Development and evaluation of Clinical Decision Support System (CDSS) for the diagnosis of irritable bowel syndrome (IBS)

Abstract

Introduction IBS manifestations are similar to heartburn, making diagnosis difficult for physicians. To diagnose this illness, doctors now rely on their experiences and therapeutic guidelines. Misdiagnosis, added costs, and extended treatment times are possible outcomes of this method. Researchers believe CDSS can help clinicians solve problems when used to make decisions. The CDSS is used in this current study to diagnose IBS. Methods The fuzzy-logic algorithm was optimized in this applicable modeling research using particle swarm optimization (PSO). Input data, an inference engine, and output data comprised this fuzzy-logic model-based system. Classification algorithms and the PSO method were used to select the input variables. PSO and "If-then" rules were used in the inference engine to extract data from the dataset. Patients experiencing IBS and normal people make up the output. The accuracy, sensitivity, precision, specificity, confusion Matrix, kappa test, and F-measure values of this model were used to assess its performance. Results The recommended model had a mean score of 96.5% accuracy, 100% sensitivity, 95.2% precision, and 89.4% specificity. Conclusion The optimized model was found that effectively diagnosed IBS cases. To improve the accuracy of this disease's diagnosis, healthcare organizations can implement the aforementioned model into their strategic scheduling at a reasonable expense.