State-space modeling based on principal component analysis and oxygenated-deoxygenated correlation to improve near-infrared spectroscopy signals

摘要: Near infrared spectroscopy (NIRS) is currently becoming an effective technique for noninvasive functional brain imaging. Therefore, the methods to improve quality of measured NIRS signals play important role make broadly accepted in practical applications. Previously, there have been approaches using state-space modeling recover from basic component eliminate artifacts presented measurements. However, proposed approach requires us onset vector determine starting position stimulus that not always available situation. In this work, we provide a new way find components efficient implementations modeling. We apply principal analysis estimate eigenvector-based basis presents compact information whole signals. utilize oxygenated-deoxygenated correlation another set enhance The based on Kalman filter used reconstruct these components. tested algorithm with actual data and showed significant improvements contrast-to-noise (CNR) after filtered by our approach.