Fuzzy logic framework applied to medical diagnosis

Khawla BanyDomi

Gabriel Castro

Abstract

• This paper dependent on clinical applications with Fuzzy logic being applied various philosophies to create keen Fuzzy demonstrative frameworks in the most recent decade. In this audit, practically we tend to perform Fuzzy frameworks practically. Type-2 fuzzy logic has been applied by certain author s to decrease the standards for direct just as perplexing issues. Authors of related works attempted to get the dataset from perceptions, UCI Repository and existing records from different medical clinics. There are various sorts of calculations like Genetic, Back-Propagation, Decision Tree, Self-Organizing Map, LM, Modified LM, Perceptron, and Gradient Descent to assess the recognition of different infections with fuzzy logic. Yet, Back-Propagation and Genetic Algorithm considered being best among for giving better outcomes precision in every single acknowledgment of an illness. There are different devices utilized for analysis however MATLAB instrument is considered as the best supplier for ideal outcomes when contrasted with Tangara and WEKA apparatus. The analysis would be vastly improved if Back-Propagation or Genetic Algorithm was actualized with MATLAB by utilizing in excess of 7 boundaries as a result of the most extreme number of rules for ideal exactness for illness discovery. The end depends on test ways to deal with building up these sorts of frameworks and the rising need where explicit mastery is not accessible and the client can anticipate the illness from existing software’s.

• Because of cover of indications among various illnesses it’s hard to comprehend the specific infection without getting research facility result which generally need time and cost. To center this perspective, the job of creating fuzzy clinical finding frameworks was wanted requirement to serve the general public. There are bunches of analysts working in each side of the world to manage clinical issues with fuzzy logic for the improvement of wise machines to foresee various infections. nowadays, there are different fields in medication that require frameworks for illness expectation based on indicative or asymptomatic methodology like in gluten affect ability, viral contamination, tuberculosis, cervical malignant growth, celiac infection, neurological issues, Alopecia and so on Each illness has been spoken to in plain configuration with reasonable system and their discoveries.

• Fuzzy logic has been applied in all aspects of our daily life . One of the most important field is the medical field which has found in fuzzy logic a method with which to express in a more realistic way the language full of vagueness that experts use. Within the field of fuzzy logic applied to medical diagnosis there are many more applications. This work will deal with cardiovascular diseases. Since these are the main cause of death in the world according to the latest report of the World Health Organization (WHO) on the main causes of death in 2012. And it is that 3 out of 10 deaths this year were attributed to this cause. If we also move to developed countries where other diseases that greatly affect third world countries have been eradicated, the percentage of deaths caused by heart attacks is even higher. An expert system that has been used several times based on the implementation of a system in MatLab’s * Fuzzy Logic Designer * will be described. Designed using data collected by the VA Medical Center in Long Beach and the Cleveland Clinic Foundation.

** Keywords: ** fuzzy logic, cardiovascular diseases.

Introduction

what is Fuzzy-Logic?

• Fuzzy Logic resembles the human decision-making methodology. It deals with vague and imprecise information. This is gross oversimplification of the real-world problems and based on degrees of truth rather than usual true/false or 1/0 like Boolean logic; Fuzzy logic affects the whole aspects of our daily life since it has various applications in many fields.

Fuzzy Logic applications:

1. firstly, In aerospace, fuzzy logic is used in the following areas − Altitude control of spacecraft, Satellite altitude control, Flow and mixture regulation in aircraft deicing vehicles.

2. secondly, In automotive, fuzzy logic is used in the following areas − Trainable fuzzy systems for idle speed control, Shift scheduling method for automatic transmission, Intelligent highway systems, Traffic control, Improving efficiency of automatic transmissions.

3. thirdly, In business, fuzzy logic is used in the following areas −Decision-making support systems, Personnel evaluation in a large company.

4. fourthly, In defense, fuzzy logic is used in the following areas −Underwater target recognition, Automatic target recognition of thermal infrared images, Naval decision support aids, Control of a hyper-velocity interceptor, Fuzzy set modeling of NATO decision making.

5. fifthly,In electronics, fuzzy logic is used in the following areas −Control of automatic exposure in video cameras, Humidity in a clean room, Air conditioning systems,Washing machine timing,Microwave ovens,Vacuum cleaners.

