Evaluation of drug and disease influence on occurrence of hyperkalemia: a data mining approach
Volk Markovic Petra, Tomislav Laptoš, Robnik-Sikonja Marko, Andreja Čufar · Studies in health technology and informatics · 2015
The retrospective study was conducted on 150 patients with hyperkalemia during either the admission or hospitalization or their visit to outpatient clinics. Medical records were examined by clinical pharmacists. The demographic characteristics, diagnosis, serum potassium level and medication therapy administered to the patient one day before the detection of hyperkalemia were collected. Medical data and additional background knowledge were analysed using two different machine learning algorithms for association rule mining to extract knowledge in the form of association rules [2, 3]. Detected rules were reviewed and evaluated by clinical pharmacists.