It is very important to inform the patient and his/her relatives about the risk of mortality before a cardiovascular operation. For this respect, a model called EuroSCORE (The European System for Cardiac Operative Risk Evaluation) has been developed by European cardiovascular surgeons. This system gives the risk of mortality during or 30 days after the operation, based on the values of some parameters measured before the operation. The model used by EuroSCORE has been developed by statistical data gathered from large number of operations performed in Europe. No research has been performed for operations on Turkish cardiovascular patients. When the genetic characteristics and life styles of Turkish patients taken into consideration, it is highly probable that the mortality risks of Turkish patients will be different than European patients. At least four risks are considered as of primary concern for the patient and the surgeon in cardiovascular operations. These are mortality, morbidity, use of national resources and the patient satisfaction. Considering the recent developments in surgical techniques, although the risk of mortality has been reduced in a large extent. In the context of this project, a model that can be used for mortality will be developed. In this research project, a data base system for storing data about cardiovascular operations performed in Turkish hospitals, a web application for gathering data, and a machine learning system on this database to learn a risk model, similar to EuroSCORE, will be developed. In this work, the feature projections based machine learning approach that has been successfully applied to several medical domains will be used as the basis. In this thesis, a risk estimation system for predicting the mortality risk of patients undergoing cardiovascular surgeries by maximizing the area under the ROC curve.

DATE: 2 September, 2008, Tuesday@ 16:00
PLACE: EA 409