TELKOMNIKA Telecommunication Computing Electronics and Control

Vol. 21, No. 2, April 2023, pp. 324~332

ISSN: 1693-6930, DOI: 10.12928/TELKOMNIKA.v21i2.22785  324

Design AI platform using fuzzy logic technique to diagnose

kidney diseases

Haider M. Al-Mashhadi1, Abdulhusein Latef Khudhair2

1
Department of Information Systems, College of Computer Science and Information Technology, University of Basrah, Basrah, Iraq
2
Department of Computer Science, Shatt Al-Arab University College, Basra, Iraq

Article Info ABSTRACT

Article history: Artificial intelligence (AI) is an advanced scientific technology that can
provide strong ability to assist in analysis and diagnosis of almost every type
Received Jan 11, 2022 of data, therefore; AI widely used in medical fields, which is applied in the
Revised Jan 02, 2023 diagnosis and early detection of diseases. Kidney disease is one of the
Accepted Jan 24, 2023 common diseases that are diagnosed and the necessary treatments are
suggested by artificial intelligence. In this research, a logic system was used.
The fuzzy logic system (FLS) is one of the artificial intelligence systems for
Keywords: diagnosing kidney diseases, where the fuzzy logic system divided into five
variable inputs, namely urea, creatinine, glucose, bun, and uric acid, and
Artificial intelligence they represented laboratory tests of the patients, this variables and also three
Electronic-Doctor outputs were identified, which are chronic inflammation and kidney failure,
Fuzzy logic system stones and salts, acute inflammation of the kidneys and bladder, which is the
Healthcare result of the medical diagnosis of the disease. Five memberships for inputs
Kidney diseases and three memberships for outputs are used in FLS. Diseases are concluded
based on the values of the inputs, and thus the system proved its
effectiveness and accuracy in diagnosis and this system is considered an aid
to the specialized doctors in the field of kidney diseases.
This is an open access article under the CC BY-SA license.

Corresponding Author:
Haider M. Al-Mashhadi
Department of Information Systems, College of Computer Science and Information Technology
University of Basrah, Basrah, Iraq
Email: [email protected]

1. INTRODUCTION
Artificial intelligence (AI) has a multiple impact in different fields. Accordingly, there are many
projects that are being carried out to study the “future work” of the modern industrial revolutions. Therefore,
artificial intelligence technologies, big data, block-chain and the Internet of Things are all now part of daily
working life in many fields, such as smart phones, smart televisions, smart watches, or various electrical
appliances and also the use of security and medical research techniques [1], Figure 1 shows the branches of AI.
Artificial intelligence is entering the field of medical healthcare by developing clinical decision
support for patients. The large and rapid developments in improving computation in statistical techniques
have resulted in the use of artificial intelligence techniques to identify hidden patterns and interactions in the
large data set, which is usually complex and has multiple levels [2]. Kidney illness may be a major open
wellbeing problem in portion since of its common etiology caused by diabetes, hypertension, corpulence,
and maturing; the incidence of these conditions is expanding. According to the Global Burden of Diseases,
Injuries, and Risk Factors Study 2015, 750 million individuals worldwide suffered from kidney infection [3].
Electronic medical records (EMR) that contains large-field clinical data, more realistic, and this data is
the basis for the development of artificial intelligence technology in the field of multiple diseases. This big data
is very difficult for a person to analyze it directly; the reasons are the large and consuming time for analysis and

Journal homepage: http://telkomnika.uad.ac.id

TELKOMNIKA Telecommun Comput El Control  325

the accuracy required to avoid human errors, as well as the difficulty in generating ideas or information in a
deep and accurate manner [4]. In recent studies on the development of artificial intelligence in the field of
health and medical care, it has been shown that the clinical applications of artificial intelligence. Relatively
common in many fields such as ophthalmology, oncology and others [5], but we see it rare to use artificial
intelligence in kidney diseases, despite the availability of large data in this area [6] and one of the most
common kidney diseases [7]. The hemodialysis center (HD) is performed three times a week for 3−5 hours,
and thus there will be large amounts of clinical data that will be recorded in the EMR. This is the
accumulated clinical data are ideal for AI applications [2].
AI application mechanisms (AITs) were used in many fields, including medical fields such as diagnosing
diseases, treating diseases, following up on patients’ cases, predicting disease risks, and others. The use of artificial
intelligence technologies (AITs) enables the design of systems that allow building an intelligent system as a model
for predicting the patient’s diagnosis and response to specific treatment, as well as knowing the prediction of
disease severity. It is smart forms that can be used for this purpose is fuzzy logic [8]. The natural developments
that happen to various diseases, which is characterized by lack of clarity and ambiguity in that medical data may
lead to the need for a more effective framework to deal with that ambiguity through the use of variables and
multiple levels of memberships functions and approximate values Boolean. Therefore, fuzzy logic gives value
to describing medical concepts by treating them as fuzzy sets [9], [10]. Use fuzzy logic in multiple medical
systems [11]. The Figure 2 show the using fuzzy logic in healthcare.

Figure 1. AI branches Figure 2. Fuzzy logic in healthcare

After fuzzy logic was introduced by Lotfi Zadeh in 1965 [12]; however, it has recently generated
important applications in this area it was interesting [13], [14]. The diagnosis of an illness could be an issue
in pharmaceutical since a few patients may have comparable indications, but the specialist may diagnose
different maladies. So, this work will offer assistance specialist when he or she has fluffiness in that
considering prepare [15], [16]. This study presents straightforward and productive strategy to create fuzzy
expert systems for restorative determination.

2. RELATED WORKS
Ledbetter et al. [17] a small piece of kidney tissue is taken, which is called a kidney biopsy, through
which diseases that affect the kidneys are diagnosed, and there are detailed studies to monitor and observe the
pathological symptoms that result from damage to the glomeruli of the kidneys, which is called idiopathic
nephrotic syndrome in patients who are at high risk. Biopsy samples were taken from more than 80 patients
with kidney disease. The neural network inputs used were slices of kidney patients’ images to do network
learning.
Gallego et al. [18] the method of learning was used to know the diseases that affect the glomeruli in
the kidneys after taking a biopsy of the kidneys, and deep artificial neural networks were used to segment the
glomeruli and detect kidney diseases. Convolutional neural networks were used to develop two methods that
have the ability to classify patches into glomerular and non-glomerular categories. This network consists of
eight layers, five of which are convolutional and three of which are fully connected for reducing the number
of iterations.
Ayyar et al. [19] used a set of medical images of the glomeruli of the kidneys and extracted a set of
data from those images and created a glomeruli classification database (GCDB), the images were divided into
two categories, the first is normal and the other is abnormal. This extracted data is trained by the method of
deep neural network by these images. The final result is compared for the network with the categories that
are under supervision.

Design AI platform using fuzzy logic technique to diagnose kidney diseases (Haider M. Al-Mashhadi)
326  ISSN: 1693-6930

Chimwayi et al. [20] the neural-fuzzy algorithm was applied hierarchical clustering algorithm to
predict early kidney disease using specific network features. It gave an accuracy of about 70%. Then pooled
the results of that prediction, to determine the percentage of patients with a high risk of developing kidney
disease that are likely to have a high risk of diabetes.
Kunwar et al. [21] they presented research for the detection and analysis of chronic kidney disease
through the use of fuzzy C-means (FCM) algorithm in which the clustering process is effective in the process
of mining complex data that is ambiguous in nature among clinicians. This algorithm is concerned with the
verification and collection of data of patients with chronic kidney disease (CKD) and not the disease itself.
This algorithm was simulated in Matlab.
Jamal et al. [22] the z-number method was used together with the fuzzy clustering algorithm to
increase reliability and human knowledge and to develop information that is uncertain. This algorithm
collects the data of kidney patients and chooses the group to which the disease belongs (group 0, group 1,
group 2, group 3, and group 4) based on the membership functions defined in the algorithm, which is the
trapezoidal function. Using the z-number process with the fuzzy clustering algorithm gave high strength and
reliability of the results of the proposed algorithm.
Agusta et al. [23] after collecting kidney patient data from University of California Irvine (UCI)
groups, the system was developed using MySQL databases with larval framework, and the system used the
waterfall method, which analyzes the system, design, coding and test it. Fuzzification was used in multiple
stages. And configure the base rule for certain factor (CF), calculate CF expert, calculate CF values, set CF
values, and also find CF maximum. After testing the fuzzy system, an accuracy rate of 92.25% was obtained
for a data set of approximately 400, and also an accuracy rate of 97.25% for the certainty factor, and when
combining the fuzzy logic and the certainty factor, an accuracy of 99% was obtained.
Hamedan et al. [24] they used an expert system that relies on fuzzy logic in identifying and
predicting kidney disease, and then assessing the validity of the system with unclear or confusing data.
The characteristics and criteria for kidney disease were defined by consulting 18 nephrologists to form rules
for all these characteristics and cases and serve as a set of rules for fuzzy logic. An expert fuzzy logic system
was designed to infer about disease and diagnose cases by taking random samples from the records and data
of 216 patients registered in the medical records of previous auditors, some of whom had the disease and
others who were not infected, in order to test the validity of the work of the expert system. Also, some noisy
data were added to the system to compare the results with the clear and original data, and then compare the
performance of the expert system with the original results proven in the doctor’s records, and the results
indicated a significant correlation between them.
Pirmoradi et al. [25] the machine learning method was used to identify miRNAs and classify the types
of cancerous diseases that affect the kidneys automatically. This algorithm works on two basic steps, the first is
to apply the feature selection algorithm to select miRNAs for each subtype, and the inequality of arithmetic and
geometric means (AMGM) measurement feature, which is characterized by high discrimination power, was
used. The second step is implementing a self-organization deep neurofuzzy system to classify subgroups of
renal cancer disease.
Iraji [26] used fuzzy neural networks. The network was based on 24 characteristics of chronic
kidney disease and 450 samples. Proposed a network to predict and classify disease through two adaptive
neuro fuzzy and this method called (TLA-ANFIS).
Sheikhtaheri et al. [27] designed a fuzzy expert system using the Mamdani method with Matlab
language software. A group of samples were taken from the medical records of (216) patients with chronic
kidney disease as well as without disease. The expert system for those samples was evaluated and the
accuracy, quality and sensitivity of the system were very high, indicating the effectiveness of the system
compared to the final medically recorded diagnosis of the patient. Jindal et al. [28] a smart system was
designed to diagnose kidney cancer disease by using fuzzy logic and neurofuzzy. Two layers of the system
were used, the first layer concludes whether the patient has kidney cancer or not, and the second layer
determines the current stage of the patient who actually has kidney cancer. Gaussian function was used for
the inputs in a neuro fuzzy system.

3. KIDNEY DISEASE
CKD is one of the diseases that are defined as the changes that occur to the shape or structure of the
kidneys, as well as the changes that occur in its performance and function, and its impact on human
health [29], [30]. Chronic kidney disease is known by the changes that you get in the urine continuously or the
occurrence of some abnormalities in the kidneys’ structure. Most of the individuals who suffer from kidney disease
are susceptible to other diseases such as (heart and blood vessels), which causes death. In many cases, the cause of
kidney disease is high blood sugar and also high blood pressure [31]. People with chronic kidney disease have

TELKOMNIKA Telecommun Comput El Control, Vol. 21, No. 2, April 2023: 324-332
TELKOMNIKA Telecommun Comput El Control  327

an increased heart rhythm and greater atrial fibrillation compared to patients who do not have such disease,
which makes them more likely to have a stroke [32]. Figure 3 show the distinguish between normal kidney and
disease kidney.

