Diagnosis of the diseases––using a GA-fuzzy approach

doi:10.1016/j.ins.2004.03.004

Information Sciences

Volume 162, Issue 2, 17 May 2004, Pages 105-120

https://doi.org/10.1016/j.ins.2004.03.004 Get rights and content

Abstract

The objective of our study is to design an expert system by modelling the knowledge and thinking process of a doctor. A fuzzy logic controller (FLC) is used to model the process and a genetic algorithm (GA) helps to select a number of good rules from a manually constructed large rule base of an FLC, based on the opinion of 10 doctors. The GA-based tuning is done off-line. Once the optimized rule base of the FLC is obtained, it can diagnose the disease, on-line. The scope of the present work has been extended to two diseases, namely Pneumonia and Jaundice. The symptoms of each disease are fed as inputs to the FLC and the output, i.e., grade of a disease is determined.

Introduction

Diagnosis plays a vital role in medical treatment. A disease has to be correctly diagnosed by a doctor before he/she prescribes some medicines. It is a common knowledge that if a patient with a set of certain symptoms goes to different doctors, he may get different opinions regarding the type or grade of the underlying disease. Also, two different persons with similar symptoms going to the same doctor may be diagnosed differently. This indicates that there is a certain degree of fuzziness in the thinking process of a doctor.

Fuzzy logic controller (FLC), a successful application of Zadeh's fuzzy set theory [13], is a potential tool for dealing with uncertainty and imprecision. The working principle of an FLC is explained in Appendix A. Thus, the knowledge of a doctor can be modelled using an FLC. The performance of an FLC depends on its knowledge base which consists of a data base and a rule base. It is observed that the performance of an FLC mainly depends on its rule base, and optimizing the membership function distributions stored in the data base is a fine tuning process [10]. The main disadvantage of using the FLC is that it requires an extensive knowledge of the system to be controlled. Such an extensive knowledge may be difficult to obtain beforehand. Based on the user's knowledge, he/she can design the rule base of an FLC but there is no guarantee that it will perform well. Thus, tuning is required of the manually constructed rule base of an FLC so that it can perform well. In this paper, we use a genetic algorithm (GA) to optimize an already-designed rule base of an FLC which will be used to diagnose the disease. The working cycle of a GA is discussed in Appendix B.

Several attempts were made by various researchers to develop suitable expert systems for making automatic diagnosis of the diseases. MYCIN system [11] was developed at Stanford University to aid physicians in diagnosing and treating patients with infectious blood diseases caused by bacteria and meningitis. The MYCIN system did not use fuzzy logic and it is a rule-based expert system based on Boolean algebra. Fuzzy logic technique was also used to design several expert systems, namely CADIAG [1], SPHINX [3], RENOIR [2], CLINAID [6] and others, to generate the diagnostic information. Moreover, FLORIDA [9] was another expert system which determines the physiological condition of patients in ICU using fuzzy logic and knowledge bases. Soula et al. [12] developed an expert system using the fuzzy logic for treatment of diabetes. Fuzzy logic controllers had also been used for controlling drug infusion to maintain an adequate level of anesthesia by monitoring blood pressure and muscle relaxation [5]. Another system based on the fuzzy logic was proposed by Kovarlerchuk et al. [8] which aids the diagnosis of breast cancer by analyzing the lobulation.

The rest of the text is organized as follows: Section 2 describes the problem and explains the developed algorithm to solve it. The results are discussed in Section 3 and conclusions are made in Section 4. Section 5 suggests the scope for future work.

Section snippets

Description of the problem and proposed algorithm

Two different diseases, namely Pneumonia and Jaundice are considered here.

Results and discussion

The effectiveness of the developed algorithm is tested on two diseases, namely Pneumonia and Jaundice. Two different approaches are developed as discussed in Section 2. The rule base of the FLCs obtained using Approach 2 for modelling the cases of Pneumonia and Jaundice are shown in Table 3, Table 5, respectively. Table 3 shows that the GA has selected 42 rules out of a total of 81, whereas 39 rules are found to be good by the GA, in case of Jaundice (refer to Table 5). It is important to

Conclusion

Realizing the fact that there is enough fuzziness in the thinking process of a doctor, doctor's knowledge of the disease has been modelled using a fuzzy logic controller. A GA-based tuning is adopted to improve the knowledge base of an FLC. The optimized FLC is able to diagnose a disease effectively for a set of input symptoms, in the absence of a doctor. Thus, an expert system has been developed to diagnose the disease automatically.

Scope for future work

The present work touches only the tip of the iceberg. A vast scope exists to improve upon it. The number of input factors taken up for Jaundice and Pneumonia can be greatly increased for a more comprehensive study of the symptoms. The data collected from the 10 doctors could be replaced by a data-bank collected from a large number of doctors. This will help to give a more realistic solution to the problem. The effectiveness of the developed algorithm will be tested, in future, on some other

Acknowledgements

The authors are profoundly grateful to the anonymous reviewers whose suggestions helped to significantly improve the quality of the paper. Special thanks are due to the doctors who shared their ideas and without whose help this project would not have been a possibility.

