alexa Intelligent System for Predicting, Diagnosis and Treatment of Breast Cancer | Open Access Journals
ISSN: 2090-4924
International Journal of Biomedical Data Mining
Make the best use of Scientific Research and information from our 700+ peer reviewed, Open Access Journals that operates with the help of 50,000+ Editorial Board Members and esteemed reviewers and 1000+ Scientific associations in Medical, Clinical, Pharmaceutical, Engineering, Technology and Management Fields.
Meet Inspiring Speakers and Experts at our 3000+ Global Conferenceseries Events with over 600+ Conferences, 1200+ Symposiums and 1200+ Workshops on
Medical, Pharma, Engineering, Science, Technology and Business

Intelligent System for Predicting, Diagnosis and Treatment of Breast Cancer

Hamouda SKM1*, Abo El-Ezz HR2 and Wahed ME3

1Zagazig University, Zagazig, Egypt

2Al-Azhar University, Cairo, Egypt

3Suez Canal University, Ismailia, Egypt

*Corresponding Author:
Hamouda SKM
Computer Laboratories
Zagazig University, Zagazig, Egypt
Tel: +201124776320
E-mail: [email protected]

Received Date: March 10, 2017; Accepted Date: March 20, 2017; Published Date: March 27, 2017

Citation: Hamouda SKM, Abo El-Ezz HR, Wahed ME (2017) Intelligent System for Predicting, Diagnosis and Treatment of Breast Cancer. Int J Biomed Data Min 6: 128. doi: 10.4172/2090-4924.1000128

Copyright: © 2017 Hamouda SKM, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Visit for more related articles at International Journal of Biomedical Data Mining

Abstract

Cancer is a disease that makes the cells in the body grow out of control. When cancer starts in the breast, it is called breast cancer. Breast cancer is one of the major death causing diseases of the women in the world due to delays and inaccuracies in diagnosis of the disease. The high accuracy in cancer prediction is important to improve the treatment quality and the survivability rate of patients. In this paper, we hope to reduce the risk of the breast cancer at the earlier stage. So, we have propose a contain two parts: First: We use a Rough Set Theory (RST) as an efficient and intelligent technique, to analyze breast cancer dataset, evaluate approximate sets, and improve the accuracy of diagnosis with reduce redundancies, and evaluate the importance attributes of data. Second: We will construct MATLAB program to diagnosis and treatment of breast cancer depending on the decision rules which can generate by Rough Set Theory from dataset, and the information’s taken from records of patient’s data base. This system may help doctors to predict and diagnose breast cancer early, also helps in follow-up of the record and medical history of any patient.

Keywords

Breast cancer predictive; Diagnosis; Survivability; Rough set theory; MATLAB; Treatment

Introduction

Breast cancer [1,2] is characterized by the uncontrolled growth of abnormal cells of the breast. Most of the breast cancers start from the ducts that carry milk to the nipple called ductal cancers. Other type of cancer start in the glands that make breast milk called lobular cancers. There are also other types of breast cancer that are less common. There is no sure way to prevent from the breast cancer, but the women can reduce their breast cancer risk, breast cancer also develops in men. Earlier detection of cancer is curable and may increase the survivability rate of patients. Women must be checking breasts for cancer before she has any symptoms, there are various tests available for prediction and diagnosis of breast cancer [1-4].

Medical databases contain large amounts of data about the patients with their medical conditions. Analyzing all these databases is one of the difficult tasks in the medical environment. Thus, we need to extract the knowledge about the patients from these databases.

The rough set theory (RST) is a mathematical tool for extracting the knowledge from uncertain and incomplete database information [5,6]. Rough set methods utilize the comparison between elements, e.g., discernibility, indiscernibility, and similarity. Additionally, we construct program by MATLAB [7-10] to diagnosis and treatment of breast cancer based on the decision rules which are generated by Rough Set Theory from patient’s data base.

Breast Cancer Overview

Breast cancer

The anatomy of normal breast [1-3] is shown in Figure 1. This picture shows the lobes and ducts inside the breast, it also shows lymph nodes near the breast.

biomedical-data-mining-breast

Figure 1: Normal breast anatomy.