6. sixth, In the finance field, fuzzy logic is used in the following areas −Banknote transfer control,Fund management,Stock market predictions,Industrial Sector.

7. seventh, In industrial, fuzzy logic is used in following areas −Cement kiln controls heat changer control, Activated sludge waste-water treatment process control, Water purification plant control,Quantitative pattern analysis for industrial quality assurance,Control of constraint satisfaction problems in structural design,Control of water purification plants.

8. eighth, In the manufacturing industry, fuzzy logic is used in following areas −Optimization of cheese production,Optimization of milk production.

9. ninth, In the marine field, fuzzy logic is used in the following areas −Autopilot for ships,Optimal route selection,Control of autonomous underwater vehicles,Ship steering.

10. tenthly, In securities, transportation, Pattern Recognition and Classification,Psychology, fuzzy logic is used in following areas −Decision systems for securities trading,Various security appliances, In transportation, fuzzy logic is used in the following areas −Automatic underground train operation, Train schedule control, Railway acceleration,Braking and stopping,In Pattern Recognition and Classification, fuzzy logic is used in the following areas −Fuzzy logic based speech recognition,Fuzzy logic based Handwriting recognition, Fuzzy logic based facial characteristic analysis, Command analysis Fuzzy image search, In Psychology, fuzzy logic is used in following areas − Fuzzy logic based analysis of human behavior,Criminal investigation and prevention based on fuzzy logic reasoning.

• Last but not least, In the medical field, fuzzy logic is used in the following areas − Medical diagnostic support system Control of arterial pressure during anesthesia, multiple variable control of anesthesia, Modeling of neurological- pathological findings in Alzheimer’s patients, Radiology diagnoses, Fuzzy inference diagnosis of diabetes and prostate cancer.

• In medicine, specialists sometimes use precise variables such as weight (kilograms, pounds …), height (centimeters, feet …), heart rate at rest (beats per minute), but In most cases, he uses others with more vagueness, such as the degree of headache suffered by the patient (low, medium, high …), the exercise he performs daily (little, normal, a lot …) and even some of the precise variables mentioned above like heart rate or age. Because an individual with 30 years is a young person or begins to be considered older?

• In addition, a doctor bases most of his diagnoses on his experience and perception as an expert on the subject, not carrying out complicated accounts according to input data with high precision.

• This is why the application of fuzzy logic in this field has been quite important, which is reflected in the number of papers published on this per year [1]:

Year	Publications
2000	96
2001	151
2002	119
2003	141
2004	182
2005	194
2006	277
2007	253
2008	290
2009	312
2010	306

• This success is due to the fact that fuzzy logic deals with both the uncertainty and the vagueness present in any clinic when a diagnosis is being made.

Fuzzy-Logic in Medical field (related work analysis)

• Sengur (2008) [1] applied LDA and ANFIS approach using MATLAB Tool To identify heart valve disease , Anbarasi M (2010) [2] approach Genetic Algorithm using Weka Tool Better results as compared to existing software’s Soni J et al. (2011) [3] Decision Tree, Naïve Bayes and KNN Tangara Tool 89% Accurate results with decision tree algorithm on IHDPS software and 609 ms to predict with naïve bayes algorithm; Ansari A.Q and Gupta N.K (2011) [4] applied Mamdani Fuzzy Logic with 7 input parameters, Neuro Fuzzy Inference System by MATLAB Simulation To identify high/low risk of cardiac patients in his study Ephzibah (2012) [5] Neuro-Fuzzy approach using MATLAB Tool Development of Neuro-fuzzy as expert system; Olainiyi et al. (2016) [6] Back-Propagation, Feed Forward Algorithm with 13 features and 2 output variables MATLAB Tool Comparative Analysis with KNN as 45.67%, Decision Tree as 84.35%, Naïve Bayes as 82.31% and BPNN as 85% recognition rate Chauhan (2018) [7] Random-Forest Classifier and Decision Tree with 5 input parameters in MATLAB Tool 75.6% Accuracy 79.35% True Positive scenario with decision tree algorithm