References (13)

D.K. Pratihar et al.
A genetic-fuzzy approach for mobile robot navigation among moving obstacles
Int. J. Approx. Reason.
(1999)
L.A. Zadeh
Fuzzy sets
Inform. Contr.
(1965)
K.P. Adlassnig
Fuzzy set theory in medical diagnosis
IEEE Trans. Syst. Man Cybernet.
(1986)
M. Belmonte-Serrano et al.
RENOIR: an expert system using fuzzy logic for rheumatology diagnosis
Int. J. Intell. Syst.
(1994)
M. Fieschi et al.
SPHINX––a system for computer-aided diagnosis
Methods Inform. Med.
(1982)
D.E. Goldberg
Genetic Algorithms in Search, Optimization, Machine Learning
(1989)

There are more references available in the full text version of this article.

Cited by (19)

Bidirectional approximate reasoning-based approach for decision support
2020, Information Sciences
Citation Excerpt :
Fuzzy logic based approximate reasoning offers a practically applicable approach to dealing with uncertainty and uncertainty management in decision support [7]. For instance, a fuzzy logic based expert system has been developed to diagnose common diseases such as pneumonia and jaundice, which is capable of returning an identified disease as output given observed symptoms [22]. Also for medical decision support, there have been many hybrid systems which work by integrating fuzzy logic and other computational intelligence mechanisms.
Fuzzy rule-based systems are widely applied for real-world decision support, such as policy formation, public health analysis, medical diagnosis, and risk assessment. However, they face significant challenges when the application problem at hand suffers from the “curse of dimensionality” or “sparse knowledge base”. Combination of hierarchical fuzzy rule models and fuzzy rule interpolation offers a potentially efficient and effective approach to dealing with both of these issues simultaneously. In particular, backward fuzzy rule interpolation (B-FRI) facilitates approximate reasoning to be performed given a sparse rule base where rules do not fully cover all observations or the observations are not complete, missing antecedent values in certain available rules. This paper presents a hierarchical bidirectional fuzzy reasoning mechanism by integrating hierarchical rule structures and forward/backward rule interpolation. A computational method is proposed, building on the resulting hierarchical bidirectional fuzzy interpolation to maintain consistency in sparse fuzzy rule bases. The proposed techniques are utilised to address a range of decision support problems, successfully demonstrating their efficacy.
Knowledge discovery in medicine: Current issue and future trend
2014, Expert Systems with Applications
Citation Excerpt :
Vatankhah, Asadpour, and Fazel-Rezai (2012) and Bojarczuk, Lopes, and Freitas (2000) try to address pain identification by adaptive neuro fuzzy inference system + SVM hybrid and a combination of decision tree and genetic programming. Roychowdhury, Pratihar, Bose, Sankaranarayanan, and Sudhahar (2004) diagnosed jaundice and pneumonia based on an expert system and genetic fuzzy algorithm. Magoulas, Plagianakos, and Vrahatis (2004) investigated computer-assisted colonoscopy image diagnosis by evolutionary algorithm + artificial neural network.
Data mining is a powerful method to extract knowledge from data. Raw data faces various challenges that make traditional method improper for knowledge extraction. Data mining is supposed to be able to handle various data types in all formats. Relevance of this paper is emphasized by the fact that data mining is an object of research in different areas. In this paper, we review previous works in the context of knowledge extraction from medical data. The main idea in this paper is to describe key papers and provide some guidelines to help medical practitioners. Medical data mining is a multidisciplinary field with contribution of medicine and data mining. Due to this fact, previous works should be classified to cover all users’ requirements from various fields. Because of this, we have studied papers with the aim of extracting knowledge from structural medical data published between 1999 and 2013. We clarify medical data mining and its main goals. Therefore, each paper is studied based on the six medical tasks: screening, diagnosis, treatment, prognosis, monitoring and management. In each task, five data mining approaches are considered: classification, regression, clustering, association and hybrid. At the end of each task, a brief summarization and discussion are stated. A standard framework according to CRISP-DM is additionally adapted to manage all activities. As a discussion, current issue and future trend are mentioned. The amount of the works published in this scope is substantial and it is impossible to discuss all of them on a single work. We hope this paper will make it possible to explore previous works and identify interesting areas for future research.
Domain experts' knowledge-based intelligent decision support system in occupational shoulder and neck pain therapy
2011, Applied Soft Computing Journal
