Data mining in predicting survival of kidney dialysis patients
Shital Shah, Andrew Kusiak, Bradley S. Dixon · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2003
The number of dialysis patients due to end stage kidney disease is increasing. Finding ways to improve patient outcomes and reduce the cost of dialysis is a challenging task. Dialysis care is complex and multiple factors may influence patient survival. More than 50 parameters may be monitored while providing a kidney dialysis treatment. Understanding the collective role of these parameters in determining outcomes for an individual patient and administering individualized treatments is of importance. Individual patient survival may depend on a complex interrelationship between multiple demographic and clinical variables, medications, and medical interventions. In this research, a data mining approach is used to elicit knowledge about the interaction between these variables and patient survival. Two different data mining algorithms are employed for extracting knowledge in the form of decision rules. Data mining is performed on the individual visits of the "most invariant" patients as they form "signatures" for their decision categories. The concepts introduced in this research have been applied and tested using a data collected at four dialysis sites. The computational results are reported.