This project will involve the development and evaluation of data reduction techniques, which can be implemented as low cost microcomputer-based monitor. The objective is to show the feasibility of using real-time signal processing to discriminate seizures from artifact in long-term EEG recorded for neonates in an intensive care unit. A second phase involves construction and testing of prototype monitors for evaluation at clinical sites.
The specific aim of this Phase 1 study is to evaluate a number of single processing techniques for their ability to detect and discriminate neonatal seizures from long-term monitoring tapes provided by the Children's Hospital of Philadelphia. Computer detection rates for the various algorithms will be compared to that of visual inspection of the EEG. Postponed for a Phase II study will be determination of the optimal number of channels and other non-EEG parameters to be recorded, which parameters should be tended, and considerations for data storage and display, in a device which could be commercially viable.The ultimate goal of this proposal is to produce a system for data reduction recording and display with a price of approximately $12,000.00.
