On-line integration of cardiological data to support medical decision making in patients with ischemic heart disease

Abstract

We have developed a system that integrates, through a virtual 3D-dynamic heart model, all the pertinent clinical and instrumental information obtained for patients hospitalized for suspected or documented ischemic heart disease (IHD). In the present study we retrospectively compared the diagnosis formulated by the cardiologist in the discharge record with that automatically provided by the system. Divergences were found in 27% of the 110 patients studied and classified into four types: I) inability of the system to provide the correct diagnosis because of the lack of pertinent diagnostic parameters in the model (3%), a cardiologist's diagnosis which was either II) not supported by objective data (3%), III) in conflict with the available information (10%), or IV) incomplete (11%). An experimental trial has been started in which the cardiologist in charge of the patient uses the automatic system during the diagnostic process and compilation of the discharge report.