A fuzzy inference system for assessment of the severity of the peptic ulcers

Dhinaharan Nagamalai et al. (Eds) : ACITY, WiMoN, CSIA, AIAA, DPPR, NECO, InWeS - 2014
pp. 263–271, 2014. © CS & IT-CSCP 2014 DOI : 10.5121/csit.2014.4527
A FUZZY INFERENCE SYSTEM FOR
ASSESSMENT OF THE SEVERITY OF THE
PEPTIC ULCERS
Kianaz Rezaei1
, Rahil Hosseini1,2
, and Mahdi Mazinani1,2
1
Department of Computing, Kharazmi University, Tehran, Iran
2
Faculty of Engineering, Islamic Azad University, Share-Qods Branch, Tehran,
Iran
kianaz.rezae@gmail.com, rahilhosseini@gmail.com
ABSTRACT
Peptic ulcer disease is the most common ulcer of an area of the gastrointestinal tract. The aim
of this study is to utilize soft computing techniques to manage uncertainty and imprecision in
measurements related to the size, shape of the abnormality. For this, we designed a fuzzy
inference system (FIS) which emulates the process of human experts in detection and analysis of
the peptic ulcer. The proposed approach models the vagueness and uncertainty associated to
measurements of small objects in low resolution images In this study, for the first time, we
applied soft computing technique based upon fuzzy inference system (FIS) for assessment of the
severity of the peptic ulcer. Performance results reveal the FIS with maximum accuracy of
98.1%, which reveals superiority of the approach. The intelligent FIS system can help medical
experts as a second reader for detection of the peptic ulcer in the decision making process and
consequently, improves the treatment process.
KEYWORDS
Soft computing, Fuzzy inference system (FIS), Peptic ulcer.
1. INTRODUCTION
The second common cause of death from malignant disease is gastric cancer around the world.
Detection and treatment of this painful disease has become one of the challenging medical
problems.
Nowadays gastric ulcer is one of the most important concerns involves many factors especially
widespread using of NSAIDs. Because of poorly understanding the pathophysiology of this
disease [6], studies investigating new active compounds are needed. As well, various
pharmaceutical products currently used for treatment of gastric ulcers are not completely efficient
and cause many adverse side effects.
Peptic ulcer disease encompassing gastric and duodenal ulcer is the most prevalent
gastrointestinal disorder [1]. They are caused by various factors such as drugs, stress or alcohol,
due to an imbalance between offensive acid- pepsin secretion and defensive mucosal factors like
mucin secretion and cell shedding [2]. Gastric ulcer therapy faces a major drawback due to the
unpredictable side effects of the long-term use of commercially available drugs. It is shown that
toxic oxygen radicals plays an important role in the etiopathogenesis of gastric damage

