Skip to Content
Find More Like This
Return to Search

Computational Method for Improved Forewarning of Critical Events

Oak Ridge National Laboratory

Contact ORNL About This Technology


Technology Marketing SummaryORNL’s computational method for analyzing nonlinear processes provides improved
forewarning of imminent critical events. This is achieved through phase space
dissimilarity analysis of data from mechanical or electrical devices, biomedical data, or
other physical processes. The need to distinguish between similar, yet distinct, states in
a nonlinear process is essential to many engineering, research, and medical applications.

DescriptionThis technology is capable of detecting failure onset from a variety of data sources,
including human brain waves and chest sounds, motors, and electrical devices.
Examples of possible biomedical forewarning include distinguishing between
preseizure and nonseizure brain waves, prefibrillation and fibrillation heart waves,
presyncope and syncope heart waves, presepsis and sepsis heart waves, and normal
and abnormal chest sounds.
BenefitsImproved forewarning of critical events
An end-of-life forewarning indicator
New method to combine several channels of time-serial data

Applications and IndustriesForewarning epileptic seizures, ventricular fibrillations, fainting, and sepsis
Forewarning failures in power systems and machines
Identification of individuals using a biometric parameter
Assessment of patient alertness through brain wave analysis
Machine failure forewarning
More InformationPatents
Lee M. Hively, Methods for Improved Forewarning of Critical Events Across Multiple Data Channels, U.S. Patent 7,209.861 B2, issued April 24, 2007.
Lee M. Hively, Methods for Consistent Forewarning of Critical Events Across Multiple Data Channels, U.S. Patent 7,139,677 B2, issued November 21, 2006.
Lee M. Hively, Paul C. Gailey, and Vladimir A. Protopopescu, Condition Assessment of Nonlinear Processes, U.S. Patent 6, 484,132 B1, issued November 19, 2002.
Lee M. Hively, and Esmond G. Ng, Integrated Method for Chaotic Time Series Analysis, U.S. Patent 5,815,413, issued September 29, 1998.
Ned E. Clapp and Lee M. Hively, Method and Apparatus for Extraction of Low-frequency Artifacts from Brain Waves for Alertness Detection, U.S. Patent 5,626,145, issued May 6, 1997.

Lead Inventor
Lee M. Hively
Computational Sciences and Engineering Division
Oak Ridge National Laboratory
Technology Status
Technology IDDevelopment StageAvailabilityPublishedLast Updated
1744, 0264, 0694, 0885, 1300, 1731, 2321DevelopmentAvailable09/25/201209/25/2012

Contact ORNL About This Technology

To: sims<simsdl@ornl.gov>