine. The knowledge base is known as the expert system

which typically involve the knowledge-based system, include

the basic knowledge of specific need or problem. And the

reasoning engine is known as constituting the complex rules

just like if-then statements and include information which is

incomplete and uncertain. The fuzzy logic is being developed

full of uncertainty and probability which is shown under

mathematical principles. This indicates the progression of

expert system in just few years. The developing nature of

fuzzy logic make it more specific and help to deduce the

uncertainty and the way human can approach to solve difficult

constituting excessive uncertainty. In South Africa logic used

in diagnosing chronic condition is fuzzy logic and diagnose

steps to minimize the chronic disease, cholera outbreaks

condition to normalize with fuzzy logics.

Application of Artificial Intelligence: Machine Learning

Artificial Intelligence include one of the applications named

Machine Learning. The data automation and data analysis

include which perform by using algorithm which conclude

different patterns in return master from them. Machine

learning categorize in to three forms like supervised learning,

with unsupervised learning and reinforcement learning. So,

the first type of machine learning is to identify the training

data. With this in contrast the applications of learning carry

ways to find the patterns that comprises data. The next level

of supervised learning is reinforcement learning in which the

reward and basic punishment are given when application

interacts with complex environment. When you talk about

machine learning that is unsupervised machine learning

relating with data mining and includes to exclude the identify

patterns in huge datasets. Like the disease used to learn and

creating logic to create machine learning in order to diagnose

the treatment and constitute various treatment but anyhow

diabetes is long lasting disease, and create full long-lasting

effects on lungs.

A. Diabetes

It considers to be long-lasting disease and effects on limbs and

vital organs in body. The ES used to identify the issue and

create such techniques or methods or treatments to control the

disease, detect it and manage to create tools and this all be

done with the help of physicians. According to Diabetes

4

Research Center, the early diagnosis of patients can prevent

them from 80% complications of type II diabetes. (Nesreen

Samer El_Jerjawi, 2018) The two types of diabetes type I and

type II. The type I diabetes refers to insulin dependent and

second type is insulin deficiency. (Mercedes Rigla, 2017) The

complications in diabetes further categorize in to two main

types, which are vascular and nonvascular complications of

diabetes. Vascular include micro vascular diseases such as eye

disease, neuropathy and macro vascular refers to coronary

disease and peripheral vascular disease. Whereas non vascular

complications lead to gastro paresis, sexual dysfunction and

skin disease.

B. Alzheimer

The most curious and integral part of human body is brain.

The most important elements of the human brain incorporate

reasoning, thinking, responsibility for coordination and parity

meaning voluntary movements, coordination and balance,

executive planning, language, memory and learning,

mathematical logic, and emotional responses. As the brain is

the center of the nervous system any abnormal conduct or

working of it might cause all out breakdown of the whole-

body functionalities. Such an abnormal conduct may cause

Alzheimer's disease. Some 4.5 million Americans have

Alzheimer's disease (AD), the eighth top reason for death in

2001. Principally older people are victims of AD, about 12.8

% of those aged more than 65 years. 35 to 40 % of those aged

more than 80 years are affected (K.S.Biju, 2017).

It causes issues with thinking, memory and behavior. It occurs

mostly in older adults about 65. Presently there is no antidote

or treatment available for this disease as of today. But

Researchers and scientists are making an effort to find

treatment for this disease which can help ease pain of the

patients.

The medical imaging technique Magnetic Resonance imaging

used basically in radiology to form pictures of the

physiological processes of the body and anatomy. MRI

scanners use strong magnetic fields, magnetic field gradients,

and radio waves to generate images of the organs in the body.

Segmentation of the MRI brain image is commonly used

technique for measuring and visualizing the brain's cellular

structures in a meaningful and easiest way to analyze the

various types of diseases such as Alzheimer's disease,

dementia and Brain tumor. Segmentation techniques are

various which are available today and, in this study, it is

described in this literature part. In general, the following

categories in which segmentation is divided: Histogram based

technique (Chinnu, 2013), Threshold based technique (Priya,

Segmentation of Brain Tissue in MR Brain Image using

Wavelet Based Image Fusion with Clustering, 2013), hybrid

technique (Palanisamy, 2010), Edge based technique (Patil,

2012), Region based technique (Jaafar, 2011), Cluster based

technique (P.Kalavathi, 2015) and Classification based

technique. In order to detect the AD, a method presented in

(Kalavathi, 2017) segmentation of WM and GM from non-

AD and ADl MRI brain scans Contour based brain

segmentation method (CBSM), Fast Fuzzy C Means (FFCM)

is used. For giving a real set (Toro, 2018) of neighborhoods

for the histone-calculation method which helps identify

Alzheimer’s disease, an overly segmented process performed

using average volume for WM, GM and cerebrospinal fluid.