Figure 3. Kidney disease

4. FUZZY LOGIC CONCEPT

The fuzzy system is used for data containing ambiguity and vagueness. Verbal terminology is one of
the ambiguities that fuzzy logic deals with. Fuzzy logic uses fuzzy sets that interact with each other with the
universe of discourse. These fuzzy sets are represented by membership functions whose values range are
between (0, 1) [33]. Basically, fuzzy logic is used in the development of expert systems for decision making
and classification as a simulation of the human mind. The fuzzy logic of decision-making is based on
knowledge bases stored from human experience. These rules are written programmatically as if-then [34], [35].
A fuzzy system contains four main components:
− Fuzzification: convert crisp inputs to fuzzy inputs.
− Rule base: present human knowledge or experience is stored in the form of if-then laws.
− Inference engine: stimulate rules and copy it’s from knowledge’s rules according to fuzzy input.
− Defuzzification: converting the fuzzy stimulate rules by inference engine to the crisp output values.
Figure 4 show the fuzzy logic system.

Figure 4. Fuzzy logic system

5. THE PROPOSED MECHANISM

The research mechanism uses medical tests for kidney diseases. There are many medical tests that
can be used in diagnosis; the most commonly used are urea, creatinine, glucose, uric acid and bun. These
tests are used as inputs to the fuzzy logic system (FLS) system, where the system analyzes those inputs and
deduces the percentage of disease in which the patient’s kidneys are affected. The proposed system helps
doctors specializing in kidney diseases in the speed and accuracy of diagnosis. Focus will be on three main
diseases related to the kidneys, namely:
− Acute inflammation of the kidneys and bladder (AIKB).
− The presence of chronic inflammation in the kidneys and kidney failure (PCIKF).
− The presence of stones in the kidneys, as well as the presence of salts (PSKS).
Figure 5 explain the Kidney disease diagnosis framework using fuzzy logic.

Design AI platform using fuzzy logic technique to diagnose kidney diseases (Haider M. Al-Mashhadi)
328  ISSN: 1693-6930

Figure 5. The proposed kidney disease diagnosis framework using fuzzy logic

The equation that can calculates the membership function is shown:

𝑥−𝑙𝑜𝑤
𝑐𝑒𝑛𝑡𝑒𝑟−𝑙𝑜𝑤 𝐿𝑜𝑤 ≤ 𝑥 ≤ 𝑐𝑒𝑛𝑡𝑒𝑟
𝜇𝐴 = { 𝑥−ℎ𝑖𝑔ℎ 𝐶𝑒𝑛𝑡𝑒𝑟 ≤ 0.5 ℎ𝑖𝑔ℎ (1)
𝑐𝑒𝑛𝑡𝑒𝑟−ℎ𝑖𝑔ℎ
𝑂𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
0
Figure 6 explain the proposed triangle membership function.
In the kidney disease diagnosis framework fuzzy logic system uses input values as shown in Table 1.
The fuzzy rule-base is (35) = 243. One of case if the system takes urea = (low), glucose = (low), creatinine = (low),
bun = (low) and uric acid = (low) then optimal value output is (high) that indicate there are a critical case. The rules
are running in the inference engine in parallel mode. in the last stage, the optimal crisp value represents the output
from the fuzzy space are found by defuzzification stage. This output value represents the kidney disease ratio.

Table1. Fuzzy rule of the proposed technique

L M H
L L H M
M H M L
H M M L

The procedure that shows the FLS are presented:

Procedure proposed FLS:

Begin
Number of membership function:
Input1: urea = 3
Input2: glucose = 3
Input3: creatinine = 3
Input4: bun = 3
Input5: uric acid = 3
Output: membership output = 3
Input the values of urea, glucose, creatinine, bun, and uric acid to the FLS;
Begin
Calculate the membership function for the urea, glucose, creatinine, bun, uric acid;
Put the result in 𝑌1...𝑌3;

𝑈 𝑘 = ∑𝑀=1 𝑀=2
𝐼=1 ∑𝑉=1 𝑌𝑖 ∗ 𝑌𝑣 (2)

Find the degree of all fuzzy sets by 𝑈 𝑘 the equation:

Using center of gravity method (COG) to find the result as a crisp value depending on the following equation:
𝑈 𝑖∗𝐶 𝑖
COG= ∑𝑖1 (3)
𝑈𝑖

End; end;

TELKOMNIKA Telecommun Comput El Control, Vol. 21, No. 2, April 2023: 324-332
TELKOMNIKA Telecommun Comput El Control  329

