Abstract: the developing new tests and for new

Abstract:

 Cancer formation is different for each type of
cancer, it has been determined by studies and research that stress also
triggers cancer types. Early precaution is important for people who have not
suffering yet and who are in beginning stage with disease like cancer which has
high chances of death. With this study Author expound that possibility of
developing such disease may be decreased and people could take measures to it.
For some of the cancer types fuzzy logic model is introduced, risk rate of
causing these cancer types and preliminary diagnosis for the person who has
chance of causing cancer will remove these risk. Using fuzzy analysis risk
outcomes are calculated, effect of stress on cancer is discussed and determined
using soft computing techniques. A fuzzy logic approach that can be adopted to
other industries as well used in health care and in effective software for
application development. Fuzzy rules can be extracted from the trained networks
and which gives information about the classification process in easily
accessible form while fuzzy logic rules point to genes that are associated with
specific type of cancer and also used for the developing new tests and for new
treatment methods.   

 

           

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1.     
Introduction

Bioinformatics
is an interdisciplinary field of science, it combines Computer Science, Biology,
Mathematics, and Engineering to analyse and interpret biological data. It
develops methods and software tools to understand the biological data and deals
with soft computing, databases, algorithms, software engineering and image
processing. The algorithms in turn depend on theoretical foundations such
as discrete mathematics, control theory, system theory, information theory, and statistics. The identification
of molecular markers and profiles is being used in cancer classification and diagnosis as well as in clinical
outcomes. The aim is to highlight some of the main molecular or bioinformatics methodologies in cancer research and their
application. In systems Biology gene regulatory networks have an
important role in advance prediction of cancer. By modelling understanding and
analysis of these gene regulatory networks dynamics. Prediction of behaviour of
the gene regulatory network will speed up in developing medicines.
Bioinformatics use the soft computing techniques for better prediction and
understanding in cancer research which is the field of bio-informatics.

Usage of numerical systems in medical field may
reduce the loss of patients. Mathematical models may be used almost everywhere
that a decision-making problem exists. Fuzzy logic plays an important role in
the field of medicine and some of the application of fuzzy logic techniques in
medicine are diagnosis of breast cancer, lung cancer or prostate cancer. If we
can use soft computing technique such as fuzzy logic, artificial intelligence,
many diseases like cancer may prevented by early diagnosis. Thus, expensive
treatments or surgeries may not even be required. In most of the cases people
concert hospital in the advanced stages of the disease so they are diagnosed
late. As a result, treatments are useless most of the time and patient dies in
short time. Future-oriented diagnosis of cancer in healthy people is the most
important issue that should be focused. 

           

            

       

 

 

 

2.     
Related
work

 Many application are available in the field of
bio-informatics which deals with classification of cancer types, prediction and
diagnosis. Many analysis and methodologies have come up which analyse the gene
expression data using data mining techniques for feature selection,
classification, clustering etc. also Soft computing methods are used for more
accuracy.

A
work titled “Evolving connectionist systems for knowledge discovery from gene expression
data of cancer tissue” explains about Microarray technique in which it is
possible to observe the expression of thousands of genes simultaneously. Author
used knowledge-based neuro computing (KBN), in particular fuzzy neural networks
are applied to classify cancer tissue, which is illustrated on the case studies
of leukaemia and colon cancer. Fuzzy logic rules are used determine type of
cancer 2. Another proposes methods to determine the risk of developing types
of cancers in the future for healthy people and preliminary diagnosis based on
the studies conducted by using artificial intelligence and fuzzy logic
technique in medicine, determining the cancer type is studied and factors
effected on types of cancer will be investigated. Here breast cancer, lung
cancer, and colon cancer are selected, reason for selecting these types is the
frequency of patient numbers and appropriateness of this study for indicated
cancer types 3. Data mining and clustering techniques are used for predicting
cancer by comparing gene expression samples that are taken from patients with
documental data. For the collected samples gene expression patterns are
designed and compared with the sample pattern to find out the affected gene
patterns. Clustering technique is used to form clusters of related gene
patterns, after analysing final prediction of cancer is done. Algorithm used in
the proposed system are Supervised Multi attribute Clustering Algorithm and Ant
Colony Optimization Algorithm 4. Neuro-fuzzy logic used develop a model to
determine the risk factors for lung cancer by analysing details of  smoke, genetic status, age, living environment
and some other factors. As
a result of this, pre-diagnosis or opportunities to remove cancer risk will be
provided to a patient by examining risk analysis for lung cancer 5.  An agent based system for risk analysis of
breast cancer has designed based on fuzzy multi agent which includes benefits form
fuzzy set theory and its linguistic information. MAS(multi agent system) decides
the