Another paper using the WM and GM technique extracted (Y-

D. Zhang, 2018) gray matter images, Afterwards, PCA were

put in for feature extraction (Kalavathi Palanisamy, 2017).

Total principal components (PC) that were extracted from

data of 3D MRI scan using singular value decomposition

(SVD) algorithm were 20. A kernel support vector machine

decision tree was built which helped in binary classification

of Alzheimer’s disease and non- Alzheimer’s disease MRI

images. For segmentation of MRI brain images an advanced

U-NET (Hanane Allioui, 2019) architecture is proposed

which can identify Alzheimer disease and brain tissue damage

and an accurate and exact Alzheimer’s disease detection using

an advanced full neural network in a 2.5D context.

Segmentation side by side with clustering also proved useful.

Fuzzy c-means clustering is an unsupervised clustering

method that segments MRI brain images into clusters

(different regions depending upon type of pixels in MRI) with

similar spectral properties. The brain tissues segmentation

using PSO based clustering techniques was developed to

detect Alzheimer’s disease in MR brain images (Yu-Dong

5

Zhang, 2014). The main idea behind this method is the mixed

use of Gaussian Mixture Model and K-Means Algorithm

(GKA). The MRI brain images will be segmented into WM,

GM and CSF by acquiring GKA method. Morphological

operations are also used in segmentation (Jyothi, 2015) for

that to work properly MRI brain images are taken as input and

morphological operations are applied to detect abnormalities

of brain tissues and cells for the diagnosis of Alzheimer’s

disease. Often one or the other methods from morphological

operations are combined with FMRI or Tomography to

provide complementary information about normal and

abnormal brain function. Edge detection is also used for

object detection.

Segmentation techniques are readily used in detection of

Alzheimer’s disease. The image segmentation plays a vital

role in diagnosing processes and analyzing diseases like

Alzheimer’s disease. A method to detect the brain tissue is

segmentation which is Alzheimer’s disease affected MRI

head scans was proposed. For detection of brain tissues and

cells in the brain MRI images SFCM was proposed in

combination with spatial fuzzy clustering algorithm. Another

Algorithm along with segmentation technique (R. Anitha,

2016) is watershed including highly reserved up to its limits

work to explore disease in an image which is scanned.

Segmentation of brain tissues are very helpful according to

(Kazuhito Sato, 2011) A brain tissue segmentation is done

using two kinds of unsupervised neural networks which are

Fuzzy Adaptive Resonance Theory (ART) and Self-

Organizing Maps (SOMs) and for classification of

Alzheimer’s disease.

C. Health Informatics and Electronic Medical Records

The health information regarding the health care issues like,

severe diseases or chronic disease will provide information to

various treatments and methods to use in treatment of disease.

The electronic media records health information that

describes the acquisition, retrieval, stored information and use

of health care information to positively affect the patient

overall health across interaction with health system. The

EMRs use to identify the resources used to provide patient as

much needed treatment in a specific time. It ensures to provide

the critical information for making sound policies and

program decisions (Karthik Seetharam MD, 2019). The

EMRs beneficial to attain information regarding population

and health information and it is the important source to get the

health information.in this era it is in great use in low resource

settings which explained the potential application of AI

platform improve public health informatics and decision

making. An important growth in AI explained with an

example of establishing Open MRS system which is used

among 15 African countries to make use of EMR system to

and implementation of MER system is known as BORA. This

focus on improving BORA to improve women maternal and

child health problems. In South Africa logic used in

diagnosing chronic condition is fuzzy logic treatment in

mostly rural areas. According to research the use of AI

become useful from many years as it helped to increase the

collection of data and closely relate the critical gap.

D. Cloud Computing

Cloud computing enhance the exploration of AI application

for health care centers. Thus, it refers to multiple functioning

in a system like to store, control and have access and doing

many processes like remote servers rather than single personal

computer or hard drive. Many businesses start to explore

cloud computing because of its phenomenal advantage in IT

system, enhance the validity and reliability with enhancing

ways to save cost (Mohamed Elhoseny, 2018). Previously it

in not accessible and unattainable in low- and middle-income

countries but now it easily accessible and provide companies

provide cloud computing in organization built in rural or

urban areas. The EMRs can be controlled with proper

functioning of privacy with security systems. The cloud

computing makes it enable to use the diverse data related to

public health. The cloud computing main advantage is to

enable the implementation of application that interlinked with

IT building in settings where its existence is rare. The cloud

computing application used in deriving the disease in patient

and aimed to improve the interactive voice response of

telephone calls for managing non-communicable disease in

Honduras. There is so less upgrading of IT building

institutions in our country but the new inventions are

6

effective. The digital health technologies reached to the

advancement level with not very high quality of resources in

high earning population of country are benefiting from

integrating AI into their healthcare ecosystem. According to

research that the AI applications results in estimated cost of

more than $145 billion in healthcare by yearly 2026

approximately in USA. Many other prove transformation with

help of AI approve the benefit of using Artificial intelligence

for countries constituting poor health resources (Giuseppe

Aceto, 2020). Today the data brings out to be vast that this

created a plethora to improve individual and group level

health care by using AI. The development in cloud computing

and substantial investments in health care information and

introduction in mobile health application will increase the

exploration of AI in health care field and approve to be

successfully in the medicine field to results in betterment of

health care problems in world.