Figure 6. Representation of inputs membership function

6. EXPERIMENTAL RESULTS AND DISCUSSIONS

In this algorithm, fuzzy logic was used in diagnosing common kidney diseases through the use of
five inputs, which represent the percentage of medical tests taken from medical laboratories after conducting
a medical examination. These medical tests that were calculated are urea, creatinine, glucose, bun, and uric
acid, which was considered as crisp inputs to the system, and after converting those crisp inputs into fuzzy
inputs through the fuzzification process and calculating the membership functions, a rule-base for all the
possibilities that can be obtained based on those inputs was established. After that, the system was tested by
entering the approved data and numbers, and the required results were obtained and compared with the real
medically approved results, and it was found that they matched with the results in the system, and
accordingly, the proposed system can be adopted to help physicians specialized in kidney diseases quickly
diagnose diseased cases.
Figure 7 shows a disease case in which a group of medical tests are urea, creatinine, glucose, bun, and
uric acid. When these values are entered into the proposed fuzzy logic system, we find that the result was giving
a special percentage for each disease case, for example, chronic inflammation and kidney failure = 0.6, kidney
stones and salts = 0.8 and acute inflammation of the kidneys and bladder = 0.6. Therefore, we conclude from
this value that the patient suffers from a large percentage of kidney stones and salts, and the remaining values
were obtained as a result of the high value of kidney stones and salts. There are infections with a ratio of 0.6.
Figure 8 shows another disease case in which the medical tests of urea, creatinine, glucose, bun and
uric acid are very high than normal rates. Then we find that the results of the diagnosis of chronic
inflammation and kidney failure = 0.8 and kidney stones and salts = 0.8, acute inflammation of the kidneys
and bladder = 0.4. Therefore, we conclude from this value that the patient suffers from a very large
percentage of kidney stones and salts, and there is likely to be kidney failure.

Figure 7. Shows the results obtained using the proposed Figure 8. Another results of proposed system
system

Design AI platform using fuzzy logic technique to diagnose kidney diseases (Haider M. Al-Mashhadi)
330  ISSN: 1693-6930

Figure 9 is also another disease case in which the medical tests of urea, creatinine, glucose, bun and
uric acid showed that there were moderate increases in the percentages of the tests than normal rates. Then
we find that the results of the diagnosis of chronic inflammation and kidney failure = 0.4 and kidney stones
and salts = 0.7 and acute inflammation of the kidneys and bladder = 0.9. Therefore, we conclude from these
values that the patient suffers from the presence of few kidney stones and salts, and also there are high
infections in the bladder with a very low probability of kidney failure.
Figure 10 is also a new disease case in which the medical tests of urea, creatinine, glucose, bun and
uric acid showed that there are very high increases in the percentages of the tests above normal rates. Then
we find that the results of the diagnosis of chronic inflammation and kidney failure = 0.8 kidney stones and
salts = 0.8 and acute inflammation of the kidneys and bladder = 0.4. Therefore, we conclude from these
values that the patient suffers from the presence of large kidney stones and salts, and also there are high
infections and acute kidney failure with the possibility of few infections in the bladder.

Figure 9. Another tests of proposed system Figure 10. A new tests of proposed system

Figure 11 the medical examinations are good because most of the percentages of the tests were
within the normal limits except for a slight rise in the percentage of urea. Therefore, we find that the results
of the diagnosis of chronic inflammation and kidney failure = 0.1, kidney stones and salts = 0.3, and acute
inflammation of the kidneys and kidneys bladder = 0. Therefore we conclude from these values that the
patient does not suffer from the presence of kidney stones and perhaps there are few salts, also there are no
infections or kidney failure as well as no infections in the bladder.