• Gallardo J et al. (2008) [13] Fuzzy Logic with Image Segmentation with 4 input and 1 output along with 72 rules Fuzzy Logic Toolbox 80% accuracy with radiologist and oncologist Adeli M and Zarabadipour H (2011) [14] Genetic Algorithm, Radial Basis Function , GRNN algorithm with 19 features Fuzzy Logic Toolbox Genetic Algorithm proved best among all as compared to other algorithms as 96.77% and 97.84% classification accuracy with and without pattern recognition. Sizilio G et al. (2012) [15] PDM-FCA (Pre-Diagnosed Module and Fine Needle Aspirate) approach and IVEMI approach Fuzzy Logic Toolbox Sensitivity 98.59% Specificity 85.43% Sagir A.M (2017) [16] ANFIS Grid Partitioning, Modified LM, GD Algorithm MATLAB Toolbox 84% Accuracy

Fuzzy-Logic in Medical field (applications)

1. Fuzzy Logic with Asthma Disease Asthma is a persistent lung issue influencing lungs because of restricted aviation routes. It is a sort of hazardous sickness making breathing issue a person. In the year 2010, Zarandi M.H et al. proposed a framework to determine asthma by allotting boundaries to have fuzzy logic. The entire portrayal was acted in Iran and closed with 100% explicitness and 94% affectability [8]. In 2014, the proposition of diagnosing grown-up asthma was given by Patra S and Thakur G.S by utilizing Neuro-Fuzzy fitting instrument with SOM, LVQ and BPNN calculations. The back-engendering viewed as the best among all at age 9 by giving 535 examples [9]. Another order was finished by Badnjevic An et al. in 2015 for asthma and constant sickness with MATLAB instrument profiler with various grouping calculations as NN and LM which furnishes 99.41% effectively characterized results with explicitness 100% and affectability 99.28% [10].

2. Fuzzy Logic with Diabetes Disease It is a kind of sickness coming about high blood glucose level in body. It considered as metabolic sickness in which cells of the body to insulin stream upset causes type 1, type 2, gestational diabetes and so forth Insulin gives glucose to cells to giving energy to the body. A lot of sugar level in the body leads different issues like harming kidney and nerves. Polat K and Gunes S in 2008 focused on proposed a model to analyze diabetes sickness dependent on PCA and ANFIS strategy with 8 information highlights actualized in MATLAB. The master framework demonstrated 89.47% exact with 85% affectability, 92% particularity and 0.262 root mean square blunder [28]. Another proposition was given on diabetic neuropathy for diagnosing diabetic illness utilizing ASP programming by Katigari et al. in 2017. The outcomes were figured dependent on poll approach with 8 info and 1 yield boundary affectability 89%, explicitness 98% and framework exactness viewed as 93% [29].

3. Fuzzy Logic with Cholera Disease Cholera is a bacterial disease mostly occurred after consumption of drinking contaminated water. It is a type of disease that can lead to dehydration, diarrhea and up to death if not tackle at right time. In view of Uduak A and Mfon a proposed model on cholera was based of Mamdani fuzzy approach using 3 inputs and 1 output parameter in 2013. The representation was given with centroid method as defuzzification and proved better results with MATLAB simulation [22]. In another study Okpor M.D in 2014 classified his analysis on cholera using fuzzy classification with 5 inputs and 1 output parameter. The outcomes were satisfactory for tackling cholera as compared to existing applications [23].

4. Bosom malignancy(Breast Cancer) is a kind of sickness caused because of irregularities found in bosom that shapes the cells. This infection viewed as the second generally deadliest in ladies when contrasted with cellular breakdown in the lungs. Gallardo J et al. in 2008 interoperate on mammographic pictures to zero in on bosom malignancy utilizing fuzzy logic and viewed as early finding to handle the illness. The philosophy was utilized as picture division alongside 4 info and 1 yield boundaries conveying 72 guidelines to close a suitable choice with 80% precision [13]. Another computerized identification approach was given by Adeli M and Zarabadipour H in 2011 with design acknowledgment by utilizing Genetic, Radial premise and GRNN calculation with 19 unmistakable highlights. Hereditary calculation viewed as best with 96.77% arrangement exactness for diagnosing bosom malignant growth [14]. Sizilio G et al. in 2012 proposed a model for pre-conclusion bosom malignant growth with Finite Needle Aspirate (FCA) approach actualized in MATLAB. The exactness of the framework was estimated as Sensitivity 98.59% and Specificity as 85.43% [15]. Another illness arrangement was finished by Sagir A.M in 2017 focused on arrangement with bosom disease by actualizing ANFIS, Modified LM and Gradient Descent calculations and demonstrated 84% precise through fuzzy logic tool compartment [16].