E. Mobile Health

The mobile health use cellphone, and radial automation to

acquire fitness care information and the accessibility of cell

phones rapidly increase in less-paying territories has designed

many ways to use automations to reinforce fitness issues. Cell

phone mostly utilized in Community Health Workers for

betterment of the provision of fitness related services with

poverty-stricken setting (Giovanna Sannino, 2019). The tool

of flexible health is currently evaluated and cell phones

utilized to coordinate the fitness issues to patient in lower

income areas at present where recourses are less and physical

cooperation are not practical. Usage of phone calls plus

messages assistance to communicate request hurdles to

inoculation and enhance potentially thoroughly written

through random controlled trials in setting in Kenya and USA.

III. BACKGROUND

A. Advent of Artificial Intelligence (AI)

In today's world, with a lot of advancement in technology and

digitization, AI is also contributing to its best to support the

decision-making process. There are infinite examples of AI

applications that are playing their efficient role in

technological advancements. Does AI have some common

constitutes that elaborate the definition of what artificial

intelligence is? Four concepts explain it well, something that

behave intellectually, according to, 1) situation for

environment and its demand, 2) is it resilient towards the

alternating goals, 3) intellectual towards experience, and 4)

make appropriate decisions towards intuitive and computing

environment.

B. Related Work on Healthcare

Artificial intelligence came into existence in the mid of 1940.

In today's era, it is known as a subfield of computer science-

related to symbolic analysis methods by computer. Alan

Turing designed AI to explain the intelligence of computers

and technology. Artificial Intelligence possesses a

tremendous and positive impact on the health sector. The

primary purpose of artificial intelligence is to use computers

in an efficient way to solve challenges and hurdles in the

healthcare sector (Ezekiel J. Emanuel, 2019). By using

computers, data can be interpreted, which can be received by

identifying chronic diseases, for instance, diabetes,

Alzheimer's well several types of cancers. In the complicated

world of health care management, artificial intelligence is

used to support human medical staff to provide better and

faster services, diagnose problems, and interoperate data to

discover data trends that would help treat specific diseases.

Artificial intelligence has been used in health care to

understand deep learning in the identification of diseases. This

is the area where artificial intelligence has shown promising

success in diagnostics. In the learning of machines and the

radiology sector, artificial intelligence has been used at a vast

level. It has been seen that in performing automotive

administrative related tasks, health AI has performed very

well. It also helped in the reduction of operational costs. By

using artificial intelligence, some hardcore challenges in

health care have been resolved. It is also seen that many

patients, through this process, find health back due to the

diagnoses of chronic diseases (Thomas Davenport, 2019).

When artificial intelligence was implemented in healthcare,

the first time was in late 1970 when Dendral was discovered

7

at Stanford University USA. It was used to guide chemists to

find some not known organic molecules. After that successful

identification, it was used in the health care sector of different

economies. Today, artificial intelligence has brought so much

worth and value in health care management by developing so

fastly and rapidly. It is also anticipated that AI will continue

to achieve success in the health care department in the near

future. It has prominently contributed to the field of diagnosis

(Mccall, 2020). To identify any disease, it needs tons of data

to analyze medical imaging, medical records of patients, and

the history of patients. It is also essential to understand that

the genetics of patients require in the diagnosis of disease. In

the last decade, it has been observed that artificial intelligence

has been more accurate in finding disease and even in

recommending effective treatment—for example, cancer

diagnoses (Marzyeh Ghassemi, 2019).

Simple radiology is not enough for the procurement of illness

i.e. cancer. Artificial intelligence is beneficial in this manner.

AI algorithm has been introduced by South Korea, through

which cancer cells can be identified on X rays images.

Moreover, mammography pictures are used to detect lungs

plus breast cancer. Artificial intelligence has 97 % accuracy

in identifying these lungs and breast cancers in the human

body. Moreover, artificial intelligence has helped in saving

around 3$ billion every year in the health care department.

