Systematic image processing for diagnosing brain tumors: A Type-II fuzzy expert system approach

doi:10.1016/j.asoc.2009.11.019

Applied Soft Computing

Volume 11, Issue 1, January 2011, Pages 285-294

https://doi.org/10.1016/j.asoc.2009.11.019 Get rights and content

Abstract

This paper presents a systematic Type-II fuzzy expert system for diagnosing the human brain tumors (Astrocytoma tumors) using T₁-weighted Magnetic Resonance Images with contrast. The proposed Type-II fuzzy image processing method has four distinct modules: Pre-processing, Segmentation, Feature Extraction, and Approximate Reasoning. We develop a fuzzy rule base by aggregating the existing filtering methods for Pre-processing step. For Segmentation step, we extend the Possibilistic C-Mean (PCM) method by using the Type-II fuzzy concepts, Mahalanobis distance, and Kwon validity index. Feature Extraction is done by Thresholding method. Finally, we develop a Type-II Approximate Reasoning method to recognize the tumor grade in brain MRI. The proposed Type-II expert system has been tested and validated to show its accuracy in the real world. The results show that the proposed system is superior in recognizing the brain tumor and its grade than Type-I fuzzy expert systems.

Introduction

The incidence of brain tumors has increased over the time and differs according to gender, age, race, and geography. Most of this increase probably is attributable to improvements in diagnostic imaging methods, increased availability of medical care and neurosurgeons, changing approaches in treatment of older patients, and changes in classifications of specific histologies of brain tumors. Survival time after brain tumor diagnosis varies greatly by histologic type and age at diagnosis. Moreover, some of malignant brain tumors such as Glioblastomas may develop suddenly or by way of malignant progression from lower grades [18]. Therefore, diagnosing the brain tumors in an appropriate time is very essential for further treatments.

In recent years, neurology and basic neuroscience have been significantly advanced by imaging tools that enable in vivo monitoring of the brain. In particular, Magnetic Resonance Imaging (MRI) has proven to be a powerful and versatile brain imaging modality that allows noninvasive longitudinal and 3D assessment of tissue morphology, metabolism, physiology, and function [44]. The information MRI provides, has greatly increased the knowledge of normal and diseased anatomy for medical research, and is a critical component in diagnosis and treatment planning [46].

The goal of this paper is to design, implement, and evaluate a Type-II Fuzzy Image Processing expert system to diagnose brain tumors, especially Astrocytoma tumors in T₁-weighted MR Images with contrast. The framework of fuzzy sets, systems, and relations is very useful to deal with the absence of sharp boundaries of the sets of symptoms, diagnoses, and phenomena of diseases. However, there are many uncertainties and vaguenesses in images, which are very difficult to handle with Type-I fuzzy sets. These fuzzy sets are not able to model such uncertainties directly because their membership functions are crisp. On the other hand, Type-II fuzzy sets are able to model such uncertainties as their membership functions are themselves fuzzy [20]. Therefore, Type-II fuzzy logic systems have the potential to provide better performance [19]. For these reasons, we have concentrated on Type-II fuzzy modeling.

The rest of this paper is organized as follows: Section 2 explains the brain tumors shortly. Section 3 addresses the Brain tumors diagnosis. Section 4 presents the fundamentals of Type-II fuzzy logic. Image processing approaches are described in Section 5. Section 6 is dedicated to the proposed image processing approach. The experimental results are presented in Section 7. Conclusions and future works are presented in Section 8. Finally, the designed software, based on the proposed method, is presented in Appendix A.

Section snippets

Brain tumors

Gliomas are groups of tumors that arise in the Central Nervous System (CNS) and are divided into four major categories: Astrocytes, Oligodendrocytes, Ependymal Cells, and Microglia [29]. In addition, tumors of Glial origin can be divided into those that are infiltrated into normal brain structures (diffuse tumors) and those with more discrete boundaries (focal tumors).