Breast cancer types

Ductal carcinoma: The first and most common type of breast cancer is ductal carcinoma (Figure 2). In this type, the cancer begins in cells that line a breast duct; about 7 of every 10 women with breast cancer have ductal carcinoma.

biomedical-data-mining-carcinoma

Figure 2: Ductal carcinoma.

Lobular carcinoma: Another type of breast cancer is lobular carcinoma. This cancer begins in a lobule of the breast. About 1 of every 10 women with breast cancer has lobular carcinoma.

Ductal and lobular carcinoma: Other women have a mixture of ductal and lobular type or they have a less common type of breast cancer.

Inflammatory breast cancer: The inflammatory cancer is other type of breast cancer. It occurs in about 1 of every 100 women with invasive breast cancer [1-4].

Breast cancer stages

The stage of breast cancer depends on the size of the tumour and determines whether it has spread and how far it has spread. Treated describe the stages of breast cancer: stage 0, stage I, stage II, stage III, and stage IV. Stage I is early-stage breast cancer, and Stage IV is advanced-stage breast cancer that has spread to other parts of the body. The stage of a cancer is one of the most important factors in selecting treatment options, and it uses the tumors, nodes and metastasis (TNM) system. When a patient’s TNM categories have been determined, then this information is combined in a process known as stage grouping to determine a woman’s disease stage [1,2,11,12]. In stage 1, tumor size less than 2 cm and the cancer hasn’t spread beyond the breast. Stage II: tumor size less than or equal 5 cm and cancer may have spread to the lymph nodes. Stage III: tumor size greater than 5 cm and tumors have spread to the lymph nodes, and possibly the chest wall. Stage IV: means cancer has reached other, remote parts of the body.

Effect of breast cancer stages on survival

The effect of breast cancer stages on the survival [13,14] vary depending on the stages of breast cancer.

Non-invasive and the early stages have a better chances of survival than that for the metastatic breast cancer (stage IV) which is the stage wherein the cancer has spread beyond the neighboring tissues. Table 1 shows the 5-year and 10-year survivability rate of a cancer patient.

Stage Description 5–year survival 10–year survival
Stage  0 T=0, LN=0, M=0 95% 90%
Stage I T<=2cm ,
LN=1:3, M=0
85% 70%
Stage II T>2cm&<=5cm,
LN=3:9, M=0
70% 50%
Stage III T>5cm, LN>9, M=0 55% 30%
Stage IV
Metastasis
T=any, LN=any, M=1 5% 2%

Table 1: Breast cancer survivability rate.

Breast cancer risk factors

Women who have one or more breast cancer risk factors never develop breast cancer, while many women with breast cancer have no known risk factors. Even when a woman with risk factors develops breast cancer, it’s hard to know just how much these factors might have contributed. Some risk factors can't be changed, like a person's age or race, other risk factors can change, such as smoking, drinking, and diet [1-3,15]. Some factors affect risk more than others, and the risk for breast cancer can change over time.

Risk factors you cannot change

Getting older: The risk of developing breast cancer increases with your age. Approximately, 75% of all breast cancers are diagnosed in women 50 years and over.

Genetic risk factors: About 5-10% of breast cancers are inherited, most commonly the genes BRCA1 and BRCA2.

Family history: If a woman has a personal or family history of breast cancer, she is at increased risk of developing breast cancer.

HER2/neu Status: Cancers that are HER2-positive have too many copies of the HER2/neu gene, these cancers tend to grow and spread more aggressively than other breast cancers. All newly diagnosed invasive breast cancers should be tested for HER2/neu.

Gender: Main risk factor of breast cancer is being a woman. The disease is about 100 times more common amongst women than men.

Risk factors you can change:

Lifestyle factors: For example, being overweight or obese after the menopause, physical inactivity, smoking, a high fat diet, and high alcohol consumption can play an important role in the development of breast cancer.

Hormone replacement therapy: Estrogen hormone therapy has been used to relieve symptoms of menopause and to help prevent osteoporosis but studies reveal that it also causes more risk of breast cancer.

A late first pregnancy: Women who have a late first pregnancy (after the age of 35) are more likely to develop breast cancer.

Breast cancer symptoms

Some people have no symptoms, other people may notice a change in the breast or doctor may find an unusual breast change during a physical examination [1,15]. The major symptoms of breast cancer are: New lump in the breast or underarm (armpit), thickening or swelling of part of the breast., irritation or dimpling of breast skin, redness or flaky skin in the nipple area or the breast, pulling in of the nipple or pain in the nipple area, nipple discharge other than breast milk, including blood, any change in the size or the shape of the breast, pain in the breast.