Moving further, the contribution of artificial intelligence also

has significance in robot surgery. This is the most critical

application in artificial intelligence, specifically in health care

management. This application provides benefits in two ways,

a tremendous amount of money-saving and more successful

and effective surgery. In this surgery, Accenture estimated

that this robot surgery could save around 40$ billion in

healthcare industry per year. Furthermore, artificial

intelligence also helps in managing a large amount of data in

the healthcare industry. While dealing with a large amount of

data, it near impossible to have proficiency in all gigabytes. A

considerable amount of data comprises of billion and trillion

of entries about patients, surgical process and procedures,

drugs, treatments, data related to different researches, and

many more. If this data is required to be used regularly, it

should be stored and managed efficiently. Artificial

intelligence work accordingly to manage this sort of data. It

provides much better and improved medical care to patients

that collecting data and storing it makes it significantly more

comfortable. Big data also helps collect information about

demographics and medical data, for example, lab tests,

medical history and condition, clinical data. However, it also

helps to provide efficient operational information, and more

importantly, it supports the research and development

proficiency. With the guidance of predictive analysis of

efficiency of staff members companies in healthcare can

boycott expensive operational costs. But by using artificial

analysis, big data can help manage workflow and give

effective forms of costs. In terms of research and

development, it can help to work on several tasks of drugs and

clinical trials. It also can find some patients with biological

factors who will indulge in specialized clinical practices

(Rigby, 2019).

It has also been observed from the past that artificial

intelligence has helped find a cure for diseases. More

availability of data can help in providing a cure at the early

stages. Hidden patterns, not known correlated diseases, and

insights can be identified early using artificial intelligence in

health care management. This credit goes to AI as big data can

help anticipate the chances and probabilities of certain

diseases in a specific human body are more or less. By

collecting data on family history and demographics can help

in detecting diseases appropriately and positively. From the

above discussion in background, indeed, it is concluded that

artificial intelligence has numerous healthcare industry

benefits. With time, AI would help achieve success and

growth in the healthcare management at a vast scale.

IV. DISCUSSION

The Artificial Intelligence (AI) has developed extensively

over the most recent 60 years. While there are currently

numerous Artificial Intelligence applications that have been

sent in top level salary nation settings, use in asset helpless

settings remains generally early. With a couple of prominent

special cases, there are restricted models of Artificial

8

Intelligence being utilized in such settings. Notwithstanding,

there are signs that this is evolving (Jarrahi, 2018). A few

prominent gatherings have been gathered lately to talk about

the turn of events furthermore, organization of Artificial

Intelligence applications to decrease neediness and convey an

expansive scope of basic public administrations. We give an

overall review of Artificial Intelligence and how it tends to be

utilized to improve wellbeing results in asset helpless settings

(Wahl, 2018). We additionally depict a portion of the current

moral discussions around quiet security also, safety. In spite

of current difficulties, Artificial Intelligence holds enormous

guarantee for changing the arrangement of medical care

administrations in asset helpless settings (Duan, 2019).

Numerous wellbeing framework leaps in such settings could

be overwhelmed with the utilization of Artificial Intelligence

and other correlative rising innovations. Further examination

furthermore, interests in the advancement of Artificial

Intelligence apparatuses custom fitted to asset helpless

settings will quicken acknowledging of the full capability of

Artificial Intelligence for improving worldwide wellbeing

(Behnam Malmir, 2017).

Many institutions and firms are developing and implementing

tools to accurately find out the problem and are proposing

decision support systems as a substitute or a solution towards

these types of problems (Shikhar Kr. Sarma, 2010). Likewise,

healthcare (Wiljer, 2019) departments are also using these

measures and methods to propose the solution for health

diseases. They are mainly focusing on the designing of tools

to detect and diagnose the diseases. This is because the

amount of information that they received using tools are

somewhat accurate but they are not perfect but contributing a

lot towards this diagnostic process. They are also trying to

design tools to take decisions about the diagnosed diseases.

These decision support systems (Balaji, 2010) will help them

a lot as the information gathered by the detection tools is so

vast and huge that human cannot understand and elaborate

easily. So in order to cater the issue, healthcare departments

are also focusing on the development of decision support

tools. Many tools have been designed and proposed to detect,

diagnose and the intervention treatment for the disease (Tan,

2016). In this research article we are going to discuss the way

researches and scientists are proposing and designing

techniques and technologies and how these developments are

contributing towards the betterment of human diseases

(Suryono, 2019).

Simulated intelligence holds huge guarantee for changing the

arrangement of medical care administrations in asset helpless

settings. A significant number of the wellbeing frameworks

obstacles in such conditions could be tended to and defeated

utilizing Artificial Intelligence upheld by other mechanical

turns of events and developing fields. The omnipresent

utilization of cell phones, joined with developing interests in

supporting innovations (e.g., m-Health, EMR and distributed

computing), give adequate occasions to utilize Artificial

Intelligence applications to improve general wellbeing results

in low-pay nation settings. While we have given a few

instances of how Artificial Intelligence is now applicable to

the intend for the betterment of wellbeing results in less-pay

nations, many positively numerous Artificial Intelligence

applications previously being executed and unquestionably

there will be more in coming years.

