Diabetic patients management exploiting case-based reasoning techniques

In this paper we propose a case-based decision support tool, designed to he lp physicians in Ist type diabetes therapy revision through the intelligent retrieval of data related to past situations (or 'cases') similar to the c urrent one. A case is defined as a set of variable values (or features) col lected during a visit. We defined taxonomy of prototypical patients' condit ions, or classes, to which each case should belong. For each input case, th e system allows the physician to find similar past cases, both from the sam e patient and from different ones. We have implemented a two-steps procedur e; (1) it finds the classes to which the input case could belong; (2) it li sts the most similar cases from these classes, through a nearest neighbor t echnique, and provides some statistics useful for decision taking. The perf ormance of the system has been tested on a data-base of 147 real cases, col lected at the Policlinico S. Matteo Hospital of Pavia. The tool is fully in tegrated in the web-based architecture of the EU funded Telematic managemen t of Insulin Dependent Diabetes Mellitus (T-IDDM) project. (C) 2000 Elsevie r Science Ireland Ltd. All rights reserved.