264 Computer Science & Information Technology (CS & IT)
[3].Currently, focus on plant research has increased all over the world and a large source of
evidence has been collected to show immense potential of medicinal plants used in various
traditional systems [4].
One of the main group of problems in medical science is related to diagnosing diseases based on
different tests on patients. However, the final diagnosis of an expert is associated with difficulties.
This matter led the physicians to apply computer aided detection and diagnosing tools in the
recent decades.
A prime target for such computerized tools is in the domain of cancer diagnosis. Specifically,
where breast cancer is concerned, the treating physician is interested in ascertaining whether the
patient under examination exhibits the symptoms of a benign case, or whether her case is a
malignant one [16].
The uncertainty issues in decisions making and medical diagnosis are related to incompleteness of
medical science. computer aided detection (CAD) tools are presented with the purpose of
facilitating the diagnosis of different diseases and acceleration of the treatment process [18]-[20].
One of the current applications of the CAD systems is to analysis the severity diagnosis of peptic
ulcer presented in [10].
This study is concerned with the severity diagnosis of peptic ulcer and uses fuzzy inference
systems for automatic diagnosis of disease. The required medical knowledge are aided by fuzzy
systems and achieved data are from tested stomachs. This method assorts patients according to
the length of ulcer.
The present study has been undertaken with the aim to assess the peptic ulcer severity using a
fuzzy system.
2. LITERATURE REVIEW
In recent years, a major class of problems in medical science involves the diagnosis of disease,
based upon various tests performed upon the patient. When several tests are involved, the ultimate
diagnosis may be difficult to obtain, even for a medical expert. This has given rise, over the past
few decades, to computerized diagnostic tools, intended to aid the physician in making sense out
of the welter of data [16].
Soft Computing techniques based on the concept of the fuzzy logic or artificial neural networks
for control problems has grown into a popular research area [11]-[13]. The reason is that classical
control theory usually requires a mathematical model for designing controllers. The inaccuracy of
mathematical modeling of plants usually degrades the performance of the controllers, especially
for nonlinear and complex control problems. Fuzzy logic has the ability to express the ambiguity
of human thinking and translate expert knowledge into computable numerical data.
A fuzzy system consists of a set of fuzzy IF-THEN rules that describe the input-output mapping
relationship of the networks. Obviously, it is difficult for human experts to examine all the input-
output data from a complex system to find proper rules for a fuzzy system. To cope with this
difficulty, several approaches that are used to generate the fuzzy IF-THEN rules from numerical
data have been proposed [11]-[13].
Today, medical endoscopy is a widely used procedure to inspect the inner cavities of the human
body. The advent of endoscopic imaging techniques allow the acquisition of images or videos
created the possibility for the development of the whole new branch of computer-aided decision

Computer Science & Information Technology (CS & IT) 265
support systems. This section summarizes related works specifically targeted at computer-aided
decision support in the gastrointestinal tract [10].
A symbiotic evolution-based fuzzy-neural diagnostic system for common acute abdominal pain
presents a symbiotic evolution-based fuzzy-neural diagnostic system (SE-FNAAPDS) for
diagnosis of common acute abdominal pain (AAP) without professional medical examination
[11]. The computer-assisted diagnostic system is formatted a multiple-choice symptom
questionnaire, with a prompt/help menu to assist user in obtaining accurate symptom data using
nothing more technologically sophisticated than a medical-type thermometer and stethoscope.
Compared to traditional methods, diagnostic decisions from SE-FNAAPDS shows 94%
agreement with professional human medical diagnosis and less CPU time for system construction.
The presented method is useful as a core module for more advanced computer-assisted diagnostic
systems, and for direct application in AAP diagnosis [11].
Non-ulcer dyspepsia (NUD) has been attributed to gastritis and Helicobacter infection in A
Quantitative analysis of symptoms of Non-Ulcer Dyspepsia as related to age, pathology, and
Helicobacter Infection. The Sydney classification enables dyspepsia symptoms assessed
quantitatively in relation to Helicobacter infection and topographic pathology in different gastric
compartments. The method presented in this study for 348 patients with the NUD. It studied the
unconfounded effects of age, pathology, and Helicobacter. It was concluded that age was the most
important determinant of dyspeptic symptoms, but not pathology or Helicobacter [7].
Computer-aided capsule endoscopy images evaluation based on color. Rotation and texture
features were used as an educational tool to physicians.
Wireless capsule endoscopy (WCE) is a revolutionary, patient-friendly imaging technique that
enables non-invasive visual inspection of the patient’s digestive tract and, especially, small
intestine. Experimental results demonstrated promising classification accuracy (91.1%) exhibiting
high potential towards a complete computer-aided diagnosis system that will not only reduce the
Wireless capsule endoscopy (WCE) data reviewing time, but also serve as an assisting tool for the
training of inexperienced physicians [9].
3. MATERIALS AND METHODS
Fuzzy set A in universe of discourse X can be defined as a set of ordered pairs of element x in X
and the grade of membership of x, µA (x), to fuzzy set A [15] as follows:
where the two dimensional membership function µA (x) is a crisp value between 0 and 1 for all
x∈X. Linguistic terms are modelled using fuzzy sets. One of the parameters in the design of a
fuzzy logic is the number of fuzzy sets associated to a linguistic term. Fuzzy inference system as a
soft computing method mimics cognitive reasoning of the human mind based on linguistic terms
for performing tasks in a natural environments.
Figure 1. Architecture of a Fuzzy Inference System