Astrocytoma is the most common type of glial tumors (30% of all brain tumors) and is a usually a malignant one. Astrocytoma can

Tumors diagnosing

History, physical and neurologic examination, and laboratory findings, especially imaging, are the common elements for diagnosing of Central Nervous System (CNS) tumors [31]. In the last three decades, the development of more sophisticated and advanced imaging techniques has led to improved diagnostic accuracy. Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) enable physicians to diagnose brain tumors that previously might have been diagnosed incorrectly as strokes or other

Type-II fuzzy logic

Fuzzy logic theory has been successfully applied to many areas, such as pattern recognition and computer vision. It has also been applied to image enhancement, image segmentation, image classification, and thresholding value selection [28]. In the field of image processing, we usually encounter with many uncertainties such as uncertainties caused by projecting a 3D object into a 2D image, digitizing analog pictures, uncertain boundaries, and non-homogeneous regions. However, Type-I fuzzy sets

Image processing

The term image processing is generally applied to the methods that receive an image as an input and produce a modified image, measurements or descriptions as an output. In the literature, there are many algorithms developed for image processing [27], [49], [33]. In the field of image processing, there may have been many uncertainties in images: uncertainties caused by [35]:

•
Image specification (such as projecting a 3D object into a 2D image, digitizing analog pictures, uncertain boundaries, and

Proposed image processing approach

Developing an efficient diagnosing method may help physicians to diagnose tumors in an appropriate time. Designing an expert system for this purpose includes two main steps: designing strategy of building Type-II fuzzy system and applying this strategy to the application area.

Step 1—designing strategy of building Type-II fuzzy system: There are two methods for generating fuzzy systems, supervised and unsupervised learning [38]. For designing the Type-II image processing system, we use

Experimental results

The proposed approach is used to design software called MRI processing, which is shortly described in appendix. Two steps are used to validate the proposed approach:

Step 1—evaluating the performance of the proposed approach and software: This step is done by using 95 patient's information. This information includes both Normal and Abnormal brain T₁-weighted MR images and patient's age. The process starts with pre-processing step, continued by Segmentation and Feature Extraction steps. The

Conclusion and future works

In this research, a Type-II expert system for brain tumor image processing has been developed. The proposed system can help the physicians to better diagnosis the human brain tumors, for further treatment. The main contributions in this paper were the aggregation of the available image pre-processing methods, development of a Type-II fuzzy cluster analysis for segmentation, and presenting a Type-II fuzzy expert system for approximate reasoning. The presented system has been encoded and its main

References (55)

A. Hassanien
Fuzzy rough sets hybrid scheme for breast cancer detection
Image and Vision Computing
(2007)
E. Nasibov et al.
A new unsupervised approach for fuzzy clustering
Fuzzy Sets and Systems
(2007)
H.D. Cheng et al.
Automated breast cancer detection and classification using ultrasound images: a survey
Pattern Recognition
(2010)
I.B. Turksen
Type 2 representation and reasoning for CWW
Fuzzy Sets and Systems
(2002)
J. Tang
A Multi-direction GVF snake for the segmentation of skin cancer images
Pattern Recognition
(2009)
J.L. Fisher et al.
Epidemiology of brain tumors
Neurologic Clinics
(2007)
J. Alcalá-Fdez et al.
Learning the membership function contexts for mining fuzzy association rules by using genetic algorithms
Fuzzy Sets and Systems
(2009)
L.A. Zadeh
The concept of a linguistic variable and its applications to approximate reasoning-I
Information Sciences
(1975)
L. Cinque et al.
A clustering fuzzy approach for image segmentation
Pattern Recognition
(2004)
L.M. Fletcher-Heath et al.
Automatic segmentation of non-enhancing brain tumors in magnetic resonance images
Artificial Intelligence in Medicine
(2001)

D. Van De Ville et al.

Noise reduction by fuzzy image filtering

IEEE Transactions on Fuzzy Systems

(2003)

D.E. Gustafson et al.

Fuzzy clustering with a fuzzy covariance matrix

E. Trucco et al.

Introductory Techniques for 3D Computer Vision

(1998)

E.E. Kerre et al.

Fuzzy Techniques in Image Processing: Techniques and Application

(2000)

H.R. Tizhoosh

Fuzzy image processing: potentials and state of the art

H.R. Tizhoosh et al.

On fuzzy enhancement of megavoltage images in radiation therapy

I.B. Turksen

An Ontological and Epistemological Perspective of Fuzzy Set Theory

(2006)

Cited by (104)