Rough Set Theory Overview

The theory of rough set theory [16] was developed by Zdzislaw Pawlak in the early 1980’s, and other researchers. The main goal of the rough set analysis is induction of (learning) approximations of concepts, it is intelligent mathematical tool for managing uncertainty that is used for the discovery of data dependencies, to evaluate the importance of attributes, to discover patterns in data, to reduce redundancies, and to recognise and classify objects.

The theory has found many interesting applications. The rough set approach seems to be of fundamental importance to AI and cognitive sciences, especially in the areas of machine learning, knowledge acquisition, decision analysis, knowledge discovery from databases, decision support, engineering, environment, expert systems, banking, medicine and others. The main advantage of rough set theory in data analysis is that it does not need any preliminary or additional information about data like probability in statistics [16,17].

Rough set basics concept

RS [16-18] starts from Information System (IS) table, the columns of which are labeled by attributes, the rows by objects of interest and entries of the table are attribute values. An Information System is a pair S=(U,A) where U is a non-empty finite set of objects called the universe and A is a non-empty finite set of attributes such that a: U→Va for every a ∈ A. The set Va is called the value set of a, elements of U are called objects.

A Decision System (DS) is a special case of information systems, S=(U, A ∪ {d}), where d (is not element of A) is the decision attribute. The elements of A are called conditional attributes or simply conditions.

Approximation sets

Let X is a subset of U, i.e., X ⊆ U.

Lower approximation: Consists of all objects which surely belong to the set.

R(X)={x ∈ U [x] R ⊆ X}.

Upper approximation: Contains all objects which possibly belong to the set.

R¯(X)={x ∈ U [x] R ∩ X≠ φ}.

Boundary region: The difference between the upper and the lower approximation constitutes the boundary region of the rough set. Boundary positive and negative regions [19] are described as below.

equations

A member of the negative region NEG (X) does not belong to X, a member of the positive region POS(X) belongs to X, and only one member of the boundary region BND(X) belongs to X, these approximation set and regions shown in Figure 3.

biomedical-data-mining-regions

Figure 3: The approximations and regions of set X.

Approximation accuracy: The accuracy of approximation (accuracy of roughness) of any subset X ⊆ U with respect to R ⊆ A , denoted αR (X) is measured by:

equations

where X denotes the cardinality of X. For empty set φ , we define α R (φ )=1.

Obviously, 0 ≤ α R(X) ≤ 1.

If αR(X )=1. Thus, the set X is crisp with respect to R, otherwise

If α R(X )<1, X is rough with respect to R.

Reducts and core

Reduct: Reducts are minimal subsets of attributes which contain a necessary portion of information from the set of all attributes [20,21]. Some attributes in decision table may be superfluous (redundant), we reduce unnecessary attributes and only keep those that are required to tell two objects that have different values on some attribute in A apart of all objects that have the same values on the attributes in B are indiscernible, or in the INDS(B)-relation to each other.

In order to express the idea of reduct, let B ⊆ A and a ∈ B in an information system I=(U, A) where U is the universe of objects, A is set of attributes, and R(B) is a binary relation: a is dispensable in B if R(B)=R(B –{a}); otherwise a is indispensable in B. Set B is independent if all its attributes are indispensable. B'⊆ B is a reduct of B if B' is independent and R(B')=R(B).

Core: The core is the set of all indispensable attributes, i.e., it is the intersection of all reducts.

Let Red (B) is the set of all reducts of B in an information system I=(U, A) where B ⊆ A then the core of B is defined as:

Core (B)=∩Red (B)

The core is included in every reduct, i.e., each element of the core belongs to some reduct. Thus, the core is the most important subset of attributes, for none of its elements can be removed without affecting the classification.

The Proposed Approach

Materials and methods

Breast cancer data are often presented as a table, columns of which are labeled by attributes, rows by objects of interest and entries of the table are attribute values, in a table containing information about patients suffering from a certain disease objects [22].