Figure 11. Normal ratio tests of proposed system

7. CONCLUSION
This paper explains the process of identifying kidney disease early and helps specialized doctors in the
process of rapid diagnosis and reducing the effort in knowing the type of disease in a more accurate and reliable
way compared to the traditional method. The goal of this study is to develop a new method in the fuzzy logic of
diagnosing kidney diseases based on skill and experience specialized doctors and recorded in approved tables for
diagnosis. The results that were approved for diagnosis in kidney diseases are chronic inflammation and kidney
failure, kidney stones and salts, acute inflammation of the kidneys and bladder. In this system, various samples of
pathological symptoms were taken according to the tables established in the medical diagnosis and it was found
that they matched the results known to the doctors who specialize in the disease and their accuracy in identifying
and diagnosing disease cases. The system can be developed in the future to include all different kidney diseases
through the use of hybrid soft computing technology.

TELKOMNIKA Telecommun Comput El Control, Vol. 21, No. 2, April 2023: 324-332
TELKOMNIKA Telecommun Comput El Control  331

REFERENCES
[1] A. Kashyap, “Artificial Intelligence & Medical Diagnosis, “Scholars Journal of Applied Medical Sciences, vol. 6, no. 12, pp. 4982-4985,
2018. [Online]. Available: https://www.researchgate.net/publication/330205451_Artificial_Intelligence_Medical_Diagnosis
[2] S. Chaudhuri et al., “Artificial intelligence enabled applications in kidney disease,” Seminars in Dialysis, vol. 34, no. 1, pp. 5-16,
2021, doi: 10.1111/sdi.12915.
[3] GBD 2015 DALYs and HALE Collaborators, “Global, regional, and national disability-adjusted life-years (DALYs) for 315
diseases and injuries and healthy life expectancy (HALE), 1990-2015: a systematic analysis for the Global Burden of Disease
Study 2015,” The Lancet, vol. 388, pp. 1603-1658, 2016, doi: 10.1016/S0140-6736(16)31460-X.
[4] Amisha, P. Malik, M. Pathania, and V. K. Rathaur, “Overview of artificial intelligence in medicine,” Journal of Family Medicine
and Primary Care, vol. 8, no.7, pp. 2328-2331, 2019, doi: 10.4103/jfmpc.jfmpc_440_19.
[5] B. X. Tran et al., “Global evolution of research in artificial intelligence in health and medicine: a bibliometric study,” Clinical
Medicine, vol. 8, no. 3, 2019, doi: 10.3390/jcm8030360.
[6] J. S. -Rodriguez, M. M. Rinschen, J. Floege, and R. Kramann, “Big science and big data in nephrology,” Kidney International,
vol. 95, no. 6, pp. 1326-1337, 2019, doi: 10.1016/j.kint.2018.11.048.
[7] Y. Xie et al., “Analysis of the Global Burden of Disease study highlights the global, regional, and national trends of chronic kidney
disease epidemiology from 1990 to 2016,” Kidney International, vol. 94, no. 3, pp. 567-581, 2018, doi: 10.1016/j.kint.2018.04.011.
[8] N. Allahverdi, “Design of Fuzzy Expert Systems and Its Applications in Some Medical Areas,” International Journal of Applied
Mathematics, Electronics and computers (IJAMEC), vol. 2, no.1, pp. 1–8, 2014, [Online]. Available:
https://dergipark.org.tr/en/download/article-file/89419
[9] F. Steimann, “On the use and usefulness of fuzzy sets in medical AI,” Artificial Intelligence in Medicine, vol. 21, no. 1-3, pp. 131-137,
2001, doi: 10.1016/s0933-3657(00)00077-4.
[10] I. Gadaras and L. Mikhailov, “An interpretable fuzzy rule-based classification methodology for medical diagnosis,” Artificial
Intelligence in Medicine, vol. 47, no. 1, pp. 25- 41, 2009, doi: 10.1016/j.artmed.2009.05.003.
[11] N. H. Phuong and V. Kreinovich, “Fuzzy logic and its application in medicine,” International Journal of Medical Informatics,
vol. 62, no. 2-3, pp.165-173, 2001. [Online]. Available: https://asset-pdf.scinapse.io/prod/2003160342/2003160342.pdf
[12] L. A. Zadeh, “Fuzzy sets,” Information and Control, vol. 8, no. 3, pp. 338-353, 1965, doi: 10.1016/S0019-9958(65)90241-X.
[13] N. Belacel, M. R. Boulassel, “Multicriteria fuzzy assignment method: A useful tool to assist medical diagnosis,” Artificial
Intelligence in Medicine, vol. 21, no. 1-3, pp. 201-207, 2001, doi: 10.1016/s0933-3657(00)00086-5.
[14] S. -M. Chen, “A weighted fuzzy reasoning algorithm for medical diagnosis,” Decision Support Systems, vol. 11, no. 1, pp. 37-43,
1994, doi: 10.1016/0167-9236(94)90063-9.