5. Fuzzy Logic with Liver Disease It is a sort of hepatic illness that makes liver forestall working and its working. Most of variables of liver infection are because of alcoholic or hereditary nature. The most well-known kinds of liver infection are greasy liver, hepatitis B or C, cirrhosis, alcoholic hepatitis, hemochromatosis and so on To anticipate liver illness, Satarkar S.L and Ali M.S in 2013 worked on finding liver infection based on master framework teamed up with fuzzy logic. As per them, the portrayal was given with Mamdani approach by utilizing 3 sources of info and 1 yield factors to recognize the danger in the people with respect to liver illness [11].

6. Fuzzy Logic with Dental Disease It is a kind of sickness that tainting encompassing teeth as tooth rot, periodontal infection, gum disease, dental plaque and so forth Allahverdi N and Akcan T analyzed on periodontal dental issue in 2011. The goal of utilizing fuzzy logic dependent on 164 fuzzy principles as fuzzy extraction with 5 information factors to limit the time taken for distinguishing proof of dental infection [24].In the following year Parewe A.M et al. spoken to dental issue with half and half fuzzy on development worldview with 8 info boundaries. The results were 82% precise when contrasted with fuzzy with 70% having RMSE of under 1 [25]. Later on, Allahverdi in 2014 proposed a model of consolidating three distinct infections shrouded in one fuzzy master approach. He chipped away at dental, heart and paleness issue with 11 information boundaries for diagnosing various infections [26].

7. Fuzzy Logic with Heart Disease It is a kind of sickness caused because of harm or blockage of veins in heart influencing less supplements and oxygen supply to heart organ. Various kinds of heart illnesses are normal like vein issue, heart failure, cardiovascular breakdown, arrhythmia, stroke and so forth fuzzy logic is continually developing to identify heart patients all through the world with the assistance of growing new programming’s based on various boundaries. From the perspective of Sengur; recognizable proof of heart valve with similar investigation of LDA and ANFIS approach was done in 2008. The entire execution was demonstrated in MATLAB programming for recognizable proof of coronary illness [1]. Considering comparative boundaries by Anbarasi in 2010proved better outcomes when contrasted with existing programming’s for coronary illness expectation based on Genetic calculation with execution dependent on Weka apparatus [2]. In 2011, Soni J et al. created IHDPS programming which indicated 89% exactness with choice tree calculation when contrasted and Naïve Bayes and KNN. It takes 609 ms to anticipate coronary illness with Naïve Bayes calculation that was executed on Tangara device [3].

• The following can be used to show the fuzzy logic toolbox in heart disease analysis

“Fuzzy-Logic-Controller-Setup-MATLAB-Fuzzy-Logic-Toolbox”

Process

• The process of a fuzzy system has the following steps [2]

Steps

Diagnosis of risk of heart disease

• Medicine is a very extensive field and as previously mentioned, the number of publications that make use of fuzzy logic to make a medical diagnosis is enormous.

• Among all the possibilities that I have found and decided to delve into a case that is repeated in several publications [3,4,5] since, in addition to that I can try to implement it myself using the * Fuzzy Logic Designer * from MatLab, it is the leading cause of death in the world in 2012 according to WHO.

• Knowing whether or not a person is going to suffer a heart attack is not something simple since a large number of variables must be taken into account and; although some are well known such as cholesterol or irregularities in the heart pulse; 50% of people who had a heart attack for the first time had none of these symptoms before [5].

• The fuzzy expert system (* fuzzy expert system , FES) developed consists of a user interface with which the user (administrator, doctor or patient) interacts to enter symptom data that is included in a database. This database is fuzzified * to enter the fuzzy knowledge base where a result can be later consulted using the knowledge base data as input data in an inference system that is then defuzzified to show the result to the user-readable.