The data sets used in our experiments consists of 60 samples, each sample consists of ten of measurement features and decision attribute. All attributes have a data type value ranging from 0 to 5. Table 2 briefs the attributes of breast cancer dataset. Information system table contains dataset about 60 patients suffering from breast cancer disease taken from NCI, Egypt, (Table 3) the breast cancer dataset. All patients have been divided by experts into five classes (stages) corresponding to their health status. The stages are: stage 0, stage I, stage II, stage III, and stage IV.

Symbol Attribute Name Discrete Values
0 1 2 3 4 5
AG Age Group - Age <=35 Age >35 - - -
FH Family History No Yes - - - -
HER HER2/neu Status No Yes - - - -
LNS Lymph Node Status No Node 1< Node &<3 3< Node <9 Node >9 - -
TS Tumor Size (cm) No Tumor Tumor<2 Tumor >2&<=5 Tumor>5 Chest Wall -
HT Histological Type No Ductal Lobular Ductal-Lobular - -
HG Histological Grade No Grade I Grade II Grade III - -
ERS Estrogen Receptor Status - <20 20-49 50-100 >100 -
PRS Progesterone Receptor Status - <5 5-30 31-50 51-100 >100
M Metastasis No Yes - - - -
D Decision (Stage) Stage0 StageI StageII StageIII StageIV -

Table 2: Description of breast cancer dataset.

Objects Condition Attribution Decision (Stage)
AG FH HER LNS TS HT HG ERS PRS M
1 2 1 0 0 3 1 1 2 3 0 2
2 2 1 1 1 4 1 2 3 5 1 4
3 1 0 0 0 0 0 0 1 1 0 0
4 1 0 0 0 1 1 1 2 2 0 1
5 1 0 0 0 2 2 0 3 2 0 2
6 2 0 0 1 1 1 1 2 2 0 2
7 2 1 1 1 1 1 1 3 3 0 2
8 2 1 1 2 4 1 2 4 5 1 4
9 2 0 1 2 4 2 3 4 4 1 4
10 2 1 0 3 1 1 1 3 5 0 3
11 2 1 1 3 1 3 2 4 4 0 3
12 2 1 0 2 1 1 1 3 2 0 2
13 2 1 1 0 3 1 2 4 5 1 4
14 2 1 1 1 4 2 2 4 5 1 4
15 2 1 1 3 2 1 1 4 3 0 3
16 2 1 1 3 3 1 2 3 4 0 3
17 2 0 1 3 4 1 3 4 4 1 4
18 2 0 1 3 3 1 2 4 5 1 4
19 2 1 1 1 1 3 3 4 5 0 2
20 2 1 1 1 2 2 2 4 3 0 2
21 2 1 1 1 3 1 2 4 5 0 3
22 1 0 0 0 0 0 0 1 2 0 0
23 1 0 0 0 0 0 1 2 1 0 0
24 1 0 0 0 1 1 1 1 2 0 1
25 2 1 0 1 2 1 2 2 3 0 2
26 1 1 0 1 3 3 1 2 2 0 3
27 2 1 1 3 2 1 2 4 4 0 3
28 2 1 1 3 3 2 2 3 3 0 3
29 1 1 0 1 2 1 1 2 3 0 2
30 2 1 0 2 3 1 3 3 4 0 3
31 2 1 0 2 3 3 2 4 3 0 3
32 2 1 1 0 4 1 3 4 3 1 4
33 1 0 0 0 0 0 0 2 2 0 0
34 1 0 0 0 1 1 1 1 2 0 1
35 1 1 0 1 2 1 2 2 3 0 2
36 1 1 0 1 1 1 1 2 2 0 2
37 1 1 0 1 2 3 1 2 3 0 2
38 2 1 0 2 3 1 1 3 4 0 3
39 2 1 0 2 3 1 2 4 3 0 3
40 1 1 0 2 1 1 3 3 4 0 3
41 1 0 0 0 2 2 1 3 2 0 2
42 2 0 1 2 2 1 2 4 4 0 3
43 1 0 0 0 1 2 1 3 3 0 2
44 2 1 1 2 1 1 2 3 4 0 3
45 1 1 1 2 3 1 3 3 3 0 3
46 1 0 0 1 1 1 1 2 3 0 2
47 1 0 0 0 0 0 0 1 2 0 0
48 1 0 0 0 2 1 1 2 2 0 2
49 2 1 1 2 3 1 3 4 5 1 4
50 2 1 1 2 2 1 2 3 4 0 3
51 2 1 0 0 3 3 1 2 2 0 2
52 1 1 0 0 3 2 0 3 3 0 2
53 2 0 0 0 3 1 1 2 2 0 2
54 2 1 1 1 4 3 1 4 5 0 3
55 2 1 1 3 1 1 3 4 5 1 4
56 1 0 0 0 0 0 1 1 1 0 0
57 2 1 0 1 2 1 1 2 2 0 2
58 2 0 0 1 1 1 0 3 3 0 2
59 2 1 0 2 2 1 2 3 4 0 3
60 2 1 0 2 2 2 1 4 4 0 3
                       

