Premium
Adaptive detection of distribution changes in medical information systems
Author(s) -
Xue Xinjian Jack
Publication year - 2007
Publication title -
pamm
Language(s) - English
Resource type - Journals
ISSN - 1617-7061
DOI - 10.1002/pamm.200700952
Subject(s) - variance (accounting) , enhanced data rates for gsm evolution , signal (programming language) , series (stratigraphy) , computer science , medical information , fault (geology) , change detection , process (computing) , distribution (mathematics) , statistics , algorithm , mathematics , artificial intelligence , information retrieval , biology , paleontology , accounting , programming language , operating system , mathematical analysis , business
The number of messages received by a medical information system was modelled as the signal in a stochastic process. Algorithms were developed for detecting changes of mean and variance and reporting the changes of status from normal to abnormal (upward edge), and vice verse (downward edge) under the assumption that the signal was distributed independently and normally in a series of time intervals. System tests have shown that the performance of the fault detecting and reporting in non‐holiday days is generally satisfactory for all laboratory data feeds. (© 2008 WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim)