A Comparison Between Fuzzy Type-1 and Type-2 Systems in Medical Decision Making: A Systematic Review

Objectives: Fuzzy logic is considered a powerful instrument for dealing with uncertainty and is implemented in both type-1 and type-2 ways. The expert systems (ESs) and decision support systems (DSSs) are applied based on type-1 and type-2 fuzzy logic since medical decision-making has always been associated with various uncertainties. The present study reviewed different types of fuzzy ES/DSS in the medical domain in order to investigate whether the fuzzy type-2 performance was better compared to that of type-1. Materials and Methods: A systematic review was conducted on PubMed, Web of sciences, Scopus, Embase, Medline, and Science Direct databases. The title, abstract, and full text of the articles, published during 2007-2017, were independently evaluated by two reviewers. The cases of disagreement were solved in a pair-work discussion. Finally, based on inclusion criteria, 12 articles were included in the study and were investigated in terms of the purpose and application, architecture and structural details, as well as the method of evaluation and the findings. Results: Type-2 expert systems were found to have a better diagnostic function compared to Type-1 systems and other different machine learning methods. Increasing the accuracy, precision, and resistance to noise was an issue that was achieved in such systems using type-2 fuzzy logic. Conclusions: In general, medical expert systems based on type-2 fuzzy logic are considered more appropriate for model uncertainty and ambiguity, therefore, they could be used in different medical domains that need to make decisions under uncertain circumstances.