Table 3: Information system for breast cancer dataset.

The problem was to find the description of each class in terms of data available for each patient of this class, check whether the set of attributes is dependent or independent, find reducts for each class, and compute the core and accuracy of descriptions. We use the MATLAB programing to compute these processes [22]. The structure of the proposed work is represented in Figure 4.

biomedical-data-mining-work

Figure 4: The structure of proposed work.

The approximation sets program

The program is written by MATLAB programing which: Computes lower, upper, and accuracy approximations.

Algorithm procedure: Input: Information system (IS) as a decision table.

T=(U,A,D,f)%where U=x1,x2 ,…,xm , A=a1,a2,…,an , D=h1,h2,…hq , f is information function

Begin,

Create matrix S ,size m x (n+1), from table T,

S={s1,s2,…,sm*(n+1)}

if any object sx=ø then //(x=1,2,…, m *(n+1))

for every object sx do replace sx by -1

if any vector X=[x1,…,xi] contain -1

then

//i=1,2,…m

delete x

end {if} ; end {for}; end {if}

for reduced table T

do compute I

// I=indiscernibility relations IND(A)

if IND(A) contain redundant values

then delete redundant values

end {for}

for T,I compute lower approximation A(X)

if xi ∈ A(X) then

create rule and insert it to Rules

end {if}

end {for}

for T,I compute upper approximation A(X)

if xi ∈ A(X) then

create rule and insert it to Rules

end {if}; end {for}

end {algorithm}

accuracy=A(X)/ A(X)

Results of approximation sets: After constructed approximation MATLAB program, we obtained the lower, upper, and accuracy attributes description of each class (stage) results. We see that stages 0, III, and IV are crisp describable in the system, and the remaining stages are roughly describable with the accuracy given in the last column. That is to say that data (symptoms) available from the patients characterize exactly classes 0, III, and IV only, and the remaining classes not are characterized exactly by this data. The accuracy of all stages (from stage 0 to stage IV) are: 100%, 60%, 92%, 100%, and 100% respectively, as shows in Table 4.

Stage
Number
Lower Approx. Upper  Approx. Accuracy
Stage  0 8 8 100%
Stage  I 3 5 60%
Stage II 23 25 92%
Stage III 22 22 100%
Stage IV 10 10 100%

Table 4: The lower, upper, and accuracy of description of each stage.

Reduction program

According to indiscernibility and reduction condition [21-23] in order to check whether the set of attributes is dependent or not to remove repeated condition attributes in the decision table based on indiscernible and reduction of rough set theory and method.

In this section we will show reducts of attributes, i.e., minimal sets of attributes necessary for the description of these classes (stages). I used a program written in MATLAB to achieve reduction of rough functions which computes the reduct and core sets of attributes.

Reduction algorithm procedure

REDUCT (C,D)

C→the set of all conditional attributes;

D→the set of decision attributes.

R ←{}

do

T←R

?χ ? (C–R)

if γ R?{x} (D)>γT(D)

T←R?{x}

R←T

Until γ R (D)=γ C(D)

Return R

Reduction results: After constructed reduction MATLAB program, we obtained that attributes 1, 2, 6, and 7 are superfluous (redundants), attributes 3, 4, 5, 8, 9, and 10. (i.e., HER2/neu status, Lymph node status, Tumor size, Estrogen Receptor status, Progesterone receptor status, and Metastasis) are the most effective symptoms in predict and accurate diagnosis of breast cancer stages, and the attributes 4, 5, and 10 (i.e., Lymph node status, Tumor size, and Metastasis) are the core of these symptoms. The summary of these results are shown in Table 5.

