Computer programs for statistical evaluation of data derived by manual analysis of the EEG: From a study with special regard to relationships between EEG and Cerebral Blood Flow
✍ Scribed by I.A. Sulg; G. Lindgren; B. Stigsby
- Publisher
- Elsevier Science
- Year
- 1970
- Weight
- 699 KB
- Volume
- 1
- Category
- Article
- ISSN
- 0010-468X
No coin nor oath required. For personal study only.
✦ Synopsis
There is an increasing need for statistical analysis of the electrical activity of the brain (EEG) especially with regard to its relationships to other parameters of cerebral activity and metabolism. In the present study new techniques for EEG quantitation and for the measurement of regional cerebral blood flow (rCBF) have provided a unique opportunity to correlate EEG and blood flow regionally. Due to the large bulk of data from more than one hundred patients the use of computer resources was imperative. For the sake of rationalization it was decided to utilize existing standard subroutines as much as possible. Only one new subroutine (for the clustering of spectral pro-Files) had to be programmed. The present computer work thus serves as an example of how to use and combine a number of different statistical subroutines, available partly from the IBM Scientific Subroutine Package and partly from the UCLA BioMedical Data Program Library, in a meaningful program with a rational coupling and recoupling between different stages of the computations.
By means of the present computer study we succeeded in demonstrating strongly significant correlations between the EEG, the cerebral blood flow and the age of patients. The electronic data processing thus offers facilities, also for neurophysiological research, which make it possible to perform statistical analysis of dimensions and complexities not obtainable before the computer era.
Neurophysiology
EEG analysis Cerebral blood flow Relationship EEG/Cerebral blood flow IBM scientific subroutine package profile analysis * The subroutines are indicated by the coded names, given in the SSP/ll30. ** By check up of the subroutines used in the present program concerning the WTEST, two errors were found in form H-20-0252-2,3 page 58: in column 1, the degrees of freedom should be M-1. In column 2, under parameter R, the statement should be -the largest is ranked M-. (Admitted by the publisher.)