[15] A. Roychowdhury, D. K. Pratihar, N. Bose, K. P. Sankaranarayanan, and N. Sudhahar, “Diagnosis of the diseases– using a GA-
fuzzy approach,” Information Sciences, vol. 162, no. 2, pp. 105-120, 2004, doi: 10.1016/j.ins.2004.03.004.
[16] I. Saritas, N. Allahverdi, I. U. Sert, “A fuzzy expert system design for diagnosis of prostate cancer,” in Proc. of the 4th International
Conference on Computer Systems and Technologies - CompSysTech, 2003, pp. 345-351, doi: 10.1145/973620.973677.
[17] D. Ledbetter, L. V. Ho, and K. V. Lemley, “Prediction of kidney function from biopsy images using convolutional neural
networks,” arXiv: Machine Learning, 2017. [Online]. Available:
https://www.semanticscholar.org/reader/ad0427befa82c7a031b05f6a6c767f6d66a7de21.
[18] J. Gallego et al., “Glomerulus classification and detection based on convolutional neural networks,” Journal of Imaging, vol. 4,
no. 1, 2018, doi: 10.3390/jimaging4010020.
[19] M. Ayyar, P. Mathur, R. R. Shah, and S. G. Sharma, “Harnessing AI for Kidney Glomeruli Classification,” in 2018 IEEE
International Symposium on Multimedia (ISM), 2018, pp. 17-20, doi: 10.1109/ISM.2018.00011.
[20] K. B. Chimwayi, N. Haris, R. D. Caytiles, and N. Ch. S. N. Iyengar, “Risk Level Prediction of Chronic Kidney Disease Using
NeuroFuzzy and Hierarchical Clustering Algorithm (s),” International Journal of Multimedia and Ubiquitous Engineering, vol. 12,
no. 8, pp. 23-36, 2017, doi: 10.14257/ijmue.2017.12.8.03.
[21] V. Kunwar, A. S. Sabitha, T. Choudhury and A. Aggarwal, “Chronic Kidney Disease Using Fuzzy C-Means Clustering
Analysis,” International Journal of Business Analytics, vol. 6, no. 3, 2019, doi: 10.4018/IJBAN.2019070104.
[22] N. J. M. Jamal, K. M. N. Ku Khalif and M. S. Mohamad, “The implementation of z-numbers in fuzzy clustering algorithm for
wellness of chronic kidney disease patients,” 2nd International Conference on Applied & Industrial Mathematics and Statistics, 2019,
vol. 1366, doi: 10.1088/1742-6596/1366/1/012058.
[23] A. Agusta, F. Y. Arini, and R. Arifudin, “Implementation of Fuzzy Logic Method and Certainty Factor for Diagnosis Expert
System of Chronic Kidney Disease,” Journal of Advances in Information Systems and Technology, vol. 2, no.1, pp. 61-68, 2020.
[Online]. Available: https://journal.unnes.ac.id/sju/index.php/jaist/article/view/44369
[24] F. Hamedan, A. Orooji, H. Sanadgol, and A. Sheikhtaheri, “Clinical Decision Support System to Predict Chronic Kidney Disease: A
Fuzzy Expert System Approach,” International Journal of Medical Informatics, vol. 138, 2020, doi: 10.1016/j.ijmedinf.2020.104134.
[25] S. Pirmoradi, M. Teshnehlab, N. Zarghami, and A. Sharifi, “A Self-Organizing Deep Neuro-Fuzzy System Approach for Classification of
Kidney Cancer Subtypes Using miRNA Genomics Data,” Computer Methods and Programs in Biomedicine, vol. 206, 2021,
doi: 10.1016/j.cmpb.2021.106132.
[26] M. S. Iraji, “Chronic Kidney Disease Prediction Using Two Layer Adaptive Neuro-Fuzzy Inference System Topology,”
International Journal of Computer Science and Information Security (IJCSIS), vol. 14, no. 8, 2016. [Online]. Available:
https://www.academia.edu/29039488/Chronic_Kidney_Disease_Prediction_Using_Two_Layer_Adaptive_Neuro_Fuzzy_Inferenc
e_ System Topology
[27] A. Sheikhtaheri, F. Hamedan, H. Sanadgol, and A. Orooji, “Development of a fuzzy Expert System to diagnose chronic kidney
disease,” Razi Journal Medical Science (RJMS), vol. 25, no. 10, pp. 46-60, 2019. [Online]. Available:
https://rjms.iums.ac.ir/browse.php?a_id=5200&sid=1&slc_lang=en&ftxt=0
[28] N. Jindal et al., “Fuzzy Logic Systems for Diagnosis of Renal Cancer,” Applied Sciences, vol. 10, no. 10, 2020,
doi: 10.3390/app10103464.
[29] G. Eknoyan et al., KDIGO 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease,
Kidney International Supplements, 2013, vol. 3, no. 1. [Online]. Available: https://kdigo.org/wp-
content/uploads/2017/02/KDIGO_2012_CKD_GL.pdf
[30] C. Zoccali et al., “The systemic nature of CKD,” Natural Reviews Nephrology, vol. 13, pp. 344-358, 2017, doi: 10.1038/nrneph.2017.52.
[31] P. Romagnani et al., “Chronic kidney disease,” Nature reviews disease primers, vol. 3, no. 17088, 2017. [Online]. Available:
https://www.nature.com/articles/nrdp201788.
[32] M. P. Turakhia et al., “Chronic kidney disease and arrhythmias: conclusions from a Kidney Disease: Improving Global Outcomes
(KDIGO) Controversies Conference,” European Heart Journal, vol.39, no. 24, pp. 2314-2325, 2018, doi: 10.1093/eurheartj/ehy060.

