ABSTRACT

BACKGROUND/AIMS

In medicine, one of the most important applications of intelligent systems is the fuzzy neural network (FNN) framework, which can be used in the diagnosis and treatment decision-making. We investigated a novel procedure in an integrated fuzzy neural structure (multiinput and -output) based on the Takagi-Sugeno-Kang (TSK)-type rule for the classification of erythemato-squamous diseases.

MATERIAL and METHODS

Designing an FNN system was intelligently aimed at the differential diagnosis of erythemato-squamous disease. Dataset explored for this research included detailed records of diagnosed patients. From the training dataset, our proposed algorithm learns from the domain to differentiate a new case.

RESULTS

Total performance of the inference system was empirically evaluated in terms of classification accuracy, with a total accuracy of 98.37%. Comparison of this result with those of other algorithms by other researchers on the same domain showed that our algorithm was considerably outstanding.

CONCLUSION

The goal of this study is to investigate the capability of the FNN classifier tested on a real-world dataset for the diagnosis of erythematosquamous diseases. To simplify the uncertainties discovered in the dataset, we integrated the learning capabilities of both the fuzzy logic and neural network. The integrated classifier was used to adequately classify the input space of the domain into the corresponding classes of the erythemato-squamous diseases. The high performance accuracy recorded by our proposed system depicts that the system could be applied to classifying new cases of erythemato-squamous diseases.