Reliable image metrics-based brain tumor analysis using sensor deep learning technologies
2023, Measurement: Sensors
In this paper, an attempt has been made to employ deep learning technique to predict the phase of brain tumors. The deep learning methods help the practitioners to correlate patients’ status in hand with similar subjects and assess as well as predict the future anomalies due to brain tumor. This certainly emerges as a handy tool for remote medication, especially citing the current pandemic needs. On the other hand, the digitization of the images followed by intelligent analysis can provide a first impression in a quick which usually consumes lot manual computations and cross verifications. Deep learning has been the most successful tool to handle supervised classification while dealing with complex patterns. A fully automated deep learning method can be developed with the existing toolboxes in the simulation environment. The purpose of the study is to apply this machine learning technique to classifying images of brains with different types of tumors: meningioma, glioma, and pituitary. The simulation is carried in Matlab environment and analysis is carried out using standard metrics. The confusion matrix shows a maximum of 95% and minimum of 89% accuracy. The results are compared with the Convolutional Neural Network (CNN) and Artificial Neural Network models (ANN).
Image processing: an early detection tool for Alzheimer’s disease
2023, Smart Diagnostics for Neurodegenerative Disorders: Neuro-sensors
Alzheimer’s disease (AD) is a progressive irreversible neurodegenerative condition that typically manifest an irreparable damage to the brain with long-lasting functional impairment. It has been severely affecting older population worldwide and thus, its early-stage detection has become a topic of major research interest. Over the past few decades, advances in neuroimaging research such as positron emission tomography (PET), computed tomography (CT), magnetic resonance imaging (MRI), fluoro-deoxy-D-glucose (FDG) and Pittsburgh Compound-B (PiB), amyloid tracers, has played a variety of roles in the study of AD. These imaging techniques are largely complementary consisting strengths and weakness of their own, expensive, time-consuming and has limited accessibility. In this regard, artificial intelligence (AI) technology has fueled interest which has opened its door to the scientific community. It has become a promising approach and cemented itself in the fields of image analysis, language processing, computer vision, big data interpretation as well as in the healthcare industries with the potential to handle enormous quantities of clinical, biochemical, environmental, and lifestyle data, AI techniques for AD research have demonstrated promising results, and offers precise, efficient, and practical ways for assessing AD diagnoses. The move from the application of sequential imaging modalities to multi-modal imaging would help impart solutions to systemic factors. This chapter describes the recent advances and significance of AI technology for image analysis for the early detection of AD.
An overview of artificial intelligence techniques for diagnosis of Schizophrenia based on magnetic resonance imaging modalities: Methods, challenges, and future works
2022, Computers in Biology and Medicine
Schizophrenia (SZ) is a mental disorder that typically emerges in late adolescence or early adulthood. It reduces the life expectancy of patients by 15 years. Abnormal behavior, perception of emotions, social relationships, and reality perception are among its most significant symptoms. Past studies have revealed that SZ affects the temporal and anterior lobes of hippocampus regions of the brain. Also, increased volume of cerebrospinal fluid (CSF) and decreased volume of white and gray matter can be observed due to this disease. Magnetic resonance imaging (MRI) is the popular neuroimaging technique used to explore structural/functional brain abnormalities in SZ disorder, owing to its high spatial resolution. Various artificial intelligence (AI) techniques have been employed with advanced image/signal processing methods to accurately diagnose SZ. This paper presents a comprehensive overview of studies conducted on the automated diagnosis of SZ using MRI modalities. First, an AI-based computer aided-diagnosis system (CADS) for SZ diagnosis and its relevant sections are presented. Then, this section introduces the most important conventional machine learning (ML) and deep learning (DL) techniques in the diagnosis of diagnosing SZ. A comprehensive comparison is also made between ML and DL studies in the discussion section. In the following, the most important challenges in diagnosing SZ are addressed. Future works in diagnosing SZ using AI techniques and MRI modalities are recommended in another section. Results, conclusion, and research findings are also presented at the end.
A concrete reformulation of fuzzy arithmetic
2021, Expert Systems with Applications
Advancement in fuzzy arithmetic is essential to advancing the applicability of fuzzy set theory (FST) to approximate reasoning problems arose in engineering, medicine, managerial science and many other domains. In the recent FST literature, many of the noteworthy progresses in fuzzy arithmetic have been built upon increasingly sophisticated mathematical concepts and structures, leading to highly non-trivial results. In this article, we take a different approach, of offering a simple, concrete framework that could serve as an entry point for the classically-trained engineers (those who might not be well-acquainted with deep theoretical FST advances) to understand and use fuzzy arithmetic in their applications. Specifically, the contributions of this work are: (i) We extend a foundational model previously proposed in Ngan (2018) to arrive at a simple fuzzy arithmetic framework, in which each key element within the framework possesses simple, concrete meaning to these classically-trained engineers; (ii) We describe how to use these key elements as building blocks to create various concrete fuzzy arithmetic operators that are adaptive to various assumptions. This means that the classically-trained engineers can now easily customize and use fuzzy arithmetic robustly in their application domains; (iii) We demonstrate that the proposed framework can overcome the possibility of generating pathological results that were seen in other fuzzy arithmetic approaches in the FST literature; (iv) We demonstrate the utility of the proposed framework in solving multiple-criteria-decision-making (MCDM) problems concretely and in an easy-to-interpret manner (which are highly important considerations for the users of MCDM). Last but not least, we offer a new avenue for extending this fuzzy arithmetic framework to higher-order fuzzy settings, such as the type-2 fuzzy and other type-2-like fuzzy settings. Thus, our framework can potentially provide a unified, robust approach for developing fuzzy arithmetic in a wide range of FST settings.
T2-FDL: A robust sparse representation method using adaptive type-2 fuzzy dictionary learning for medical image classification
2020, Expert Systems with Applications
Citation Excerpt :
Neural networks, deep learning, and some of their extensions are taken into consideration by Hermessi, Mourali, and Zagrouba (2019), Roy, Sadhu, Bandyopadhyay, Bhattacharyya, and Kim (2016), Gao, Hui, and Tian (2017), Havaei et al. (2017), Zhou et al. (2019), Thiyagarajan and Pandurangan (2015), Anaraki, Ayati, and Kazemi (2018) and Patil, Naik, Pai, and Gad (2018). In Sharma and Mukharjee (2012), Papageorgiou et al. (2008), Zarinbal, Zarandi, Turksen, and Izadi (2015), Zarandi, Zarinbal, and Izadi (2011), Selvapandian and Manivannan (2018) and Raju, Suresh, and Rao (2018) fuzzy systems are evaluated as efficient ways for tumor classification. Dictionary learning is a powerful method in classification, detection, and analysis of tumor images and has shown acceptable accuracy (Deshpande, Maurel, & Barillot, 2015; Al-Shaikhli, Yang, & Rosenhahn, 2016).
In this paper, a robust sparse representation for medical image classification is proposed based on the adaptive type-2 fuzzy learning (T2-FDL) system. In the proposed method, sparse coding and dictionary learning processes are executed iteratively until a near-optimal dictionary is obtained. The sparse coding step aiming at finding a combination of dictionary atoms to represent the input data efficiently, and the dictionary learning step rigorously adjusts a minimum set of dictionary items. The two-step operation helps create an adaptive sparse representation algorithm by involving the type-2 fuzzy sets in the design process of image classification. Since the existing image measurements are not made under the same conditions and with the same accuracy, the performance of medical diagnosis is always affected by noise and uncertainty. By introducing an adaptive type-2 fuzzy learning method, a better approximation in an environment with higher degrees of uncertainty and noise is achieved. The experiments are executed over two open-access brain tumor magnetic resonance image databases, REMBRANDT and TCGA-LGG, from The Cancer Imaging Archive (TCIA). The experimental results of a brain tumor classification task show that the proposed T2-FDL method can adequately minimize the negative effects of uncertainty in the input images. The results demonstrate the outperformance of T2-FDL compared to other important classification methods in the literature, in terms of accuracy, specificity, and sensitivity.
An improved interval type-2 possibilistic C-means clustering algorithm for interclass maximization
2024, Journal of Intelligent and Fuzzy Systems

