Probabilistic unit life status estimation (PULSE)

Subrata Das, Josh Introne, Dave Lawless, Reed Hoyt, Steve Muza

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

We present here the Probabilistic Unit Life Status Estimation (PULSE) methodology and software architecture. The PULSE approach leverages Temporal Belief Networks (TBNs) to model the subject's fundamental physiological dynamics (e.g. shock response) over time, to perform life status estimation in a robust manner that accounts for sensor hardware malfunctions. We developed a limited-scope prototype using our in-house TBN engine to estimate clinical status, which served as a framework to demonstrate the PULSE approach for two separate but related clinical assessment tasks. In the first task, the PULSE prototype provides a life status assessment of a simulated unit, allowing for the principled introduction of noise to demonstrate the system 's ability to detect sensor failure. In the second task, the PULSE prototype provides a clinical assessment of a unit 's degree of acclimatization, using pre-recorded data from studies of soldiers at altitude.

Original languageEnglish (US)
Title of host publicationProceedings of the Seventh International Conference on Information Fusion, FUSION 2004
EditorsP. Svensson, J. Schubert
Pages959-966
Number of pages8
StatePublished - Nov 2 2004
Externally publishedYes
EventProceedings of the Seventh International Conference on Information Fusion, FUSION 2004 - Stockholm, Sweden
Duration: Jun 28 2004Jul 1 2004

Publication series

NameProceedings of the Seventh International Conference on Information Fusion, FUSION 2004
Volume2

Other

OtherProceedings of the Seventh International Conference on Information Fusion, FUSION 2004
CountrySweden
CityStockholm
Period6/28/047/1/04

Keywords

  • Acclimatization Assessment
  • Life Status Assessment
  • Temporal Belief Networks

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint Dive into the research topics of 'Probabilistic unit life status estimation (PULSE)'. Together they form a unique fingerprint.

Cite this