Computer assisted medical decision-making system using genetic algorithm and extreme learning machine for diagnosing allergic rhinitis

Abstract

Allergic rhinitis (AR) is an antigen-mediated inflammation of the nasal mucosa that might extend into the paranasal sinuses. Rhinorrhea, nasal obstruction or blockage, nasal itching, sneezing, and postnasal drip that reverse spontaneously or after treatment are symptoms of AR. Allergic conjunctivitis frequently accompanies AR. For diagnosis of AR, intradermal skin tests remain the gold standard. This paper presents a clinical decision-making system that assists the clinicians to diagnose whether a patient suffers from AR. Feature selection is done using a wrapper approach that employs genetic algorithm (GA) and extreme learning machine (ELM). The selected features are trained and tested using an ELM classifier. For experimenting, the outcome of the symptoms observed in 872 patients for diagnosing the presence or absence of AR has been used. The experimental result shows that the system has achieved an accuracy of 97.7%.