Detecting of Preeclampsia by Expert System: A Case Study in Tehran University of Medical Sciences Hospitals

Diagnosis of preeclampsia has an essential role in applying appropriate treatment plan for the patients. The aim of this study was to design an expert system in order to diagnos preeclampsia in order to assist the clinicians.

This was a cross-sectional study which resulted in developing a new system. The study population consisted of all patients admitted to three Maternity hospitals affliated to Tehran University of Medical Sciences (TUMS). Sample size included 215 medical records which were randomly selected. The results obtained were compared with the diagnosis from experts by kappa test using SPSS software.

First of all, input parameters fuzzificated and entered into inference engine. Outputs were categorized in two groups as patients and healthy, with the final diagnosis and clinical explanation. The results obtained from system evaluation showed that accuracy, specificity and sensitivity of the system were 98.2 percent, 100 percent and 96.4 percent respectively.

Based on evaluation results, it could be concluded that fuzzy logic is an efficient method for designing of expert systems in the field of obstetrics and gynecology. Also, due to the similarity of the logic used in the proposed system with workflow and medical decision making, it will be accepted by the physicians.