The fuzzy inference system is a rule-based system that uses fuzzy logic, rather than Boolean
logic, to reason about data. Its basic structure includes four main components, as depicted in
Figure 1: (1) a fuzzifier, which translates crisp (real-valued) inputs into fuzzy values; (2) an
inference engine that applies a fuzzy reasoning mechanism to obtain a fuzzy output; (3) a
defuzzifier, which translates this latter output into a crisp value; and (4) a knowledge base, which
contains both an ensemble of fuzzy rules, known as the rule base, and an ensemble of
membership functions, known as the database[16].
The fuzzy inference system is a popular computing framework based on the concepts of fuzzy set
theory, fuzzy if-then rules, and fuzzy reasoning. It has found successful applications in a wide
variety of field, such as automatic control, data classification, decision analysis, expert systems,
and pattern recognition.
This Mapping is accomplished by a number of fuzzy if-then rules, each of which describes the
local behavior of the mapping. In particular, the antecedent of a rule defines a fuzzy region in the
input space, while the consequent specifies the output in the fuzzy region.
Fuzzy logic models can be developed from expert knowledge or from process (patient) input-
output data. In the first case, fuzzy models can be extracted from the expert knowledge of the
process. The expert knowledge can be expressed in terms of linguistics, which is sometimes
faulty and requires the model to be tuned. This process requires defining the model input
variables and the determination of the fuzzy model system parameters.
Sugeno, or Takagi-Sugeno-Kang, method of fuzzy inference. Introduced in 1985, it is similar to
the Mamdani method in many respects. The first two parts of the fuzzy inference process,
fuzzifying the inputs and applying the fuzzy operator, are exactly the same. The main difference
between Mamdani and Sugeno is that the Sugeno output membership functions are either linear
or constant [21]. This study applies the Sugeno fuzzy inference model in order to present a
measure of the sevirity of the peptic ulcer in the output of the FIS.
4. PEPTIC ULCER SPECIFICATION
This section explains the chemical process and animals used in laboratory experiments. The
features extracted in the experiments were considered as the input of the FIS. These features are
explained in details in this section.
The present study was tested on male Wistar rats for 15 days protected the gastric mucosa against
the damage induced by indomethacin (25, 50 and 100 mg/kg) [17]. Male Wistar rats weighing
175 - 220 g were used in the study. The animals were in 6 separate groups consisting of 5 rats.
The quantitative evaluation of experimentally induced gastric lesions is a problematic and error-
prone task due to their predominantly multiple and irregularly shaped occurrence. The simplest
type of lesion index for quantification of chemically induced ulcers were described as the
cumulative length (mm) of all hemorrhagic erosions. The width of lesion has also been taken into
account (ulcer index = length -width) [17]. Figure 2 shows Microscope views of a sample
stomach of a rat with ulcer.
Figure 2. Microscope views of the rats stomach with ulcer