Many vague solutions have been proposed from the past

years. Artificial intelligence (Elkin, 2018) is helping mainly

in this regard. Multiple solutions have been proposed for the

detection of diseases and the ways and methods used for their

detection using artificial intelligence techniques. It has been

already discussed in detail that how artificial intelligence is

playing its role in healthcare. Now we are going to discuss

specifically some diseases in which it is contributing.

Researchers are working on different diseases like diabetes

(Jacobs, 2020), Alzheimer (K.S.Bijua, 2017), heart, cancer,

skin diseases etc. We are going to discuss some of them i.e.

Diabetes, Alzheimer and tumor detection.

A. Use of Artificial Intelligence in Diabetes Detection

Many decision support systems are being proposed in

healthcare departments for different diseases. A vague theory

have been proposed in for the detection of Diabetes using

artificial intelligence techniques. It is broadly pointed that the

traditional ontologies can't adequately deal with uncertain

9

what's more, unclear information for some true applications,

however vague cosmology can successfully resolve data and

information issues with vulnerability. The article (Chang-

Shing Lee, 2011) presents a novel vague master framework

for diabetes related procurement system. An architecture,

consisting fiver layers, vague metaphysics, counting a vague

information layer, vague gathering connection layer, vague

gathering space layers, vague individual connection layer, and

vague individual space layer, is created in the vague master

framework to depict information with vulnerability. The five

layer approach is well demonstrated by the below figure.

Fig. 1. Architecture of unique vague layer five system

By applying the novel vague philosophy to the diabetes space,

the semantic decision support agent is characterized to

demonstrate the diabetes information. The overall

architecture of vague diabetes philosophy is also designed in

the form of a diagram that demonstrates the concept.

Fig. 2. Architecture vague expert system for diabetes

application.

Moreover, a semantic decision support agent, counting an

information development component, vague philosophy

creating system, and semantic vague dynamic component, is

likewise evolved. The information development system

develops the vague ideas and relations dependent on the

structure of the vague philosophy creating system. The cases of

this system are created by the vague philosophy creating

component. At last, in view of above-mentioned system and the

vague philosophy, the semantic vague dynamic component

imitates the semantic illustration of clinical staff for diabetes

related application. Significantly, the suggested vague master

framework can turn out viably for diabetes choice help

application. In the end, the proposed methods is evaluated in

terms of precision, recall, accuracy, true negative rate and

specified threshold value.

10

Fig. 3. Measuring performance of proposed method

B. Use of Artificial Intelligence in Alzheimer Detection

Disease of Alzheimer is one of the neurological confusions in

which nerve cells (neurons) connecting each other injures and

dies throughout the brain. The demise of cells of brain causes

loss of memory and diminish and leads to shrinkage in the brain

tissues. It’s a type of dementia. Indications for the most part

grow continuously and weaken after some time, getting

adequately outrageous to interfere with step by step exercises

and it leads towards death. The center foundations for

Alzheimer's infections are low mind movement and blood

stream. Magnetic Resonance Imaging (Priya, 2013) is

comprehensively utilized in distinguishing Alzheimer's

infection. In their examination work they intended to build up a

computational technique to measure the cerebrum tissue

misfortune in MRI checks utilizing Contour based mind division

strategy and Threshold based Segmentation. These techniques

separate White Matter from Gray matter which helps identify

Alzheimer's disease. To evaluate and estimate the accuracy of

the method proposed in their paper, the dataset used of MRI

brain images is obtained from kaggle website. In this dataset the

skull of the brain was already removed from a total of more than

500 MRI brain images. Otherwise the skull can be removed with

Edge detection and morphological operation mentioned in

(Shanthi kj, 2013).

Disease of Alzheimer is mainly due to various kind of

dimension reduction in the cells of brain and cells normally

known as White Matter (WM) and Grey Matter (GM), therefore

segmentation of these brain tissue using MRI brain images leads

to identifying Alzheimer. Normally, partitioning the cells of

brain from Magnetic Resonance Imaging brain image is a tough

and tricky work, thus first the skull from the mind image must

be removed. Luckily the dataset they found and experimented

with was already processed and gone through this step. One of

automated skull removing methods is mentioned in the dataset

heading. In the proposed method, they used contour-based brain

segmentation method along with thresholding technique, it

plays a vital role in segmentation and pattern identification. This

is a simple Approach of dividing the MRI brain image into

different parts and regions. Some grayscales values are

automatically selected as thresholds and then used in binary

classification the image into two regions: disease of Alzheimer

and not the disease of Alzheimer. System architecture of overall

proposed method is demonstrated in the below diagram.