“FES”

A * dataset * based on data obtained at the VA Medical Center in Long Beach and the Cleveland Clinic Foundation and classified by the University of California has been used.

• The purpose of this dataset is to diagnose the presence or absence of heart disease risk that a patient may have. Originally, the * dataset * had 76 variables and 303 patients, but to develop this fuzzy system, only the six most significant ones have been used (5 input and 1 output).

• The input variables are:

Variable	Unit of measure
Blood pressure	mmHg
LDL cholesterol	mg / dl
Heart rate	beats / m
ST segment depression	mV
Age	years

• And the output indicates the percentage of risk of heart disease of the individual; the objective is to be able to diagnose a person’s risk of suffering from heart disease in the short to medium term using variables that have a lot of influence. Because it is not something that can be determined precisely by taking into account only cholesterol, blood pressure or average heart rate.

Fuzzify

• The first step of all is to define the * membership functions * of each variable to convert them to fuzzy numbers. We have to convert the obtained values ​​to objects that are in one or more sets with a degree of membership. For example:

• 170 mmHg is a medium pressure with grade 0.13 and high with grade 0.94.

• 192 mg / dl of cholesterol in the blood is a low level with 0.11 grade and a medium with 0.07 grade.

To obtain these degrees of membership, the membership functions that will be explained later will be used.

Inputs

• These are all the input parameters of a patient that are * fuzzified * as determined below:

Systolic blood pressure There are two movements when the heart beats. Systole, when it contracts, and diastole, when it dilates.

When measuring blood pressure with a blood pressure monitor, we obtain two values, commonly called high and low. The high is the one we will use and it is the maximum value of blood pressure when the heart goes through systole. It is measured in millimeters of mercury (mmHg).

We divide it into four fuzzy sets (two trapezoidal and two triangular)

“Fuzzy systolic blood pressure”

“Systolic blood pressure function”

LDL cholesterol

• One of the biggest risks of having a heart attack is the presence of a high level of LDL (* low-density lipoprotein *) cholesterol in the blood. Since it blocks the arterial walls, preventing blood circulation.

! [“Cholesterol”] (img/cholesterol.png)

• A higher amount of LDL cholesterol translates into a higher risk of having a heart attack. It is measured in milligrams per deciliter of blood (mg / dl).

• We divide it into four fuzzy values ​​(two trapezoidal and two triangular):

“Fuzzy cholesterol”

! “Cholesterol function”

Frequency heart rate Heart rate is how many times your heart beats per minute. The average at rest is between 60 and 100 beats per minute. When exercising, the heart rate increases linearly and determines the cardiorespiratory fitness of a person; We divide it into three fuzzy sets (two trapezoidal and one triangular)

:!Fuzzy heart rate"] (img/heart-rate.png)

["! [“Heart rate function”] (img/heart-rate-function.png)

ST segment depression

• The electrocardiogram (ECG) is a representation of the electrical activity of the heart. It is obtained with ultrasound using an electrocardiograph.

“ECG”

Measuring ST segment depression during exercise can help predict heart disease. The higher this, the more chances of having a heart attack. It is measured in millivolts (mV).

Three fuzzy sets will be used (two trapezoidal and one triangular)

:

["

Age

The older an individual is, the greater the likelihood that they will have a heart attack.

Four fuzzy variables will be used (two trapezoidal and two triangular)

:

“Fuzzy age”

Output

The system it will provide a single output with the percentage of attack risk.

For this, five triangular fuzzy sets will be used

:! [“Fuzzy risk”] (img/risk.png)

Inference mechanism

• The Mamdani method is used as a fuzzy inference mechanism since it is the one provided by MatLab. This method has a fairly simple structure with * min-max * operations:

• • OR *: max
• • AND *: min

• Aggregation: max

• Implication: min

! [“Application schema”] (img/system.png)

### Rules

• This mechanism is based on the definition of fuzzy relationship rules. For example: “If the blood pressure is very high the risk is very high.” A weight is applied to each of the rules that determines its degree of validity.

• The rules have been created taking into account both the results of the * dataset * and the opinions of experts, creating more than 1,000 rules.