Design AI platform using fuzzy logic technique to diagnose kidney diseases (Haider M. Al-Mashhadi)
332  ISSN: 1693-6930

[33] T. Faisal, M. N. Taib, and F. Ibrahim, “Adaptive Neuro-Fuzzy Inference System for diagnosis risk in dengue patients,” Expert
Systems with Applications, vol. 39, no. 4, pp. 4483-4495, 2012, doi: 10.1016/j.eswa.2011.09.140.
[34] M. Marashi, A M. Torkashvand, A. Ahmadi, and M. Esfandyari, “Adaptive neuro-fuzzy inference system: Estimation of soil
aggregates stability,” Acta Ecologica Sinica, vol. 39, no.1, pp. 95-101, 2019, doi: 10.1016/j.chnaes.2018.05.002.
[35] M. K. Hussein, K. R. Hassan, and H. M. Al-Mashhadi, “The quality of image encryption techniques by reasoned logic,”
TELKOMNIKA (Telecommunication, Computing, Electronics and Control), vol. 18, no. 6, pp. 2992-2998, 2020,
doi: 10.12928/TELKOMNIKA.v18i6.14340.

BIOGRAPHIES OF AUTHORS

Haider M. Al-Mashhadi He is a professor in computer information systems

department since 2003, university of Basrah, Iraq. His research interests in the network and
information security, cybersecurity, IoT, embedded systems, cloud computing, AI and image
processing. He can be contacted at email: [email protected].

Abdulhusein Latef Khudhair He is a lecturer in computer science department

since (2006), University of Basra in Iraq. His current research interests included soft
computing and AI. His research interests are in Fuzzy Logic, Fuzzy Control, Neuro-Fuzzy and
Genetic-Fuzzy hybrid approaches. He can be contacted at email: [email protected].

TELKOMNIKA Telecommun Comput El Control, Vol. 21, No. 2, April 2023: 324-332