Fig. 4. Architecture of Proposed Alzheimer's disease

detection method

Proposed method consists of two processes. In the first process,

they did shape analysis and object detection on MRI using

Contour based brain segmentation method. Drawing a proper

boundary around the skull to quantify tissue loss later on, which

11

gave us information about brain cells excluding the black

background of MRI which later on could affect the final result

and classification of disease of Alzheimer and not the disease of

Alzheimer patients.

In the second process, Binarization Technique was applied to

get White Matter count from MRI of the brain. Extracting the

count of Grey Matter was not as simple as White Matter, to get

Grey Matter count they had to invert the background of the MRI

image first and then used Binary inversion Technique to get the

count at the end (which in fact is quantification of White Matter

& Grey Matter). The segmented Grey Matter and White Matter

was analyzed to separate MRI scans of disease of Alzheimer and

not the disease of Alzheimer patients by a condition:

If count of GreyMatter > count of WhiteMatter,

It indicates the patient has Alzheimer else the patient does not

have Alzheimer

In the proposed algorithm, we used MRI brain images obtained

from Kaggle database to evaluate the calibration of the

suggested technique. Confusion Matrix is used with total 213

MRI brain scans of Alzheimer's disease patients and 256 MRI

brain scans of normal brains with accuracy: (TP + TN)/All =

93 %. Out of 238 Medical Resonance Imaging scans of

Alzheimer's disease Patients 210 were correctly identified as

Alzheimer's disease patients and 28 were miss-identified as

non-Alzheimer's disease patients, similarly out of 231 non-

Alzheimer's disease patients only 3 were miss-classified as

Alzheimer's disease patients.

Fig. 5. Confusion Matrix for Proposed Method

The existing works in the domain of Alzheimer’s disease

detection have also experimented with segmentation techniques

mentioned in detail in the literature review of paper, most of

them using similar but not identical methods discussed in this

paper. Thus, we can't say much about which method is better

than which but can compare time complexity of algorithms

which are the basis of these methods and are used in one or

another way in every paper. Time complexity using vague-c

mean is almost O(NC), K-means (Kalavathi Palanisamy, 2018)

is O(ncdi) and time complexity of FCM (Priya, 2013) is

O(ndc2i). The discussed paper used contour-based brain

segmentation method and threshold-based segmentation for

object detection, shape analysis and quantifying Grey Matter &

White Matter (segmentation of Grey Matter & White Matter).

This method is simple with low time complexity which is almost

linear depending upon the size of data and accuracy of up to 93%

in classifying (disease of Alzheimer and not the disease of

Alzheimer patients Medical Resonance Imaging images)

Medical Resonance Imaging images to the right class.

C. Artificial Intelligence in Diagnosis of Heart Diseases

With regards to infection conclusion, experts may have various

conclusions (Saba, 2012), which lead to various choices and

activities. Then again, the measure of accessible data, even for

a situation of an average infection is tremendous to the point that

quick and accurate decision might be troublesome. For instance,

specialists may recommend a few costly tests so as to analyze a

coronary illness though a large number of those tests probably

won't be needed. Appropriately, a medical expert system

(Nazari, 2018) could be exceptionally useful there. Specifically,

such a medical expert system can be created as a specialist

framework for those persons who have a major probability of

creating heart illnesses. The investigation builds up a specialist

framework dependent on vague Systematic Ranking method and

Vague reasoning method such that as to assess the state of

persons that are getting inspected for heart illnesses. Vague

Systematic Ranking method is utilized to ascertain loads for

various standards that effect creating heart illnesses, and the

vague reasoning method is utilized for survey & assess the

probability of creating heart sicknesses in a patient. The created

framework has been executed in an emergency clinic. The

results show productivity and exactness for created approach.

The paper introduced a crossover philosophy dependent on the

Vague Systematic Ranking method and vague reasoning method

to plan a medical expert system with the point of assessing

probability of creating heart illnesses. The created medical

expert system causes professionals and masters to just

12

recommend progressed finding checks for highest probability of

creating heart illnesses has resolved. The significant advantage

of such a medical expert system is that an underlying analysis

will be led prior to endorsing any, more probable costly, clinical

tests. Consequently, it extraordinarily decreases the related

expenses and assets, while the ideal results are destined to be

delivered.

V. CONCLUSION

A quickly developing demand to guarantee is, all wellbeing

control experts have the capacities needed to explore compound

universe of Artificial Intelligence inside the medical services

environment. Such ideas should be coordinated in all training

events which are centered on establishing Artificial Intelligence

information and abilities of doctors, medical control experts,

and administrative employees that may connect with Artificial

Intelligence advances. To be applicable and successful, safe, and

merciful care, clinical administrations specialists must develop

the abilities to use these advancements. All the more

significantly, medical care experts possesses an occasion to

educate, formalize, and take into consideration by building up

the information, aptitudes and perspectives needed to advance

and empower Artificial Intelligence for better patient results.