This research develops a fuzzy knowledge-based decision support system (FKBDSS) that measures and predicts the degree of severity of the work-related risk associated with shoulder and neck pain (SNP) that is a prevalent pain complaint in an occupational environment. Assessing the harshness of SNP is a dreary chore, since the risk factors are featured with imprecision, uncertainty and vagueness. Predicting SNP subjective risk level provides key decision support information to medical practitioners in diagnosis. The objective involves knowledge acquisition performed through literature analysis, traditional and concept mapping interviews with domain experts comprising neurologist, orthopaedist, psychologist and physiotherapist to identify risk factors that include mechanical, physical and psychosocial categories. The determination of ranking the relative factor importance has accomplished using analytic hierarchy processing (AHP) analysis. The linguistic variables that qualify risk levels are quantified using fuzzy set theory (FST) that provides linguistic and numeric value outputs to predict the hazard level of SNP.
Knowledge and intelligent computing system in medicine
2009, Computers in Biology and Medicine
Knowledge-based systems (KBS) and intelligent computing systems have been used in the medical planning, diagnosis and treatment. The KBS consists of rule-based reasoning (RBR), case-based reasoning (CBR) and model-based reasoning (MBR) whereas intelligent computing method (ICM) encompasses genetic algorithm (GA), artificial neural network (ANN), fuzzy logic (FL) and others. The combination of methods in KBS such as CBR–RBR, CBR–MBR and RBR–CBR–MBR and the combination of methods in ICM is ANN–GA, fuzzy–ANN, fuzzy–GA and fuzzy–ANN–GA. The combination of methods from KBS to ICM is RBR–ANN, CBR–ANN, RBR–CBR–ANN, fuzzy–RBR, fuzzy–CBR and fuzzy–CBR–ANN. In this paper, we have made a study of different singular and combined methods (185 in number) applicable to medical domain from mid 1970s to 2008. The study is presented in tabular form, showing the methods and its salient features, processes and application areas in medical domain (diagnosis, treatment and planning). It is observed that most of the methods are used in medical diagnosis very few are used for planning and moderate number in treatment. The study and its presentation in this context would be helpful for novice researchers in the area of medical expert system.
Encoding fuzzy possibilistic diagnostics as a constrained optimization problem
2008, Information Sciences
This paper discusses a knowledge-base encoding methodology for diagnostic tasks, transforming such knowledge into constrained optimization problems. The methodology is based on a reinterpretation of the consistent causal reasoning paradigm [D. Dubois, H. Prade, Fuzzy relation equations and causal reasoning, Fuzzy Sets and Systems 45 (2) (1995) 119–134] as an equivalent problem of feasibility subject to equality and inequality constraints (in the binary case). Then, it is extended to the fuzzy case. Preferences under uncertain knowledge are incorporated by transforming the feasibility problem into an optimization one, which may be interpreted in possibilistic terms. The problem is solved by efficient, widely-known, linear and quadratic programming tools, which are able to cope with large-scale problems. Examples illustrating some of the concepts and possibilities of the proposed procedure, as well as a summary comparison with other approaches are also discussed.
Optimal two-degree-of-freedom fuzzy control for locomotion control of a hydraulically actuated hexapod robot
2007, Information Sciences
Locomotion control of legged robots is a very challenging task because very accurate foot trajectory tracking control is necessary for stable walking. An electro-hydraulically actuated walking robot has sufficient power to walk on rough terrain and carry a heavier payload. However, electro-hydraulic servo systems suffer from various shortcomings such as a high degree of nonlinearity, uncertainty due to changing hydraulic properties, delay due to oil flow and dead-zone of the proportional electromagnetic control valves. These shortcomings lead to inaccurate analytical system model, therefore, application of classical control techniques result into large tracking error. Fuzzy logic is capable of modeling mathematically complex or ill-defined systems. Therefore, fuzzy logic is becoming popular for synthesis of control systems for complex and nonlinear plants. In this investigation, a two-degree-of-freedom fuzzy controller, consisting of a one-step-ahead fuzzy prefilter in the feed-forward loop and a PI-like fuzzy controller in the feedback loop, has been proposed for foot trajectory tracking control of a hydraulically actuated hexapod robot. The fuzzy prefilter has been designed by a genetic algorithm (GA) based optimization. The prefilter overcomes the flattery delay caused by the hydraulic dead-zone of the electromagnetic proportional control valve and thus helps to achieve better tracking. The feedback fuzzy controller ensures the stability of the overall system in the face of model uncertainty associated with hydraulically actuated robotic mechanisms. Experimental results exhibit that the proposed controller manifests better foot trajectory tracking performance compared to single-degree-of-freedom (SDF) fuzzy controller or optimal classical controller like state feedback LQR controller.

View all citing articles on Scopus

View full text

Diagnosis of the diseases––using a GA-fuzzy approach

Abstract

Introduction

Section snippets

Description of the problem and proposed algorithm

Results and discussion

Conclusion

Scope for future work

Acknowledgements

Int. J. Approx. Reason.

Inform. Contr.

Fuzzy set theory in medical diagnosis

IEEE Trans. Syst. Man Cybernet.

RENOIR: an expert system using fuzzy logic for rheumatology diagnosis

Int. J. Intell. Syst.

SPHINX––a system for computer-aided diagnosis

Methods Inform. Med.

Genetic Algorithms in Search, Optimization, Machine Learning