A Fuzzy-Based Clinical Decision Support System for Coeliac Disease

Coeliac disease (CD) is a permanent inflammatory of the small intestine characterized by destruction mucous membrane this intestinal tract. represents most frequent food intolerance and affects about 1% population, but it severely underdiagnosed. Currently available guidelines require CD-specific serology atrophic histology in duodenal biopsy samples to diagnose CD adults. In paediatric CD, recently adults also, non-invasive diagnostic strategies have become increasingly popular. order increase rates correct diagnosis without use biopsy, researchers been using approaches based on artificial intelligence techniques. work, we present Clinical Decision Support System (CDSS)system for supporting diagnosis, developed context Italy-Malta cross-border project ITAMA. The implemented CDSS has neural-network-based fuzzy classifier. system was tested Virtual Database Real acquired during ITAMA project. Analysis 10,000 virtual patients shows that achieved an accuracy 99% sensitivity 99%. On 19,415 real patients, which 109 with confirmed coeliac disease, 99.6% accuracy, 85.7% sensitivity, specificity 96% precision. Such results show can be used effectively support reducing appeal invasive techniques such as biopsy.