The paper of diabetes innovation has introduced a unique

architecture consisting of five layers vague cosmology in order

to show the discipline information with vulnerability and stretch

out the vague metaphysics to the diabetes area. At long last, the

outcomes are put away in the diabetes decision making support.

Test results demonstrate that the proposed strategy can examine

information and further exchange the gained data into the

information to imitate the considering cycle of people. Our

outcomes further show that the proposed technique works more

adequately for diabetes application than recently created ones.

The second discussed paper of Alzheimer’s disease, proposes a

simple method, to detect Alzheimer’s disease in MR brain

images. They applied a widely used Contour based brain

segmentation method for segmentation of White Matter and

Grey Matter from brain image scans to classify either as person

with Alzheimer or not Alzheimer infected MRI images. Their

method is a simple methodology with low time complexity and

accuracy of up to 93% in classifying the Magnetic Resonance

Imaging images to the right class. They tested the proposed

algorithm with the image dataset taken from Kaggle and found

better segmentation results than the existing methods. The above

procedure is capable of classification into the 3 types of

Alzheimer’s disease using segmentation. For this they need to

find an accurate threshold before Binarization. Quantifying loss

of brain tissues using this technique can be used for

classification of Alzheimer’s disease. This is the future

recommendation proposed by the evaluating the above method.

It is very important to note that while the past examinations have

constraints in the quantity of variables they considered, they

didn't research the communication between factors. In actuality,

the current investigation centers on building up an exact medical

expert system, where the constraints of the past examinations

are survived. Also, the created medical expert system evaluates

patients' conditions at a less expensive cost in light of the fact

that extra costly conclusion and clinical tests may be

recommended once the medical expert system reports a high

probability of creating heart infections. To assess the proposed

medical expert system, they examined 100 patients. While the

masters at first recommended progressed conclusion checks for

81 persons, from which just 20 experienced heart diseases, the

medical expert system inferred that lone 26 persons, as well as

the past 20 persons, have a high probability of creating heart

infections. In synopsis, the suggested medical expert system

contributed into a generous sparing in expenses, and assets, and

in the request for some amount for a solitary patient.

REFERENCES

1. Arash Shaban-Nejad, M. M. (2018). Health

intelligence: how artificial intelligence transforms

population and personalized health. Digital

Medicine.

2. Balaji, S. V. (2010). Fuzzy logic based leanness

assessment and its decision support system.

3. Behnam Malmir, M. A. (2017). A Medical Decision

Support System for Disease Diagnosis under

Uncertainity.

13

4. Brian Wahl, A. C.-G. (2018). One of the most

curious and integral parts of the human body is the

brain. The most important elements. BMJ Glob

Health.

5. Chang-Shing Lee, S. M.-H. (2011, February). A

Fuzzy Expert System for Diabetes Decision Support

Application. 41(1).

6. Chinnu, A. (2013). MRI Brain Tumor Classification

Using SVM and histogram based image

segmentation. International Journal of Computer

Science and Information Technologies, 6(2), 1505-

1508.

7. David Wiljer, M. P. (2019). Developing an Artificial

Intelligence–Enabled Health Care Practice:

Rewiring Health Care Professions for Better Care.

Journal of Medical Imaging and Radiation Sciences,

59-68.

8. Duan, Y. J. (2019). rtificial intelligence for decision

making in the era of Big Data–evolution, challenges

and research agenda.

9. Elkin, P. L. (2018). Artificial intelligence: bayesian

versus heuristic method for diagnostic decision

support.

10. Giovanna Sannino, N. B. (2019). Artificial

Intelligence for Mobile Health Data Analysis.

Mobile Information Systems.

11. Giuseppe Aceto, V. P. (2020). Industry 4.0 and

Health: Internet of Things, Big Data, and Cloud

Computing for Healthcare 4.0. Journal of Industrial

Information Integration.

12. Hanane Allioui, M. S. (2019). Deep MRI

Segmentation: A Convolutional Method Applied to

Alzheimer Disease Detection.

13. Jaafar, H. F.-N. (2011). Jaafar, H. F., Nandi, A. K.,

& Al-Nuaimy, W. (2011, August). Detection of

exudates from digital fundus images using a region-

based segmentation technique.

14. Jacobs, N. S. (2020). Artificial Intelligence in

Decision Support Systems for Type 1 Diabetes.

15. Jarrahi, M. H. (2018). Artificial intelligence and the

future of work: Human-AI symbiosis in

organizational decision making.