• These rules have not been published in their entirety, so the simulation that I have personally carried out using MatLab includes a few rules that I have created

:! “Rules”

• Here you can see how the system works . First, the alpha cuts in the fuzzy set are defined using each of the membership functions and the input data. For the rules in which these generated areas have influence, it is moved to the exit area of ​​that rule taking into account its weight. Finally, an aggregation of the areas generated by all the rules is carried out to obtain the fuzzy result.

Defuzzify

To obtain a result in the form of a percentage and not in the form of a fuzzy variable, it is necessary to defuzzify, for this, the centroid method is used. It is a very simple method that calculates the center of the area obtained after applying the rules to the input data.

! “Methods to defuzzify”

Examples

• As the * dataset * with which the fuzzy system has been built has not been released, I can’t see how much it resembles reality . But if it can be compared with a mathematical model also used to calculate this percentage based on a function in which each entry is multiplied by a factor depending on its weight to obtain the result:

Age	Pressure	Cholesterol	Heart rate	ST segment depression	Risk (fuzzy)	Risk (mathematical)
71	120	265	130	0.24	40	32
49	130	188	139	2	40	28
54	135	129	126	0.1	10	8
59	140	187	152	0.1	20	17
57	128	229	150	0.14	40	24
61	122	260	170	3.6	60	53
39	165	219	150	1.2	60	48
61	145	277	186	1	60	61
56	125	249	144	1.2	40	41
45	130	164	135	0.16	20	13
56	190	288	153	4	100	121
54	160	239	146	1.8	60	57
41	120	200	130	0.1	20	16
61	124	209	163	0.1	20	18
58	120	258	137	0.14	20	28
51	122	227	124	0.1	20	21
29	130	204	202	0.1	20	20
51	140	241	186	0.1	20	29
43	122	213	165	0.12	20	20
57	167	299	164	1	80	85

Conclusions and future works

The system has been tested by medical experts and comes to simulate a doctor as it coincides 94% with the results obtained by this. The main advantage is that having the input data, it is not necessary to consult an expert and the user himself can know the risk he has of suffering a heart attack. It can also support a doctor with little experience in making their first successful diagnoses, Just as this system has been created specifically for heart problems, others could be created for any other type of disease.

And it is that the application of fuzzy logic in medicine, as already mentioned in the introduction, is enormous and could also be applied to distinguish between various diseases given symptoms.

Another future worldview is to distinguish different medications to be prescribed notwithstanding sickness identification or forecast and further infections can be analyzed in a new way. It is accepted that fuzzy logic will modify existing frameworks into half breed wise frameworks in future.

References

[1] Sengur, A. “An expert system based on linear discriminant analysis and adaptive neuro-fuzzy inference system to diagnosis heart valve diseases”, Expert Systems with Applications 2008; 35 (1–2): 214–222.

[2]. Anbarasi, M., Anupriya, E., Iyengar, N. “Enhanced Prediction of Heart Disease with feature subset selection using Genetic Algorithm”. International Journal of Engineering Science and Technology 2010; 2(10):5370- 5376.

[3] Soni, J., Ansari, U., Sharma, D., Soni, S. “Predictive data mining for medical diagnosis: An overview of heart disease prediction”. International Journal of Computer Applications 2011; 17(8):43-48.

[4]. Ansari, A.Q, Gupta, N.K. “Automated diagnosis of coronary heart disease using neuro fuzzy integrated system”. IEEE 2011. DOI:10.1109/wict.2011.6141450.

[5] Ephzibah, E.P, Sundarapandian, V. “An expert system for heart disease diagnosis using neuro-fuzzy technique”. International Journal on Soft Computing, Artificial Intelligence and Applications 2012; 1(1).

[6]. Olaniyi, E.O., Oyedotun, O.K., Helwan, A., Andan, K. “Neural Network diagnosis of heart diseases”. 2016, DOI:10.1109/ICABME.2015.7323241.

[7]. Chauhan, R., Jangade, R., Rekapally. “Classification model for prediction of heart disease”. Springer 2018; 707-714. [8]. Zarandi, M.H., Zolnoori, M., Moin, M., Heidarnejad, H. “A fuzzy rule based expert system for diagnosing asthma”. Transaction E: Industrial Engineering 2010; 17(2):129-142.

