Allergenius, an expert system for the interpretation of allergen microarray results

Abstract

Background An in vitro procedure based on a microarray containing many different allergen components has recently been introduced for use in allergy diagnosis. Recombinant and highly purified allergens belonging to different allergenic sources (inhalants, food, latex and hymenoptera) are present in the array. These components can either be genuine or cross-reactive, resistant or susceptible to heat and low pH, and innocuous or potentially dangerous. A large number of complex and heterogeneous relationships among these components has emerged, such that sometimes these interactions cannot be effectively managed by the allergist. In the 1960s, specialized languages and environments were developed to support the replacement of human experts with dedicated decision-making information systems. Currently, expert systems (ES) are advanced informatics tools that are widely used in medicine, engineering, finance and trading.Methods We developed an ES, named Allergenius ®, to support the interpretation of allergy tests based on microarray technology (ImmunoCAP ISAC ®). The ES was implemented using Flex, a LPA Win-Prolog shell. Rules representing the knowledge base (KB) were derived from the literature and specialized databases. The input data included the patient’s ID and disease(s), the results of either a skin prick test or specific IgE assays and ISAC results. The output was a medical report.Results The ES was first validated using artificial and real life cases and passed all in silico validations. Then, the opinions of allergists with experience in molecular diagnostics were compared with the ES reports. The Allergenius reports included all of the allergists’ opinions and considerations, as well as any additional information.Conclusions Allergenius is a trustable ES dedicated to molecular tests for allergy. In the present version, it provides a powerful method to understand ISAC results and to obtain a comprehensive interpretation of the patient’s IgE profiling. Keywords: 3 TO 10 DIFFERENT ImmunoCAP ISAC, Allergen microarray, Artificial intelligence, Expert system, ISAC interpretation, Genuine component, Cross-reactive component, Computer supported diagnosis