A Framework for Fuzzy Expert System Creation—Application to Cardiovascular Diseases

Abstract

A methodology for the automated development of fuzzy expert systems is presented. The idea is to start with a crisp model described by crisp rules and then transform them into a set of fuzzy rules, thus creating a fuzzy model. The adjustment of the model's parameters is performed via a stochastic global optimization procedure. The proposed methodology is tested by applying it to problems related to cardiovascular diseases, such as automated arrhythmic beat classification and automated ischemic beat classification, which, besides being well-known benchmarks, are of particular interest due to their obvious medical diagnostic importance. For both problems, the initial set of rules was determined by expert cardiologists, and the MIT-BIH arrhythmia database and the European ST-T database are used for optimizing the fuzzy model's parameters and evaluating the fuzzy expert system. In both cases, the results indicate an escalation of the performance from the simple initial crisp model to the more sophisticated fuzzy models, proving the scientific added value of the proposed framework. Also, the ability to interpret the decisions of the created fuzzy expert systems is a major advantage compared to "black box" approaches, such as neural networks and other techniques.

ECG-based heartbeat classification for arrhythmia detection: A survey

“…Many other methods for arrhythmia classification have been developed using other machine learning and data mining algorithms, such as decision trees [125,126,68], nearest neighbors [127,128,129], clustering [73,130,131], hidden Markov models [132,133], hyperbox classifiers [105], optimum-path forest [134], conditional random fields [8] and rules-based models [135,67,136].…”

“…The set of rules presented by Tsipouras et al [135,67,136] was obtained together with cardiologists and are related to a morphological tachogram for arrhythmic events. Methods constructed in conjunction with rules usually present a worser performance, in terms of effectiveness, when compared to other methods proposed in literature.…”

A decision-making mechanism for context inference in pervasive healthcare environments

“…A paper by [26] describes a methodological framework for the automated generation of fuzzy expert systems, which are based on an initial crisp model for cardiovascular domain problems. The fuzzy models produced are tuned using global optimization, i.e., optimizing their parameters to fit an arrhythmia database.…”

“…In this context, one interesting approach is to use a hybrid technique for model generation, i.e., neither entirely based on medical knowledge nor entirely based on an annotated dataset but making use of both. Some studies (e.g., [2,26]) have explored this approach, which can involve the following steps: (i) creating initial medical rules, (ii) fuzzification of these fuzzy rules to create the model, and (iii) updating the parameters of the fuzzy model based on the analysis of an annotated dataset.…”

Fuzzy Expert Systems (FES) for Medical Diagnosis

“…[140][141][142] The architectures are proposed for the implementation and improving performance of a rule based diagnostic decision support systems related to medical diagnosis. [143][144][145] The powerful, generic as well as hybrid expert systems for diagnosis and remodeling of existing systems are also been reported.…”