Has Microsoft left behind risk modeling in cardiac and thoracic surgery?

This concept paper examines a number of key areas central to quality and risk assessment in cardiac surgery. The effect of surgeon and institutional factors with regard to outcomes in cardiac surgery is utilized to demonstrate the need to sub analyze cardiac surgeons performance in a more sophisticated manner than just operation type and patient risk factors, as in current risk models. By utilizing the mathematical/engineering concept of Fourier analysis in the breakdown of cardiac surgical results the effects of each of the core components that makes up the care package of a patient's experiences are examined. The core components examined include: institutional, regional, patient, and surgeon effects. The limitations of current additive (Parsonnet, Euroscore) and logistic (Euroscore, SouthernThoracic Society) regression risk analysis techniques are discussed. The inadequacy of current modeling techniques is demonstrated via the use of known medical formula for calculating flow in the internal mammary artery and the calculation of blood pressure. By examining the fundamental limitations of current risk analysis techniques a new technique is proposed that embraces modern software computer technology via the use of structured query language.