Managing and mining clinical outcomes

In this paper, we describe clinical outcomes analysis for data in Memorial Sloan-Kettering Cancer Center Sarcoma Database using relational data mining and propose an infrastructure for managing cancer data for Drexel University Cancer Epidemiology Server (DUCES). It is a network-based multi-institutional database that entails a practical research tool that conducts On-Line Analytic Mining (OLAM). We conducted data analysis using relational learning (or relational data mining) with cancer patients' clinical records that have been collected prospectively for 20 years. We analyzed clinical data not only based on the static event, such as disease specific death for survival analysis, but also based on the temporal event with censored data for each death. Rules extracted using relational learning were compared to results from statistical analysis. The usefulness of rules is also assessed in the context of clinical medicine. The contribution of this paper is to show that rigorous data analysis using relational data mining provides valuable insights for clinical data assessment and complements traditional statistical analysis and to propose an infrastructure to manage and mine clinical outcomes used in multi-institutional organizations.