2010

In Medical Informatics, there is an increasing awareness that temporal information plays a crucial role, so that suitable database approaches are needed to store and support it. Specifically, most clinical data are intrinsically temporal, and a relevant part of them are now-relative (i.e., they are valid at the current time). Even if previous studies indicate that the treatment of now-relative data has a crucial impact on efficiency, current approaches have several limitations. In this paper we propose a novel approach, which is based on a new representation of 'now', and on query transformations. We also experimentally demonstrate that our approach outperforms its best competitors in the literature to the extent of a factor of more than ten, both in number of disk accesses and of CPU usage.

Presented at Conferences

  • AMIA 2008 (2008)

    Washington, DC USA