Description Attrib. No. Attributes (Symptoms)  Name
The necessary attributes(Most effective symptoms)
in predict & accurate diagnosis of breast cancer
3,4, 5, 8,9,10 HER2/neu Status
LymphNode Status,
Tumor Size,
EstrogenReceptorStatus,
Progesterone ReceptorStatus , and Metastasis.
The core of attributes (Symptoms) 4, 5,10 LymphNode Status,
Tumor Size,and Metastasis.
The superfluous or redundants attributes
(Unnecessary)
1, 2,
6 ,7
Age Group, Family History,
Histological Type,
Histological Grade.

Table 5: Summary of reduction operations.

The results of reduction operations after removing the redundant attributes and merge the same objects are simplified in the decision table as shown in Table 6.

Objects Condition Attribution Decision (Stage)
HER LNS TS ERS PRS M
1 0 0 3 2 3 0 2
2 1 1 4 3 5 1 4
3,56 0 0 0 1 1 0 0
 4 0 0 1 2 2 0 1
5,41 0 0 2 3 2 0 2
6,36 0 1 1 2 2 0 2
7 1 1 1 3 3 0 2
8 1 2 4 4 5 1 4
9 1 2 4 4 4 1 4
10 0 3 1 3 5 0 3
11 1 3 1 4 4 0 3
12 0 2 1 3 2 0 2
13 1 0 3 4 5 1 4
14 1 1 4 4 5 1 4
15 1 3 2 4 3 0 3
16 1 3 3 3 4 0 3
17 1 3 4 4 4 1 4
18 1 3 3 4 5 1 4
19 1 1 1 4 5 0 2
20 1 1 2 4 3 0 2
21 1 1 3 4 5 0 3
22,47 0 0 0 1 2 0 0
23 0 0 0 2 1 0 0
24,34 0 0 1 1 2 0 1
25,29,35,37 0 1 2 2 3 0 2
26 0 1 3 2 2 0 3
27 1 3 2 4 4 0 3
28 1 3 3 3 3 0 3
30,38 0 2 3 3 4 0 3
31,39 0 2 3 4 3 0 3
32 1 0 4 4 3 1 4
33 0 0 0 2 2 0 0
40 0 2 1 3 4 0 3
42 1 2 2 4 4 0 3
43 0 0 1 3 3 0 2
44 1 2 1 3 4 0 3
45 1 2 3 3 3 0 3
46 0 1 1 2 3 0 2
48 0 0 2 2 2 0 2
49 1 2 3 4 5 1 4
50 1 2 2 3 4 0 3
51,53 0 0 3 2 2 0 2
52 0 0 3 3 3 0 2
54 1 1 4 4 5 0 3
55 1 3 1 4 5 1 4
57 0 1 2 2 2 0 2
58 0 1 1 3 3 0 2
59 0 2 2 3 4 0 3
60 0 2 2 4 4 0 3

Table 6: The decision table after reduction algorithm.

The Decision Rules

All decision rules which can be generated from objects represented in decision table are listed below:

Rule 1: if (TS=0) ? (NS=0) ? (M=0) ? (HER=0) ? (ERS=1 ? ERS=2) ? (PRS=1 ? PRS=2) then (Decision=Stage 0)→objects (patients) {3,22,23,33,47,56}

Rule 2: if (TS=1) ? (NS=0) ? (M=0) ? (HER=0) ? (ERS=1 ? ERS=2) ? (PRS=2) then (Decision=Stage 1) → objects (patients) {4,24,34,48}

Rule 3: if (TS=0 ? TS=1) ? (NS=1) ? (M=0) ? (HER=0 ? HER=1) ? (ERS=2 ? ERS=3) ? (PRS=2 ? PRS=3) then (Decision=Stage 2)→ objects (patients) {6,7,19,36,46,58}

Rule 4: if (TS=2) ? (NS=0 ? NS=1) ? (M=0) ? (HER=0 ? HER=1) ? (ERS=2 ? ERS=3) ? (PRS=2 ? PRS=3) then (Decision=Stage2)→objects (patients) {5,20,25,29,35,37,41,48,57}

Rule 5: if (TS=3) ? (NS=0) ? (M=0) ? (HER=0 ? HER=1) ? (ERS=2 ? ERS=3) ? (PRS=2 ? PRS=3)) then (Decision=Stage 2)→objects (patients) {1,51,52,53}

Rule 6: if (TS=0 ? TS=1 ? TS=2) ? (NS=2) ? (M=0) ? (HER=0 ? HER=1) ? (ERS=3 ? ERS=4) ? (PRS=3 ? PRS=4 ? PRS=5) then (Decision=Stage 3)? objects (patients) {12,40,42,44,50,59,60}

Rule 7: if (TS=3) ? (NS=1 ? NS=2) ? (M=0) ? (HER=0 ? HER=1) ? (ERS=3 ? ERS=4) ? (PRS=3 ? PRS=4 ? PRS=5) then (Decision=Stage 3)→objects (patients) {21,26,30,31,38,39,45}

Rule 8: if (TS=4) ? (NS=0 ? NS=1 ? NS=2 ? NS=3) ? (M=0) ? (HER=0 ? HER=1) ? (ERS=3 ? ERS=4) ? (PRS=3 ? PRS=4 ? PRS=5) then (Decision=Stage 3)→objects (patients) {54}

Rule 9: if (TS=0 ? TS=1 ? TS=2) ? (NS=3) ? (M=0) ? (HER=0 ? HER=1) ? (ERS=3 ? ERS=4) ? (PRS=3 ? PRS=4 ? PRS=5) then (Decision=Stage 3)→objects (patients) {10,11,15,16,27,28}

Rule 10: if (TS=0 ? TS=1 ? TS=2 ? TS=3 ? TS=4) ? (NS=0 ?NS=1 ? NS=2 ? NS=3) ? (M=1) ? (HER=1) ? (ERS=3 ? ERS=4) ? (PRS=3 ? PRS=4 ? PRS=5) then (Decision=Stage4) →objects (patients) {2,8,9,13,14,17,18,32,49,55}

Breast Cancer Treatment

Breast cancer treatment [1,24-27] involves some combination of surgery, radiation, chemotherapy, hormone therapy, and targeted therapy. The treatment that’s best for one woman may not be best for another.

Sequence of breast cancer treatment

The common treatment sequence (pathway):

First: Surgery to remove the breast cancer and to reconstruct the breast happen during the same operation, and Axillary lymph node removal (if there).

Second: Radiation therapy usually follows surgery.

Third: Chemotherapy almost always recommended.

Fourth: Hormonal therapy if hormone receptor-positive (ERpositive or PR-positive).

Fives: Targeted therapy if HER2-positive.

There are many exceptions to this sequence, however, other treatments given. There are also many other variations in sequence and timing [24-27].

Treatment of breast cancer by stage

Breast cancer treatments [24-27] for all stages are summarized in Table 7.

TREATMENT OPTIONS STAGE 0 STAGE I STAGE II STAGE III STAGE IV
The Breast None MastectomyOR Lumpectomy 
plus Radiation
MastectomyOR Lumpectomy 
plus Radiation
MastectomyOR Lumpectomy 
plus Radiation
SurgeryRadiation,
or bothmay be used, depending on many individual factors
The Lymph Nodes None Axillary lymph
node biopsy
Axillary lymph node biopsy Axillary lymph node removal AND radiation Enlarged lymph nodes may be treatedif they are producing  uncomfortable symptoms
Chemotherapy None None Commonly recommended Almost always recommended Almost always recommended
Hormonal Therapy None If hormone-receptor-positive If hormone-receptor-positive If hormone-receptor-positive If hormone-receptor-positive
Targeted Therapy None If HER2-positive. If HER2-positive. If HER2-positive. If HER2-positive.
Other Parts of the Body None None None None Radiation most commonly used
Other Check up Doctor decide Doctor decide Doctor decide Doctor decide

Table 7: Treatment of breast cancer by stage.

Treatment during pregnancy

Breast cancer during the pregnancy [25,28] can be treated by using certain chemo drugs during the second and third trimesters (the fourth to ninth months), because chemo drugs does not increase the risk of birth defects. The safety of chemo during the first trimester (the first 3 months) of pregnancy has not been studied.

Both hormone therapy and targeted therapy can affect the fetus and should not be started until after the baby is born. Treatment of pregnancy is summarized in Figure 5.

biomedical-data-mining-pregnancy

Figure 5: Treatment during pregnancy.

