Fuzzy-based framework for diagnosis of acid–base disorders

Abstract

The main objective of this research is to develop a fuzzy-based framework for diagnosis of different acid–base disorders. There are several acid–base disorders that cause many clinical complications and their proper diagnosis is the only way for their efficient treatment. The common disorders are metabolic acidosis, metabolic alkalosis, non-anion gap acidosis, anion-gap acidosis, acute respiratory alkalosis and chronic respiratory alkalosis. The proposed fuzzy-based framework was used to diagnose all of these disorders using four parameters directly measured in blood: hydrogen-ion concentration (pH), arterial blood carbon dioxide partial pressure (paCO2), sodium ions concentration (Na+) and chloride ions concentration (Cl−) along with 12 features extracted from the directly measured parameters. The validation results showed that the developed framework has an accuracy of 94%, an average sensitivity of 88% and a specificity of 93%. These results imply that the developed fuzzy-based framework is accurate and reliable one and can be used to help clinicians specially the non-expert ones to provide correct and rapid diagnosis of acid–base disorders.