Processing EEG signals to predict brain damage
US6931274B2 · kind B2 · utility
Assignee
Inventor
Key dates
| Filing date | May 6, 2002 |
| Grant date | Aug 16, 2005 |
| Priority date | — |
| Expiry date | May 6, 2022 |
Classification
- Technology area (CPC A)Human Necessities
- CPC primaryA61B5/726
- WIPO fieldMedical technology
- WIPO sectorInstruments
Abstract
Rapid and accurate in-vivo assessment of cerebral white matter injury particularly for pre-term infants, for timely treatment and/or prediction of outcomes has been very limited. This invention exploits the discovery that reduced high-frequency EEG intensity, particularly as shown by the upper spectral edge frequency, is a good indicator of cerebral white matter neural injury and is well correlated with MRI results. With more experience of clinical cases, a set of simple rules such as “if the spectral edge value is below 8 Hz there is a high likelihood of injury” may be validated, yet the EEG technology involved is largely invisible to the user. In the invention, EEG signals are processed by software to obtain, store, and graphically display bilaterally collected EEG spectral edge and intensity values over from hours to weeks. Rejection of corrupted signals by filtering and gating means is responsive to incoming signal characteristics, to additional inputs such as motion sensors or impedance tests, and to patient data (gestational age in particular). The invention includes the software and methods of use.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.