¹Department of Software, Shenyang University of Technology, Shenyang, China
²Shenyang HeXing Testing Equipment Co., Ltd., China
 Department of information science and technology, Shenyang University of Technology, Shenyang, China

---

Abstract—the main aim of BCI builds a communicating bridge between brain and peripheral devices. NIRS is dependent on changes of blood flow, as it measures oxygenated and deoxygenated hemoglobin’s in the superfacial layers of the human cortex. We are able to detect HbO and HbR of imaged movement and movement on the surface of the brain with NIRS. If we want to achieve the control of external devices with HbO and HbR, the change of these data must be analyzed, and be extracted. In this paper, we present a new method to achieve in the analysis of the data of HbO and HbR, realize the removal of high frequency and achieve preliminary extraction the characteristics of the data. Secondary analysis of the extracted feature points could reduce the number of feature points. Designing a new compensated interpolation algorithm achieve completely new feature points to replace the original feature points to represent the data .The interpolated data curves response the change of original data, and realize the removal of high frequency to smooth the output curve.

Index Terms—interpolation algorithm, time series analysis, near-infrared spectroscopy, oxygenated hemoglobin

[PDF]

Cite: Yan Jiang, Xin Jin, Xin Li, "Features Extraction from NIRS Data using Extreme Decomposition," Journal of Computers vol. 9, no. 7, pp. 1671-1677, 2014.