Cancer
risk rate whether it is high or low. Here actual data are gathered from various
  hospitals and analysed by comparing the
test reports 6. Adaptive fuzzy Petri net(AFPN) reasoning algorithm used to
develop a system which predict outcomes of esophageal cancer, this system performs
fuzzy reasoning to evaluate risk degree value for cancer 7.

 

3.     
 Current practice using Fuzzy Values

Fuzzy logic
technique is used in most of the medical concepts and it is convenient for many
medical application due to the relationship between the concepts and uncertain
nature of medical concepts. Using fuzzy logic can find solution for even
uncertain problems. Since the traditional analysis approaches are not
appropriate and complexity of concepts in medicine is comparatively more.    

a.     
 Fuzzy Interface System

“Fuzzy Interface System
is a process that uses fuzzy logic to map inputs and outputs”. Some of the
important concepts of fuzzy logic includes fuzzification of variables, usage of
fuzzy operators such as intersection (AND), Union (OR), additive complement.
Fuzzification converts the inputs and outputs to linguistic variables or natural
language.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

           Fig.1 Input and output map for the
indoor health risk level in fuzzy inference system.

 

 

 

 

 

 

Risk level of
specified matters classified according to the Table (1), which include the
criterion levels as very low, low, moderate, high, very high.

 

Criterion
levels

Fraction of threshold limit

DA (very low)

X<0.1 DB (low) 0.12

 

 

 

 

 

 

 

 

 

C(health risk levels) = (CPM2.5 × WPM2.5) + (CPM10 × WPM10) + (CTSP
× WTSP)

 

Determining criteria
of the risk level of specified matter (PM25, PM10 and TSP) are classified according to
the effects on health and each of the factors determine based on risk level
according to equation.

 

 

 

 

3.2. Fuzzy rules for Lung Cancer Risk
Analysis

 

               Sample Values:

fi = ai
* age + bi* cigarette + ci
* genetic + di*environment + ei*skin
color + hi
 

Age = young
& Cigarette = low & Genetic = low
& Environment = well & Skin color = white
skinned ->
f1 = a1*age
+ b1*Cigarette + c1*Genetic
+ d1*Environment + e1*Skin
Color + h1
 

f1 = a1*age
+ b1*Cigarette + c1*Genetic
+ d1*Environment + e1*Skin
Color + h1
Age
= young & Cigarette = low
& Genetic = low & Environment = worse
& Skin Color = black skinned->

 

Based
on details filled in the form Risk rate will be calculated, entries in the form
details such as age, if cancer caught before, intestine status genetic status,
Consumption of alcohol and smoking etc. This calculation of degrees enables the
rules. Output of active rule become output calculation of the whole system.

 

 

 

 

 

 

 

 

 

 

 

 

 

Conclusion

 

Aim of this
study is determining the risk rate and factors influencing on it for the occurrence
of the cancer. Usage of fuzzy logic in medicine to find the solutions for many
uncertain problems, Fuzzy rules play an important role while decision making
and analysing the factors effecting on the risk level of cancer. There are many
methods proposed for predicting the rate of causing cancer and pre-diagnosis
using soft computing approaches. Reason for selecting fuzzy technique is it
gives effective results by analysis the verbal inputs so the chances of mistakes
happing by the medical equipment is comparatively low. Probabilistic results
obtained from the fuzzy logic are compared with various test results hence
author’s proposed method can be used for private health system also as public
health system for expertise and for researchers. Many researches are going on
in this area and system can also be improved by including preventive
suggestions regarding heath habits and preventive methods.