16. Jyothi, S. (2015). Detection of Brain Abnormality

for Alzheimer's Disease Using Image Processing

Techniques.

17. K.S.Biju, S. K. (2017). Alzheimer’s Detection Based

on Segmentation of MRI Image. Procedia Computer

Science, 474-481.

18. K.S.Bijua, S. K. (2017). Alzheimer’s Detection

Based on Segmentation of MRI Image.

19. Kalavathi Palanisamy, A. A. (2017). Detection of

Alzheimer Disease in MR Brain Images using FFCM

Method. National Conference on Computational

Methods, Communication Techniques and

InformaticsVolume: Computational Methods,

Communication Techniques and Informatics.

20. Kalavathi Palanisamy, N. S. (2018). White Matter

and Gray Matter Segmentation in Brain MRI Images

Using PSO Based Clustering.

21. Kalavathi, P. C. (2017). Detection of Alzheimer

disease in MR brain images using FFCM method.

22. Karthik Seetharam MD, S. S. (2019). Artificial

Intelligence in Cardiovascular Medicine. Current

Treatment Options in Cardiovascular Medicine

volume.

23. Kazuhito Sato, M. F. (2011). Brain tissues

segmentation for diagnosis of Alzheimer-type

Dementia.

24. Mercedes Rigla, G. G.-S. (2017). Artificial

Intelligence Methodologies and Their Application to

Diabetes. Journal of Diabetes Science and

Technology.

25. Mohamed Elhoseny, A. A. (2018). A hybrid model

of Internet of Things and cloud computing to manage

big data in health services applications. Future

Generation Computer Systems, 1383-1394.

26. Nazari, S. M. (2018). A Fuzzy Inference- Fuzzy

Analytic Hierarchy Process-Based Clinical Decision

Support System for Diagnosis of Heart Diseases.

14

27. Nesreen Samer El_Jerjawi, a. S.-N. (2018). Diabetes

Prediction Using Artificial Neural Network.

International Journal of Advanced Science and

Technology, 55-64.

28. P.Kalavathi, T. (2015). MRI Brain Tissue

Segmentation using AKM and FFCM Clustering

Techniques.

29. Palanisamy, K. (2010). A hybrid method for

automatic skull stripping of magnetic resonance

images (MRI) of human head.

30. Patil, J. &. (2012). Development of digital image

processing using Fuzzy Gaussian filter tool for

diagnosis of eye infection.

31. Priya, P. K. (2013). Segmentation of Brain Tissue in

MR Brain Image using Wavelet Based Image Fusion

with Clustering.

32. Priya, P. K. (2013). Segmentation of Brain Tissue in

MR Brain Image using Wavelet Based Image Fusion

with Clustering.

33. Priya, P. K. (2013). Segmentation of Brain Tissue in

MR Brain Image using Wavelet Based Image Fusion

with Clustering.

34. R. Anitha, P. S. (2016). A segmentation technique to

detect Alzheimer's disease using image processing.

35. Saba, T. S.-Z. (2012). Expert System for Offline

Clinical Guidelines and Treatment.

36. Shanthi kj, R. D. (2013, December). Image

segmentation an early detection to Alzheimer's

disease.

37. Shikhar Kr. Sarma, K. R. (2010). An Expert System

for diagnosis of diseases in Rice Plant.

38. Shizhen Wang, Y. Y. (2019). The Current Research

Landscape of the Application of Artificial

Intelligence in Managing Cerebrovascular and Heart

Diseases: A Bibliometric and Content Analysis.

International Journal of Environmental Research

and Public Health.

39. Suryono, S. B. (2019). Real-time decision support

system for carbon monoxide threat warning using

online expert system.

40. Tan, C. F. (2016). THE APPLICATION OF

EXPERT SYSTEM: A REVIEW OF RESEARCH

AND APPLICATIONS.

41. Toro, C. A.-M.-P. (2018). Supervoxels-Based

Histon as a New Alzheimer’s Disease Imaging

Biomarker.

42. Trishan Panch, P. S. (2018). Artificial intelligence,

machine learning and health systems. Journal of

Global Health.

43. Wahl, B. A.-G. (2018). Artificial intelligence (AI)

and global health: how can AI contribute to health in

resource-poor settings?

44. Wiljer, D. a. (2019). Developing an Artificial

Intelligence–Enabled Health Care Practice:

Rewiring Health Care Professions for Better Care.

45. Y-D. Zhang, S. W. (2018). Classification of

Alzheimer Disease Based on Structural Magnetic

Resonance Imaging by Kernel Support Vector

Machine Decision Tree.

46. Yu-Dong Zhang, S. W. (2014). Classification of

Alzheimer Disease Based on Structural Magnetic

Resonance Imaging by Kernel. Progress In

Electromagnetics Research, 185-191.