Separation of fNIRS signals into functional and systemic components based on differences in hemodynamic modalities

Abstract

In conventional functional near-infrared spectroscopy (fNIRS), systemic physiological fluctuations evoked by a body's motion and psychophysiological changes often contaminate fNIRS signals. We propose a novel method for separating functional and systemic signals based on their hemodynamic differences. Considering their physiological origins, we assumed a negative and positive linear relationship between oxy- and deoxyhemoglobin changes of functional and systemic signals, respectively. Their coefficients are determined by an empirical procedure. The proposed method was compared to conventional and multi-distance NIRS. The results were as follows: (1) Nonfunctional tasks evoked substantial oxyhemoglobin changes, and comparatively smaller deoxyhemoglobin changes, in the same direction by conventional NIRS. The systemic components estimated by the proposed method were similar to the above finding. The estimated functional components were very small. (2) During finger-tapping tasks, laterality in the functional component was more distinctive using our proposed method than that by conventional fNIRS. The systemic component indicated task-evoked changes, regardless of the finger used to perform the task. (3) For all tasks, the functional components were highly coincident with signals estimated by multi-distance NIRS. These results strongly suggest that the functional component obtained by the proposed method originates in the cerebral cortical layer. We believe that the proposed method could improve the reliability of fNIRS measurements without any modification in commercially available instruments.

Figure 1. Specially designed optodes and their holder system.

Optode arrays consisting of a source () and four detectors (–) were fixed directly above the left and right primary motor areas. The detector optodes on both sides were linearly aligned at distances of 10, 20, 30, and 40 mm from the source optode.

Figure 2. The

dependency of the mutual information for participant 1. Upper two rows: body-tilting task. Middle two rows: breath-holding task. Bottom two rows: finger-tapping task. Left and Right indicate the measurement positions. 10 mm, 20 mm, 30 mm, and 40 mm indicate the source-detector distances.

Figure 3. Histogram of estimated

values from all experiments.

Figure 4. Hemodynamic changes during the body-tilting task for all participants.

Left two columns: hemodynamics estimated by the conventional method. Middle two columns: systemic component by the proposed method. Right two columns: functional component by the proposed method. Data were block averaged. Distance between source and detector was 30 mm. Red and blue lines indicate oxy- and deoxyhemoglobin changes, respectively. Red and blue bands indicate SDs for oxy- and deoxyhemoglobin changes, respectively. Green line indicates the task period. “Left” and “Right” indicate the measurement positions.

Figure 5. Hemodynamic changes during the breath-holding task for all participants.

Left two columns: hemodynamics estimated by the conventional method. Middle two columns: systemic component by the proposed method. Right two columns: functional component by the proposed method. Data were block averaged. Distance between source and detector was 30 mm. Red and blue lines indicate oxy- and deoxyhemoglobin changes, respectively. Red and blue bands indicate SDs for oxy- and deoxyhemoglobin changes, respectively. Green line indicates the task period. “Left” and “Right” indicate the measurement positions.

Figure 6. Hemodynamic changes during the finger-tapping task for participants except participant 3.

Left two columns: hemodynamics estimated by the conventional method. Middle two columns: systemic component by the proposed method. Right two columns: functional component by the proposed method. Data were block averaged. Distance between source and detector was 30 mm. Red and blue lines indicate oxy- and deoxyhemoglobin changes, respectively. Red and blue bands indicate SDs for oxy- and deoxyhemoglobin changes, respectively. Green line indicates the task period. “Left” and “Right” indicate the measurement positions. “L” (left) and “R” (right) indicate the side used during finger tapping.

Figure 7. Hemodynamic changes at various source-detector distances during the finger-tapping task for participant 1.

Upper two rows: hemodynamics estimated by the conventional method. Middle two rows: functional component by the proposed method. Bottom two rows: systemic component by the proposed method. Data were block averaged. Red and blue lines indicate oxy- and deoxyhemoglobin changes, respectively. Red and blue bands indicate SDs for oxy- and deoxyhemoglobin changes, respectively. Green line indicates the task period. “L” (left) and “R” (right) indicate the side used during finger tapping. “Left” and “Right” indicate the measurement positions. “10 mm,” “20 mm,” “30 mm,” and “40 mm” indicate the source-detector distances.

Figure 8. Hemodynamic trajectories of the three methods for participant 1.

Data during the three kinds of tasks are shown. Green line: the conventional method. Red and blue lines: the functional and the systemic components of the proposed method. Black line: the CWMD method. “Left” and “Right” indicate the measurement positions.

Figure 9. Comparison of estimated temporal hemodynamics of the functional component and the CWMD result.

Data during the three kinds of tasks for participant 1 were used. Red and blue lines correspond to oxy- and deoxyhemoglobin changes of the functional component, respectively. Black line: the corresponding hemoglobin changes measured by the CWMD method. The CWMD hemodynamics was rescaled to obtain the best fit to that of the functional component under each condition. “Left” and “Right” indicate the measurement positions. “L” (left) and “R” (right) indicate the side used during finger tapping.

Figure 10. FFT analysis of the functional and systemic components of the three tasks for participant 1.

For detailed analysis, the low-frequency range is also shown in the second, fourth, and sixth columns.

Figure 11. Optical attenuation ratio of the functional and systemic components across different source-detector distances.

Optical attenuation changes of the functional and systemic components at a wavelength of 780 nm were calculated by using an absorption coefficient matrix . Ratios in the optical attenuation changes against that at the distance of 30 mm was calculated by a least mean square method such that the shape of optical attenuation change at the distance best coincided with that at 30 mm when that was multiplied by the ratio. A gray line shows ratios across different source-detector distances for each task and each participant. Lines for all tasks for all participants are presented. Black lines indicate the simulated partial optical path lengths of the gray matter (left figure) and scalp (right figure). All values were normalized by the value at the source-detector distance of 30 mm.

Figure 12. Difference in signal separation under three different

conditions. Values of : −0.4 (left column), −0.6 (middle column), and −0.8 (right column) were used for the data at a source-detector distance of 30 mm during the finger-tapping task for participant 1. Upper two rows: hemodynamics estimated by the conventional method. Middle two rows: functional component by the proposed method. Bottom two rows: systemic component by the proposed method. Data were block averaged. Red and blue lines indicate oxy- and deoxyhemoglobin changes, respectively. Red and blue bands indicate SDs for oxy- and deoxyhemoglobin changes, respectively. Green line indicates the task period. “Left” and “Right” indicate the measurement positions. “L” (left) and “R” (right) indicate the side used during finger tapping.