[9]. Patra, S., Thakur, G.S.M. “A proposed neuro fuzzy model for adult asthma disease diagnosis”. Computer Science and Information Technology 2014; pp: 191-205. DOI:10.5121/csit.2013.3218.

[10] Badnjevic, A., Cifrek, M., Koruga, D., Osmankovic, D. “Neuro-Fuzzy classification of asthma and chronic obstructive pulmonary disease”. BMC Medical Int. J. Advanced Networking and Applications Volume: 10 Issue: 04 Pages: 3944-3950 (2019) ISSN: 0975-0290 3950 Informatics and Decision Making 2015; 15(Suppl3): S1:1-9.

[11]. Satarkar, S.L., Ali, M.S. “Fuzzy expert system for the diagnosis of common liver disease”. International Engineering Journal for Research and Development 2013, 1(1).

[12]. Hashmi, A., Khan, M.S. “Diagnosis blood test for liver disease using fuzzy logic”. International Journal of Sciences: Basic and Applied Research 2015; 20(1):151- 183.

[13]. Gallardo, J., Hernandez-Vera, B., Lasserre, A.“Interpretation of Mammographic using Fuzzy logic for early diagnosis of breast cancer”. IEEE Computer Society 2008; 278-283. DOI:10.1109/MICAI.2008.58.

[14]. Adeli, M., Zarabadipour, H. “Automatic disease diagnosis systems using pattern recognition based genetic algorithm and neural networks”. International Journal of Physical Sciences 2011; 6(25):6076-6081.

[15]. Sizilio, G., Leite, C., Guerreiro, A., Neto, A.D.“Fuzzy method for pre-diagnosis of breast cancer from the fine needle aspirate analysis”. BioMedical Engineering Online 2012;11(83):1-21.

[16]. Sagir, A.M., Sathasivam, S. “Intelligence system based classification approach for medical disease diagnosis”. AIP Conf

[17]. Patel, A., Gupta, SK, Rehman, Q., & Verma, MK (2013). Application of Fuzzy Logic in Biomedical Informatics. * Journal of Emerging Trends in Computing and Information Sciences, 4 * (1), 57-62.

[18]. Tarhini, A., Kočiš, L ’. (1998). Fuzzy Approach in Psychiatry. * Psychiatria *, 80-86.

[19] . Sikchi, SS, Sikchi, S., & Ali, MS (2013). Generic medical fuzzy expert system for diagnosis of cardiac diseases. * International Journal of Computer Applications, 66 * (13).

[20]. Adeli, A., & Neshat, M. (2010, March). A fuzzy expert system for heart disease diagnosis. In * Proceedings of International Multi Conference of Engineers and Computer Scientists, Hong Kong * (Vol. 1).

[21]. Barman, M., Choudhuy, JP (2012). A Fuzzy Rule Base System for the Diagnosis of Heart Disease. * International Journal of Computer Applications, 57 * (7).

[22]. Uduak, A., Mfon, M. “Proposed fuzzy framework for cholera diagnosis and monitoring”. International Journal of Computer Applications 2013; 82(17):1-10.

[23] Okpor, M.D. “Using Fuzzy classifier for cholera analysis”. International Journal of Science and Research 2014; 3(8):314-317.

[24] Allahverdi, N. and Akcan, T. “A fuzzy expert system design for diagnosis of Periodontal Dental Disease” IEEE 2011.

[25]. Parewe, A.M., Mahmudy W.F., Ramdhani F., Anggodo, Y. “Dental disease detection using hybrid fuzzy logic and evolution strategies”. Journal of Telecommunications, Electronic and Computer Engineering; 10(10).

[26]. Allahverdi. “Design of Fuzzy expert system and its applications in some medical areas”. International Journal of Applied Mathematics, Electronics and Computers 2014; 2(1):1-8.

[27]. Ambara, B., Putra, D., Rusjayanthi, D. “Fuzzy expert system of oral and dental disease with certainity factor”. International Journal of Computer Science 2017, 14(3):22-30. DOI: 10.20943/01201703.2230.

[28]. Polat, K., Gunes, S. “An expert system approach based on principal component analysis and adaptive neuro fuzzy inference system to diagnosis of diabetes disease”. Elsevier, Digital Signal Processing 2008; 17:702-710.