Abstract
Breast cancer diagnosis (WBCD) is an important, real-world medical problem. There are different
artificial Intelligence techniques try to classify WBCD to help to minimize the errors that might occur when
the doctors do not have adequate experience or because of stress . In this work , fuzzy genetic tool is used
to present diagnostic system that classify WBCD cases automatically .The system provides two prime
features: first, it attain high classification performance ; second, the resulting system consists of a few
simple rules, and are therefore interpretable.
Keywords: WBCD ;Fuzzy systems; Genetic algorithms; Breast cancer diagnosis; GFRBS
الخلاصة
من المشاكل الواقعية المهمة في المجال الطبي . هناك العديد من تقنيات الذكاء (WBCD) يعد تشخيص سرطان الثدي
بهدف المساعدة في تقليص الأخطاء التي يمكن أن تقع عندما لا يمتلك الطبيب الخبرة الكافية WBCD الاصطناعي التي حاولت تصنيف
أو بسبب الإجهاد . في هذا العمل تم استخدام تقنية هجينة تجمع بين المنطق الضبابي و الخوارزميات الجينية لتقديم نظام تشخيص يصنف
أوتوماتيكياً ( بالاعتماد على الحاسوب بصورة كلية) . هذا النظام يوفر خاصيتين مميزتين : WBCD حالات
الخاصية الأولى : أن النظام يحقق أداء عالي في التصنيف .
الخاصية الثانية : يعتمد النظام على قواعد ضبابية بسيطة و قليلة و لهذا فهو يقدم تفسي ا ر لآلية صنع الق ا رر في النظام.
1. Introduction
Breast cancer is the most dangerous disease that threaten women and even men all
over the world. After lung cancer breast cancer is the second leading cause of cancer
death in women. Over the past few decades, Researchers have been tried to present
computerized diagnostic tools to help the physician in diagnosing this cancer .
A good computer-based diagnostic system should possess two important features:
1-The system should attain the highest possible performance providing a numeric value
that represents the degree to which the system is confident about its response.
2- Interpretability i.e. the system gives explanation about how the decision is made.
In this work, fuzzy logic and genetic algorithm are used to produce automatically breast
cancer diagnosis systems. Fuzzy logic makes the system interpretable while the genetic
algorithm makes production of fuzzy systems automatic. In the next two sections, a brief
overview of fuzzy systems and genetic algorithms is presented. In Section 4 describes the
WBCD problem, which is the main focus of this work. This is followed by an explanation
of GFRBS (fusion between GA and fuzzy logic). Section 6 describes in details GFRBS
that is used to solve WBCD problem in this work. In Section 7, the results obtained by
the system are displayed, followed by conclusion in Section 8.
1110
2. Fuzzy systems
This section explain Fuzzy logic concepts briefly . A more in depth explanation can
be found in [Lee, 1990] [Zadeh, 1965].
2.1. Linguistic variables
A linguistic variable [Jain and Abraham, 2004] is defined by its name and its value
which is called fuzzy values or labels ,each fuzzy label has a membership function that
assign membership degree ?Label(x) to a crisp element x that is belong to a predefined
range of discrete or continuous values , this range known as universe of discourse (UOD)
or simply universe.
In classical set theory an element must either belong or not belong to the set and there is
no possibility to partial belonging. In contrast, in fuzzy set theory, elements can belong
by a certain degree (membership degree) The value of membership degree ranges from 0
to 1 .Let x be an element belong to UOD called X , and A is a fuzzy label :