View all citing articles on Scopus

View full text

Systematic image processing for diagnosing brain tumors: A Type-II fuzzy expert system approach

Abstract

Introduction

Section snippets

Brain tumors

Tumors diagnosing

Type-II fuzzy logic

Image processing

Proposed image processing approach

Experimental results

Conclusion and future works

Image and Vision Computing

Fuzzy Sets and Systems

Pattern Recognition

Fuzzy Sets and Systems

Pattern Recognition

Neurologic Clinics

Fuzzy Sets and Systems

Information Sciences

Pattern Recognition

Artificial Intelligence in Medicine

Pattern Recognition

Pattern Recognition Letters

Fuzzy Sets and Systems

Expert Systems with Applications

Pattern Recognition

Neuroimage

Pattern Recognition

International Journal of Approximate Reasoning

Current methods in the automatic tissue segmentation of 3D magnetic resonance brain images

Current Medical Imaging Reviews

Toward breast cancer diagnosis based on automated segmentation of masses in mammograms

Pattern Recognition

Noise reduction by fuzzy image filtering

IEEE Transactions on Fuzzy Systems

Fuzzy clustering with a fuzzy covariance matrix

Introductory Techniques for 3D Computer Vision

Fuzzy Techniques in Image Processing: Techniques and Application

Fuzzy image processing: potentials and state of the art

On fuzzy enhancement of megavoltage images in radiation therapy

An Ontological and Epistemological Perspective of Fuzzy Set Theory