Using MATLAB for Diagnosis and Treatment of Breast Cancer

We suggest MATLAB [7,9] program to diagnosis and treatment of breast cancer patients. This program is building according the decision rules are generated from decision table (section 4.7), and dependent on the information’s taken from patients records too. The system is classifying the patients of breast cancer to: Stage 0, Stage 1, Stage 2, Stage 3, and Stage 4. In addition the system determines the appropriate treatment for each stage depending on the opinion of medical experts, researchers, and so taking into account the health status of each patient.

The front user interface GUI [28] is designed in a user friendly manner to help who use the system without any hassles. Figure 6 contains: Patient ID, Patient Name, HER2/neu Status, Lymph Node Status, Tumor Size, Estrogen Receptor Status, Progesterone Receptor Status, and Metastasis.

biomedical-data-mining-interface

Figure 6: User interface GUI.

In report screen the patients are diagnosed to stages such (stage 0, stage 1, stage 2, stage 3, or stage 4) by matching his data with the entire database, the information’s about the patients, cancer stage, and the suggested treatments all are saved in data base, and then display in Figure 7.

biomedical-data-mining-screen

Figure 7: Report screen and suggested treatment.

Conclusion

Predicting and diagnosis of breast cancer early is very important to improve the treatment quality and survivability rate of patients.

In this paper we presented a Rough Set Theory (RST) as an efficient technique for predicting and diagnosing breast cancer in early stages, where RST technique proved that it is more efficient in the field of computational biology because of the effective classification and high diagnostic capability.

In addition, People avoid detection of breast cancer because of the high cost of the tests, this leads to a diagnosed in late stages. Therefore, we have proposed this system to predict and diagnose breast cancer easy and effective way, and selects the appropriate treatment for each patient according to her/his stage. This system also stores patient records in the database to be used as a source in hospitals to help doctors in follow-up the medical history and treatments for any patient.

References

Select your language of interest to view the total content in your interested language
Post your comment

Share This Article

Article Usage

  • Total views: 168
  • [From(publication date):
    August-2017 - Nov 24, 2017]
  • Breakdown by view type
  • HTML page views : 129
  • PDF downloads : 39
 

Post your comment

captcha   Reload  Can't read the image? click here to refresh

Peer Reviewed Journals
 
Make the best use of Scientific Research and information from our 700 + peer reviewed, Open Access Journals
International Conferences 2017-18
 
Meet Inspiring Speakers and Experts at our 3000+ Global Annual Meetings

Contact Us

Agri & Aquaculture Journals

Dr. Krish

[email protected]

1-702-714-7001Extn: 9040

Biochemistry Journals

Datta A

[email protected]

1-702-714-7001Extn: 9037

Business & Management Journals

Ronald

[email protected]

1-702-714-7001Extn: 9042

Chemistry Journals

Gabriel Shaw

[email protected]

1-702-714-7001Extn: 9040

Clinical Journals

Datta A

[email protected]

1-702-714-7001Extn: 9037

Engineering Journals

James Franklin

[email protected]

1-702-714-7001Extn: 9042

Food & Nutrition Journals

Katie Wilson

[email protected]

1-702-714-7001Extn: 9042

General Science

Andrea Jason

[email protected]

1-702-714-7001Extn: 9043

Genetics & Molecular Biology Journals

Anna Melissa

[email protected]

1-702-714-7001Extn: 9006

Immunology & Microbiology Journals

David Gorantl

[email protected]

1-702-714-7001Extn: 9014

Materials Science Journals

Rachle Green

[email protected]

1-702-714-7001Extn: 9039

Nursing & Health Care Journals

Stephanie Skinner

[email protected]

1-702-714-7001Extn: 9039

Medical Journals

Nimmi Anna

[email protected]

1-702-714-7001Extn: 9038

Neuroscience & Psychology Journals

Nathan T

[email protected]

1-702-714-7001Extn: 9041

Pharmaceutical Sciences Journals

Ann Jose

[email protected]

1-702-714-7001Extn: 9007

Social & Political Science Journals

Steve Harry

[email protected]

1-702-714-7001Extn: 9042

 
© 2008- 2017 OMICS International - Open Access Publisher. Best viewed in Mozilla Firefox | Google Chrome | Above IE 7.0 version
adwords