5. FUZZY MODELING OF THE FEATURE CHARACTERIZATION OF
PEPTIC ULCER
In order to apply soft computing techniques based on the FIS for severity assessment of the peptic
ulcer, we applied two methods as follows:
1) FCM (Fuzzy C-Means Clustering): for scatter partitioning of the input space and
automatic generation of the membership functions
2) ANFIS (Adaptive Neuro-Fuzzy inference system) for learning the FIS rules and tuning of
the membership functions
The rest of this section explains the abovementioned processes in details.
5.1. Scatter partitioning of the input space
Fuzzy c-means (FCM) is a data clustering technique wherein each data point belongs to a cluster
to some degree that is specified by a membership grade. This technique was originally introduced
by Jim Bezdek in 1981 as an improvement on earlier clustering methods. It provides a method
that shows how to group data points that populate some multidimensional space into a specific
number of different clusters.
In this study we used Fuzzy Logic Toolbox in Matlab to implement the FIS. The FCM starts with
an initial guess for the cluster centers, which are intended to mark the mean location of each
cluster. The initial guess for these cluster centers is most likely incorrect. Additionally, fcm
assigns every data point a membership grade for each cluster. By iteratively updating the cluster
centers and the membership grades for each data point, FCM iteratively moves the cluster centers
to the right location within a data set. This iteration is based on minimizing an objective function
that represents the distance from any given data point to a cluster center weighted by that data
point's membership grade Input features for assessment of the peptic ulcer are as Follows:
For each input and output variable of the FIS, three linguistic terms (Low, Medium and High)
were considered. Table 1 shows all input variables of the peptic ulcer FIS.
Table 1. The FIS input variables
No. Feature Description
1 Score 1 Each fifth petechia was calculated as 1 mm
2 Score 2 lesion length between 1 and 2 mm
5 Score 5 lesion length more than 6 mm
6 Indomethacin Explained in Section IV
7 Cimitidine Explained in Section IV
The FIS output variable were considered ulcer index (UI) which represents severity of the peptic
ulcer. The ulcer index (UI) was calculated using the following formula:
UI =( 1*S1)+(2*S2)+(3*S3)+(4*S4)+(5*S5) (1)

where S1, S2, S3, S4, S5 are related to the score 1 to score 5, respectively.
5.2. ANFIS (Adaptive Neuro-Fuzzy Inference System)
This syntax is the major training routine for Sugeno -type fuzzy inference systems. anfis uses a
hybrid learning algorithm to identify parameters of Sugeno-type fuzzy inference systems. It
applies a combination of the least-squares method and the backpropagation gradient descent
method for training FIS membership function parameters to emulate a given training data set.
ANFIS can also be invoked using an optional argument for model validation. We applied the
ANFIS for learning rules in the FIS and tuning of the membership function parameters.
The flowchart of the approach applied for learning and tuning of the FIS parameters using the
ANFIS approach is shown in Figure 3.
6. EXPERIMENTS RESULTS
In the process of the FIS parameter specification using the ANFIS model, we have a dataset
including 30 real patients diagnosed with peptic ulcer information. We partitioned the dataset into
two parts:
1) Training (70%)
2) Testing (30%)
Figures 4 to 9 represent the performance results on training and testing datasets in terms of the
root mean square error (RMSE) and the histogram of the errors. The performance results are
summarized in Table 2.
Table 2. System Performance on Train and Test datasets
Accuracy
Average(train) 99.65%
Average(test) 97.74%
The rules and membership functions of the FIS for peptic ulcer risk assessment was designed
using fuzzy c-means (FCM) clustering by extracting a set of rules that models the data behaviour
using Fuzzy Logic Toolbox and ANFIS Toolbox in Matlab are shown in Figures 9 to 14. The rule
extraction method first uses the FCM method to determine the number of rules and membership
functions for the antecedents and consequents. Then ANFIS is applied to tune the FIS parameters.
Table 3 shows the resulted FIS before training and after training process using the ANFIS
approach. The RMSE was used as performance measure during evaluation process. The result of
the RMSE and the histogram of the errors on train and test datasets are shown in Figures 3 to 8.
Table 3. Comparison of Results
Methods Accuracy
FCM 94.9%
ANFIS 98.1%

