Estimation of brain tissue's hemoglobin oxygen saturation (SO2) of rats by artificial neural network.
- Author:
Lijuan DAI
1
;
Zhiyu QIAN
;
Huinan WANG
;
Guoqiang YU
Author Information
1. Department of Biomedical Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China.
- Publication Type:Journal Article
- MeSH:
Algorithms;
Animals;
Brain;
metabolism;
Computer Simulation;
Female;
Hemoglobins;
metabolism;
Neural Networks (Computer);
Oximetry;
methods;
Oxygen;
blood;
Rats;
Rats, Sprague-Dawley;
Spectrum Analysis;
methods
- From:
Journal of Biomedical Engineering
2008;25(4):917-920
- CountryChina
- Language:Chinese
-
Abstract:
A specifically designed fiber optical spectrum system was applied to measure the hemoglobin oxygen saturation (SO2) of living rat's brain tissue. Collections of in vivo visible absorption spectrum measured by the fiber optical spectrometer at different SO2 as well as their corresponding SO2 values measured by OXImeter were made on a kind of tissue model (i. e. the mixture of Intralipid and blood) and were used as the training data of an artificial neural network (ANN) which, after being trained, can put out SO2 value correctly when an absorption spectrum was put in. The SO2 values of 5 living rats' brain tissue at different depth were obtained, and the experimental error range was +/- 5%. This method is of great significance to important value for monitoring hemoglobin oxygen saturation at various depths in brain minimally invasive surgery.