Figure 3. This section shows the Train data. Figure 4. RMSE: Shows the maximum and
Almost coincides target and Output. minimum errors in the Train Data.
Figure 5. The third part is the histogram of Figure 6. This section shows the Test data.
the errors,and shows the mean and standard There is between Output and Target.
deviation of the error.
Figure 7. RMSE: Shows the maximum and Figure 8. The third part is the histogram of
minimum errors in the Test Data. the errors, and shows the mean and
standard deviation of the error.
Figure 9. Membership functions related to Figure 10. Membership functions related to
the Score 3: lesion length between 2 and the Score 4: lesion length between 4 and 6 mm
4 mm
7. CONCLUSION
In this study, for the first time, a soft computing technique based upon fuzzy inference system
(FIS) was proposed for the problem of peptic ulcer assessment. The FIS was generated using
FCM and tuned using the ANFIS model. Performance results on a dataset including real patients
reveal the FIS with maximum accuracy of 98.1%, which reveals superiority of the approach. The

intelligent FIS system can help medical experts as a second reader for detection of the peptic ulcer
in the decision making process and consequently, improves the treatment process.
REFERENCES
[1] Umashanker, M. and S. Shruti, Traditional Indian herbal medicine used as antipyretic, antiulcer, anti-
diabetic and anticancer: A review. IJRPC, 2011. 1: 1152-1159.
[2] M. Shakeerabanu, K. Sujatha, C. Praveen-Rajneesh and A. Manimaran, “The Defensive Effect of
Quercetin on In- domethacin Induced Gastric Damage in Rats,” Advances in Biological Research,
Vol. 5, No. 1, 2011, pp. 64-70.
[3] Cadirci, E., et al., Effects of< i> Onosma armeniacum</i> root extract on ethanol-induced oxidative
stress in stomach tissue of rats. Chem Biol Int, 2007. 170(1): p. 40-48
[4] Anosike, C.A., et al., Anti-inflammatory and anti-ulcerogenic activity of the ethanol extract of ginger
(Zingiber officinale). Af J Biochem Res, 2009. 3: 379-384.
[5] M. Khanavi, R. Ahmadi, A. Rajabi, S. Jabbari-Arfaee, G. Hassanzadeh, R. Khademi, A.
Hadjiakhoondi and M. Sha- rifzadeh, “Pharmacological and Histological Effects of Centaurea
bruguierana ssp. belangerana on Indometha- cin-Induced Peptic Ulcer in Rats,” Journal of Natural
Medicines, Vol. 66, No. 2, 2012, pp. 343-349.
[6] P. Malfertheiner, F. K. Chan and K. E. McColl, “Peptic Ulcer Disease,” Lancet, Vol. 374, No. 9699,
2009, pp. 1449-1461.
[7] S. T.Lai , K.P.Fung, F.H. Ng. and K. C . Lee“A Quantitative Analvsis of Svmptoms of Non-Ulcer
Dvspepsia as Related to Age pathology, and Helicobacter Infection” 1996,vol. 31,No.11,pages 1078-
1082.
[8] G.Gopu, R.Neelaveni, K.Porkumaran “Noninvasive Technique for Acquiring and Analysis of
Electrogastrogram” IJCA Special Issue on “Computer Aided Soft Computing Techniques for Imaging
and Biomedical Applications” CASCT, 2010.
[9] Vasileios S. Charisis, Christina Katsimerou, Leontios J. Hadjileontiadis, Christos N. Liatsos, George
D. Sergiadis “Computer-aided Capsule Endoscopy Images Evaluation based on Color”.
[10] M. Liedlgrubera,_, A. Uhla “A Summary of Research Targeted at Computer-Aided Decision Support
in Endoscopy of the Gastrointestinal Tract” Technical Report 2011-01.
[11] G. G. Towell and J. W. Shavlik, ―Extracting refined rules from knowledge-based neural networks,‖
Machine Learning, vol.13, pp.71-101, 1993.
[12] C. J. Lin and C. T. Lin, ―An ART-based fuzzy adaptive learning control network,‖ IEEE
Transactions on Fuzzy Systems, vol.5, no.4, pp.477-496, 1997.
[13] C. F. Juang and C. T. Lin, ―An on-line self-constructing neural fuzzy inference network and its
applications,‖ IEEE Transactions on Fuzzy Systems, vol.6, no.1, pp.12-31, 1998.
[14] Jyh-Shing Roger Jong, Chapter 4, Fuzzy Inference System (Jang, Neuro‐Fuzzy and Soft Computing:
A Computational Approach to Learning and Machine Intelligence.)
[15] Zadeh, Fuzzy Logic, Neural Networks, and Soft Computing-1994
[16] Carlos Andre´s Pen˜a-Reyes, Moshe Sipper, A fuzzy-genetic approach to breast cancer diagnosis,
Artificial Intelligence in Medicine 17 (1999) 131–155.
[17] Ghazaleh Moghaddam1, Mohammad Sharifzadeh2, Gholamreza Hassanzadeh3, Mahnaz
Khanavi4,5, MannanHajimahmoodi; Anti-Ulcerogenic Activity of the Pomegranate Peel
(Punicagranatum) ,Published Online October 2013.
[18] M. Mazinani, J. Dehmeshki, R. Hosseini, S.D. Qanadli, "Afuzzy automatic method for measuring the
volume of the stenosis by modeling thepartial volume problem in the coronary artery", USGG 2009,
Lausanne, Switzerland(2009).
[19] RahilHosseini, J. Dehmeshki, S. A. Barman, MMazinani, S.D. Qanadli, "A Fuzzy Logic System for
Classification ofthe Lung Nodule in Digital Images in Computer Aided Detection", In Proc.of Int.
Conf. of Digital Society (ICDS), IEEE, September (2010), pp. 255- 259.
[20] RahilHosseini, S. D. Qanadli, S. Barman, M. Mazinani, T.Ellis, and J. Dehmeshki, "An Automatic
Approach for Learning and Tuning Gaussian Interval Type-2 Fuzzy Membership Functions Applied
to Lung CAD Classiﬁcation System", IEEE Transactions on Fuzzy Systems, vol. 20, No. 2, April
(2012), pp. 224-234.
[21] Sugeno, M., Industrial applications of fuzzy control, Elsevier Science Pub. Co., 1985.

AUTHORS
Kianaz Rezaei (born February, 1990), obtained her B.Sc. degree in Computers-Soft
computing from Islamic Azad University, Share-e-Qods Branch, Tehran, Iran during 2012,
and currently pursuing Masters Degree from Kharazmi University, Tehran, Iran.
Rahil Hosseini is assistant professor in software engineering and soft computing at Azad
University. She is a member of the Digital Imaging Research Centre (DIRC) in the Faculty
of Science, Engineering and Computing at Kingston University, London. She received her
BSc at Teacher training University in Tehran followed by her MSc in software engineering.
She commenced her PhD at Kingston University London. Her main area of interest in
research is in the fields of soft computing, fuzzy modelling and pattern recognition in data
mining problems and medical image analysis, specifically cancer diagnosis. She has professional work
experience in the field of data mining, medical image analysis and fuzzy modelling for uncertain
environments. She has published journal and conference papers in data mining, soft computing, distributed
systems, fuzzy modelling, and pattern recognition for medical image analysis and cancer diagnosis.
Mahdi Mazinani is assistant professor in electronic engineering and image processing at
Azad University. He is a member of the Quantitative Medical Imaging International
Institutes (QMI3) and Digital Imaging Research Centre in the Faculty of Science,
Engineering and Computing at Kingston University, London. After completing his BSc
degree in Electronic engineering at Semnan University (securing the first rank) he
completed his MSc degree in the Semnan University with first rank. He commenced his
PhD. His main area of interest in research is in the field of medical image analysis. His work is focused on
research into novel image analysis techniques to enable the recognition and quantification using CTA
images.
He has published papers in the detection and quantification of the coronary arteries using fuzzy and medical
image analysis techniques.

A fuzzy inference system for assessment of the severity of the peptic ulcers

More Related Content

What's hot (18)

Viewers also liked (20)

Similar to A fuzzy inference system for assessment of the severity of the peptic ulcers (20)

Recently uploaded (20)

A fuzzy inference system for assessment of the severity of the peptic ulcers